ABSTRACT Title of Document: WATER IMERSION BALASTED PARTIAL GRAVITY FOR LUNAR AND MARTIAN EVA SIMULATION John Richard Mularski Master of Science, 2007 Directed By: Asociate Profesor David Akin Department of Aerospace Engineering The University of Maryland Space Systems Laboratory is developing the capability to simulate partial gravity levels for human operational activities through the use of balast on body segments in the underwater environment. This capability wil be important as NASA prepares to return to the Moon by the end of the next decade. This thesis discusses various forms of partial gravity simulation used in the past, and discusses applications for balasted underwater simulations. Primary application of this technique is for static or quasistatic activities, such as collecting basic anthropometric data on reach envelopes or postural control, as wel as acumulating an experience base on partial gravity habitat and vehicle design and operations. The research conducted investigated collecting postural stability data through the use of a controlled disturbance to the balasted subject. WATER IMERSION BALASTED PARTIAL GRAVITY FOR LUNAR AND MARTIAN EVA SIMULATION By John Richard Mularski Thesis submited to the Faculty of the Graduate School of the University of Maryland, College Park, in partial fulfilment of the requirements for the degre of Master of Science 2007 Advisory Commite: Asociate Profesor David Akin, Chair Asociate Profesor Robert Sanner Visiting Profesor Mary Bowden ? Copyright by Space Systems Laboratory University of Maryland 2007 i Dedication This paper is dedicated to my parents for al of their help and support. ii Acknowledgements I would like to thank everyone at the SL for al the help they have provided during my time here. I would especialy like to thank my test subjects Agnieszka, Liz, Ali, and Tim as wel as my lead safety diver Dave. I would also like to thank my adviser Dr. Akin for helping me develop this project and guiding me through. iv Table of Contents Dedication........................................................i Acknowledgements.................................................ii Table of Contents...................................................iv List of Tables......................................................vi List of Figures.....................................................vii List of Abbreviations...............................................vii Chapter 1: Introduction...............................................1 Chapter 2: Background Material and Literature Review......................3 Parabolic Flight...................................................3 Advantages....................................................4 Disadvantages..................................................4 Previous Research...............................................5 Counterweight Systems.............................................6 Advantages....................................................8 Disadvantages..................................................8 Previous Research...............................................9 Inclined Suspension...............................................11 Advantages...................................................12 Disadvantages.................................................12 Previous Research..............................................13 Water Imersion.................................................14 Advantages...................................................15 Disadvantages.................................................15 Previous Research..............................................16 1G Simulation...................................................18 Advantages...................................................18 Disadvantages.................................................19 Previous Research..............................................20 Backpack Loading................................................21 Chapter 3: Partial Gravity System Design & Development....................23 Balast System Development........................................25 Jumpsuit......................................................25 Harnes......................................................26 Balast Atachment Development.....................................28 Hard Plate System..............................................28 Pocket System.................................................32 Chapter 4: Procedures Development....................................35 Nominal Operations...............................................35 Emergency Procedures.............................................38 Normal Extraction..............................................38 Ladder Extraction...............................................39 Safety Diver Extraction..........................................40 Crane Extraction...............................................40 v Chapter 5: Experimental Procedures & Hardware..........................42 Orientation Dive.................................................42 Data Collection Dives.............................................43 Ramp........................................................44 Pusher.......................................................44 Simulation PLS Backpack.......................................46 Video Camera.................................................47 Experimental Procedures.........................................48 Chapter 6: Results & Data Analysis.....................................52 Subjects........................................................52 Quantitative Results...............................................52 Experimental Problems & Solutions...................................55 Data Analysis Problems & Solutions................................59 Chapter 7: Conclusions..............................................62 Equipment Improvements..........................................62 Procedural Improvements...........................................62 Analysis Improvements............................................63 Alternate Experimental Method......................................63 Chapter 8: Future Work..............................................65 Shirtsleve Experiments............................................65 MX-2 Space Suit Analogue.........................................66 Human-Robot Interaction...........................................68 Appendix A: Pusher Experiment Data...................................69 Appendix B: MATLAB Code........................................104 Appendix C: IRB Paperwork.........................................109 Bibliography.....................................................121 vi List of Tables Table 1 - Body Segment Parameters [22].................................24 Table 2 - Subject Extraction Breakdown.................................38 Table 3 - Subject Information..........................................52 Table 4 - Data Variation for Subject 2...................................53 Table 5 - Repeatability Test Data.......................................60 vii List of Figures Figure 1 - Parabolic Flight Trajectory [1].................................4 Figure 2 - Apollo Parabolic Flight EVA Training [3]........................5 Figure 3 - Apollo Counterweight EVA Training [6].........................7 Figure 4 ? Example of a Counterweight System [7].........................7 Figure 5 - Typical Inclined Suspension Rig [10]...........................12 Figure 6 - Aquarius Undersea Habitat [14]................................18 Figure 7 - 1 G Lunar Simulation [15]....................................19 Figure 8 - Apollo Geology Training [16].................................20 Figure 9 - The NBRF Dive Tank [21]...................................23 Figure 10 - Completed Harnes........................................27 Figure 11 - Balast Atached to Plate....................................29 Figure 12 - Balast Plate Atached to Restraint.............................29 Figure 13 - Hard Plate Retainer with Quick Adjustment Buckle................30 Figure 14 - Partial Gravity Test with Hard Plate System......................31 Figure 15 - Pocket Closed (left) and Open................................34 Figure 16 - Hookah Rig..............................................37 Figure 17 - Ladder Extraction.........................................39 Figure 18 - Safety Diver Extraction.....................................40 Figure 19 - Crane Extraction..........................................41 Figure 20 - Subject on ramp set at 20 degres.............................44 Figure 21 - Pusher..................................................45 Figure 22 - Pusher in C-clamp.........................................46 Figure 23 - Simulation PLS Backpack..................................47 Figure 24 - Camera in Housing........................................48 Figure 25 - Diver Adjusting Balast.....................................50 Figure 26 - Hand Propulsion..........................................55 Figure 27 - Hand Clasped.............................................56 Figure 28 - Photo Used in Repeatability Test..............................61 Figure 29 - MX-2 Space Suit Analog....................................67 Figure 30 - Ranger & MX-2 Working Together............................68 vii List of Abbreviations BAT ? Beam Asembly Teleoperator CG ? Center of Gravity EVA ? ExtraVehicular Activity FM ? Full Face Mask FMARS ? Flashline Mars Arctic Research Station MIT ? Masachusets Institute of Technology NASA ? National Aeronautics & Space Administration NBRF ? Neutral Buoyancy Research Facility NEMO ? NASA Extreme Environment Operations NOA ? National Oceanic & Atmospheric Administration PLS ? Portable Life Support System RATS ? Research And Technology Studies SL ? Space Systems Laboratory 1 Chapter 1: Introduction The Moon is the Earth?s nearest neighbor and the only other planetary body that humans have walked on. The United States is planning on returning to the Moon by the end of the next decade, but by then it wil have been almost 50 years since Gene Cernan left the Moon?s surface. By this point, almost everyone who was involved in the Apollo program wil have retired. In order to succesfully return to the Moon and prepare ourselves for Mars we must relearn how to work efectively in a reduced gravity environment. It is important to gain this knowledge as early as possible in the design proces, so that ?lesons learned? can be efectively applied throughout the Lunar and Martian surface infrastructure. Knowledge of how humans move and work in a reduced gravity field wil be applied to the design of everything from habitats to space suits. Designing systems that maximize the productivity of astronauts during this next phase of planetary exploration is more important than ever. Apollo had the advantage of send newly hardware to the Moon for every mision. This alowed for an incremental learning proces. After each mision astronaut fedback could be used to refine the equipment and procedures for the next mision. NASA?s building block approach wil not alow this to the same extent, as equipment from previous misions wil be reused carying forward any faults in its design. There are several ways to gain the required knowledge. These methods included reviewing the Apollo debriefings, interviewing Apollo astronauts and engineers, and simulating the partial gravity environment on Earth. Among the various simulation 2 methods are parabolic flight, counterweight systems, 1G testing and water-imersion. Water-imersion simulation was chosen for the simulations in this thesis. Chapter 2 wil explain why this choice was made by discusing the benefits and drawbacks to various methods to simulate partial gravity as wel as a look at some past experiments in this field. Chapter 3 wil discus the design and development of the system used in this thesis. This chapter wil include the various trade studies as wel as ?lesons learned? through the development proces. Chapter 4 wil focus on the basic safety procedures including their development and validation. Experimental procedures and equipment wil be covered in chapter 5. The data from these experiments wil be presented in chapter 6. Chapter 7 wil contain data analysis and conclusions. The future work section wil be contained in chapter 8. 3 Chapter 2: Background Material and Literature Review Several methods have been used to study how humans react to partial gravity environments. Each of the methods discussed in this chapter has advantages and drawbacks. They are each useful for researching certain aspects of work in partial gravity. Each method wil be useful in preparing for the first human mision to the Moon in nearly half a century. The methods discused are parabolic flight, counterweight systems, inclined suspension, 1G simulation, and water imersion. After a discussion of simulation techniques, techniques for distributing loads in a backpack wil be discused as an introduction to the experiment caried out for this thesis. Parabolic Flight Parabolic flight is a method to generate reduced gravity conditions by flying a roller coaster style patern as sen in Figure 1. As the aircraft transitions from a nose high path to a nose low path the apparent gravity in the aircraft is reduced. If the aircraft acelerates towards the Earth at the same rate as gravity, then the occupants fel weightles. The aircraft can also choose any partial gravity level including Lunar and Martian gravities. The aircraft creates true partial gravity in that an acelerometer placed in the aircraft wil read just as it would if placed on another planetary body. 4 Figure 1 - Parabolic Flight Trajectory [1] NASA and the US Air Force have been using parabolic flight to investigate space flight since the mid-1950?s. [2] During the Apollo program parabolic flight was used to help train the crew for surface operations. The astronauts found the training to be useful to acquaint them with the fel of the experimental equipment they would be handling. [4] Advantages The primary advantage of parabolic flight is that it creates true partial gravity. It is the highest fidelity mision simulator for partial gravity. The actual flight hardware can be used in parabolic flight tests, which has two primary advantages. First it alows the astronauts to become familiar with fel of the actual hardware. They can se how easy it is to move and use. Secondly parabolic flight alows engineers and astronauts to ensure that al of the interfaces work correctly in the actual gravity environment. Disadvantages The disadvantages of parabolic flight include limitations on mas, volume, and time among others. The tests occur inside of an aircraft, which restricts the size and mas of the equipment that can be used. As can be sen in Figure 2, in their presure suits the 5 astronauts reach almost to the top of the aircraft cabin. Time is also limited during these tests. When flying parabolas to simulate lunar gravity the reduced gravity lasts approximately 28 seconds before the plane must pull up. [2] This precludes testing an entire ExtraVehicular Activity (EVA), which may last as long as 8 hours. Figure 2 - Apolo Parabolic Flight EVA Training [3] At the end of the parabola when the plane pulls up the occupants are subjected to almost twice the normal force of gravity. This creates two problems that must be dealt with. First al of the personnel and equipment must be positioned to endure the pull out force. This maneuvering may cut into an already short test period. And second, the change from low to high gravity causes some personnel to become nauseous, leading to lost testing time. Lastly parabolic testing is expensive. The Apollo era tests were conducted in a KC-135A, which is a modified Boeing airliner. This large aircraft is costly to operate limiting those organizations that can independently conduct parabolic testing. Previous Research During her doctoral work at the Masachusets Institute of Technology (MIT) Dava Newman conducted partial gravity research aboard NASA?s KC-135. The parabolic 6 flight tests were used to calibrate data taken during water imersion trials at the National Aeronautics and Space Administration (NASA) Ames Research Center. The parabolic tests consisted of two subjects running on a treadmil. The treadmil was run at a range of speeds. The tests were run at two diferent gravity levels: lunar and Martian. During the treadmil tests data was collected on the biomechanics of running and walking in reduced gravities. This data included stride frequency, peak force, contact time and aerial time. This research found that peak force and stride frequency decreased as the gravity level decreased. The stride length increased as the gravity level decreased and the contact time was unafected. [5] NASA conducted lunar gravity simulations using parabolic flight as wel. Their tests were mainly aimed at testing specific hardware under lunar gravity as shown in Figure 2. Counterweight Systems Counterweight systems simulate partial gravity by offseting part of the subject?s weight through the use of springs, weights or pneumatics. A counterweight system was used during the Apollo program to alow the astronauts to run through complete tasks in simulated partial gravity an example of which can be sen in Figure 3. 7 Figure 3 - Apolo Counterweight EVA Training [6] Figure 4 ? Example of a Counterweight System [7] Figure 4 depicts a typical design for a counterweight system consisting of a harnes atached to a support point, which can move to maintain the force directly above the subject. This keeps the suspension force vector in direct opposition to the local gravity vector. The simplest way to keep the suspension point above the subject is to have it afixed to an overhead trolley, which is pulled by the subject. The trolley wil naturaly 8 sek the lowest energy position, which is directly above the subject. To keep the efects of the disturbing force the trolley creates low, it must be designed to slide as easily as possible. Methods that have been used to move the suspension point include overhead cranes, air bearing trolleys, and magnetic bearings. Advantages Counterweight systems remove the severe time and space constraints inherent to parabolic simulations. Counterweight tests can run as long as the stamina of the test subject alows. This alows full tasks and EVAs to be simulated without interuption. The size of the simulation area is constrained only by the ability to move the overhead suspension point. This can range from a smal test area with a linear monorail to a virtualy unlimited test area as alowed by the truck-mounted suspension point. Disadvantages The use of a harnes and suspension rig places some limitations on the simulations that can be conducted. In order to alow rotational degres of fredom a complex gimbaled harnes must be constructed. Even with this harnes rotation is usualy limited. The next drawback also involves the harnes. As sen in Figure 3 the harnes is typicaly atached at or near the subject?s center of gravity. This means that the subject?s limbs are not being counterweighted to provide a fully acurate simulation. Another drawback involves the overhead suspension point, which requires al tasks to be setup such that the subject does not pas under or enter into any mockups. This prevents testing of ingres/egres testing of enclosed spacecraft or rovers. The dynamics of the system are 9 also afected. If springs are used the upward force is typicaly not constant and if motors are used the force may be delayed due to the control system implemented. Previous Research Much of the research that has ben conducted using counterweight systems has focused on the use of suspension and treadmils to investigate the mechanics of running in reduced gravity. The use of the treadmil reduces the need for a large overhead rig to move with the subject. The suspension point only needs to move to acount for variations in the subject?s location on the treadmil. The treadmil also alows the subject to run without stopping or changing direction as long as the subject?s endurance lasts. A study conducted at the University of California, Berkeley by Grifin, Tolani, and Kram looked at the way energy is conserved when walking in reduced gravity. As a person walks they convert gravitational potential energy into horizontal kinetic energy through the motion of their center of gravity. This reduces the amount of energy a person must expend to walk. They hypothesized that the recovery of mechanical energy would decrease and the maximum recovery of energy would occur at a lower walking sped as gravity was reduced. They tested their hypothesis using the equipment shown in Figure 4. The experiment only offset the gross body weight using a lower body harnes, so it did not acount for the dynamics of the moving limbs. Data was collected using a 2-axis force sensor mounted under the treadmil. The sensor measured vertical and horizontal ground reaction forces. The results of the experiment proved both points of their hypothesis. [7] 10 Christopher Car conducted research at MIT to investigate how spacesuit legs contribute to energy recovery during walking and running. This research was conducted using a lower body exoskeleton with a rigid knee. This fixed exoskeleton acted like a spring much in the same way that the presurized tube of a spacesuit leg does. Both the exoskeleton and the space suit leg sek to remain in a fully straight orientation. Car?s research focused on the energy recovery, the metabolic energy required to walk/run, and the walk/run transition point. The results were described using thre quantities: Froude number, specific resistance, and cost of transport. The Froude number is a non- dimensional way of representing velocity and is defined as: ! F r = v 2 g?L Where v is velocity, g is the gravity level, and L is the subject?s leg length. Specific resistance is the eficiency of motion per unit distance in units of: ! J N?m Cost of transport is the energy expended to cary a unit mas a unit distance with units: ! J kg?m The results of his research showed that the cost of transportation, specific resistance, and the Froude number of the run-walk transition were al decreased with the use of the exoskeleton and energy recovery was increased. This implies that the exoskeleton and therefore a space suit can be beneficial for transportation in reduced gravities. The reduced Froude number of transition shows that humans would be more likely to run when in a reduced gravity environment. [8] 11 NASA, in support of the constelation program, is using a suspension system to test how far an astronaut can walk back in the event of a rover breakdown. This research alows engineers to size the thermal system of the suits to handle the increased heat generated due to a sustained run. The test is being conducted using the MK II space suit over a 10 km distance. [9] Inclined Suspension Inclined suspension simulates partial gravity by rotating the floor such that the component of gravity perpendicular to the floor is equal to the desired gravity. The angle required is calculated from the following: ! "=cos #1 g Desired ( ) Where theta is the angle betwen the local horizontal and the inclined surface and the desired gravity is a fraction of 1. In the case of the Moon, the desired gravity is 1/6. For lunar gravity the floor is rotated 80.5 degres from its normal orientation. Since the subject would naturaly fal off of such a step surface a harnes is used to offset the rest of the gravitational force. This harnes must be designed to support the entire body so that no unnatural torques are applied to the subject. It must also be designed to alow the legs to move unimpeded and independently. Figure 5 shows a typical inclined suspension rig from the Apollo era. 12 Figure 5 - Typical Inclined Suspension Rig [10] Advantages When the inclined platform is curved so that it forms a closed circuit the subject has an unlimited distance that can be traversed while staying in a relatively smal area. This alows testing of walkback distances and metabolic rates. Walkback distances were the constraining factor when using the Apollo lunar rover and this is not likely to change if a similar rover is used in the future. This wil be an important test for future spacesuit designs. Disadvantages The harneses typicaly used for inclined suspension tests do not alow for much rotation from a standing posture; this limits tests mainly to mobility studies. Even for the limited positions they alow the harneses required for inclined suspension are extremely 13 complex. As can be sen in Figure 5 the left and right leg must be balanced independently so that the subject can walk. To prevent interference the high leg is in a sling while the lower leg is atached to a rod that runs up the back. The suspension lines typicaly run from the subject up to a fixed point above. This introduces a pendulum dynamic that would not be found in true partial gravity. As the subject moves away from the platform the lines no longer pull directly paralel to the platform. The shorter the suspension lines the larger this efect becomes, so typicaly large towers are built to support the subject. Previous Research NASA has conducted research using inclined simulators. One study looked at the mobility efects of lunar versus Earth gravity and a presurized space suit versus an unpresurized space suit. The study had the subjects complete general motions such as walking, running, jumping, and climbing. For the walking and running tests the subjects selected their own velocities. They were told to select a comfortable pace. The subjects could not obtain maximum possible running velocity due to the short traverse distance of 20 fet alowed by the simulator. The maximum velocity obtained by the subjects while running in lunar gravity was approximately 50% of that obtained in Earth gravity. The authors atribute the reduction to reduced traction due to the reduced normal force. Presurizing the space suit reduced the speed under both gravity levels by approximately 25%. During the jumping trials, as expected the subjects could jump higher and farther in lunar gravity than in Earth gravity. The height was 6 ? 7 times higher for jumping straight up and the subjects jumped twice as far horizontaly in a standing broad jump. The unpresurized suit reduced the height by 10% - 15% when compared to stret clothes 14 and presurizing the suit further reduced the altitude by 30%. The climbing tests involved a ladder, a staircase, and a pole. The subjects were capable of ascending and descending the stairs easily in lunar gravity, but could not se the stairs while descending requiring careful concentration. While in Earth gravity one subject atempted to climb the stairs. The subject was succesful with the unpresurized suit, but unable to climb even one step with the suit presurized. The subjects were able to climb the ladder under al suit conditions in lunar gravity, but modifications to their technique was required when the suit was presurized. Due to the mobility of the suit it was dificult to grasp the rungs while the suit was presurized. Instead the subjects grabbed on the back of the ladder uprights as with did not require them to rol their hands from the natural position of the suit. Climbing the ladder in Earth gravity with the suit presurized was posible, but dificult and fatiguing. The subjects were unable to climb a pole in Earth gravity, but found the task easy in lunar gravity. [11] Water Imersion Water Imersion uses the natural buoyancy of water to offset the force of gravity. This type of simulation can be used for simulating any gravity level les than 1-G. Foam or weights are added to mockups and subjects to achieve the desired gravity level. In water imersion testing a mockup?s rotational characteristics must be checked in addition to its weight so that the mockup does not continualy atempt to return to a ?prefered? orientation. To achieve a rotationaly neutral object, one that does not rotate when released, the center of buoyancy must be place coincidental with the center of gravity (CG). If this is not done the object wil rotate so that the center of buoyancy is directly above the center of gravity. 15 Advantages Water imersion systems alow for a self-contained test subject unlike either of the suspension techniques. This alows for a full six degre-of-fredom simulation and alows for entry into mockups to test airlock hatches and ingres/egres procedures. Balast may be added to a subject in several locations alowing for a more acurate weighting than typicaly found in counterweight systems, which offset gross body weight. Any equipment mockups that are used in the simulation can also be weighted to achieve the desired gravity level alowing astronauts to train with equipment of realistic size and mas. This alows them to learn the best way to pick up an object and handle it. The duration of tests while limited by air supplies is significantly longer than that found in parabolic flight studies. Disadvantages The major disadvantage in water imersion is due to the water itself. The water adds drag to al movements, so that static or slow moving tests are most acurately simulated. Static tests involving posture or stability are wel suited to water imersion, as are slow moving tests such as ingres/egres studies. While underwater the human body is very close to neutral, so balast must be added to achieve the desired gravity. While the apparent weight is correct the inertial mas is higher than would be encounter in actual reduced gravity. This added mas efects creates higher inertial forces that must be overcome during any dynamic testing. The water also adds a life support concern; the human body was not designed to operate underwater so equipment must be worn to sustain life. This requires that al subjects be certified to participate in dive operations 16 and the use of safety divers to ensure the subjects have air at al times during the simulation. Previous Research As was mentioned earlier, Dava Newman of MIT conducted research using water imersion as wel as parabolic flights. The underwater trials consisted of six subjects running on an underwater treadmil. The subjects were balasted to simulate five diferent gravity levels ranging from 0G to approximately 1G. The subjects ran at thre diferent speeds under each gravity condition. A 1G land based control was also performed. The data measured during these tests included stride frequency, contact time, peak force, and stride length. The same trends found in the parabolic flight data held for the water imersion research as wel. The data however was not an exact match. The stride frequency was lower and the contact time was longer in the water imersion study than in the parabolic flights. These efects were atributed to the increased mas and inertia due to the balast required for the water imersion study. [5] Leslie Wickman and Bernadete Luna expanded upon the research conducted by Newman. They added a variable CG backpack to their study to investigate carying additional loads in reduced gravity environments. Their study consisted of six subjects walking and running on treadmils in lunar, Martian, and Earth gravity levels. The Earth gravity trials were land based with the two reduced gravity trials conducted underwater. The maximum weight of the backpack was determined by the subject?s comfort level. This weight consisted of 270% of the subject?s 1G-body weight in lunar gravity, 80% in Martian gravity, and 45% in the 1G trials. The data consisted of energetics results and 17 mechanics results such as stride length and stride frequency. The mechanics results followed the same trends as found in the Newman data. The energetics results were used to extrapolate the amount of load that a human could cary for an eight-hour workday in each gravity level and to develop a model for energy expended. The maximum load for al day work was selected utilizing the subject?s maximum oxygen uptake capacity. Other research has suggested that 35% to 50% of the maximum oxygen uptake is sustainable for an eight-hour workday. Utilizing this oxygen range, the data showed that for lunar gravity a person should be able to sustain a load factor of 2.7 while walking. Load factor is defined as the backpack mas plus the subject?s body mas divided by their body mas. In others words they should be able to cary 170% of their body weight for one EVA. For Martian gravity this drops to 50% and in Earth gravity 20%. Their energy model was developed using regresion analysis. The parameters of the model included velocity, gravity level, load factor, body mas, and leg length. [12] NASA has ben conducting water imersion simulation studies through the NASA Extreme Environment Mision Operations (NEMO) project. The NEMO misions are intended to serve as an analog to long duration spaceflight. They take place at the National Oceanic and Atmospheric Administration (NOA) Aquarius undersea habitat. The NOA Aquarius is a 15 meter by 4.5 meter cylinder similar in size to the U.S. Lab module on the International Space Station (IS). The habitat is located in 60 fet of water off the coast of Key Largo, FL. It is a presurized habitat, so the presure inside is slightly higher than the surrounding presure keeping the water out. Research conducted during the NEMO project has included telemedicine, surface robotics, and partial 18 gravity EVA. Using hardhat diving equipment divers have caried out tasks with diferent CG conditions to determine stability. [13] Figure 6 - Aquarius Undersea Habitat [14] Aquarius images are copyright ? 207, University of North Carolina Wilmington. Al rights reserved. Further reproduction prohibited. 1G Simulation 1G simulation takes a completely diferent approach than al of the preceding methods. This method ignores reduced gravity entirely. Subjects use hardware that is similar to the real equipment in size and shape, but not in mas. Using 1G simulation eliminates al of the hardware atached to the subject to offload gravity. They are fre to move about, but must contend with full gravity. These simulations are useful for procedures training and team building. Advantages The major advantage to 1G simulation is the elimination of simulation hardware atached to the subject. The subject is not encumbered by a harnes or cable system that limits 19 their mobility. Special training in the use of simulation hardware is also not required. This alows subjects to concentrate on learning the task at hand. Disadvantages The disadvantage to 1G simulation is the lack of any atempt at partial gravity. This simulation is useful for mision training and task planning, but shows nothing about human capabilities in partial gravity. Fatigue may be a major factor in any 1G simulation especialy if spacesuits are used as in Figure 7. Not only is the subject contending with their own body weight they also fel the full weight of the suit. For this reason lightweight mockups are sometimes used such as in Figure 8. Figure 7 - 1 G Lunar Simulation [15] 20 Figure 8 - Apolo Geology Training [16] Previous Research Several organizations are currently conducting 1G simulations geared towards preparing for the next generation of planetary exploration. Among these are NASA and the Mars Society. The Mars Society is currently funding two simulations. One is caled the Flashline Mars Arctic Research Station (FMARS) and is located on Devon Island near the Arctic Ocean. The simulation is based here because of the geologic and temperature similarities to Mars. The goal of the simulation is to conduct research with the same constraints found during a Mars surface mision. The Mars Society also runs Mars Desert Research Station in Utah. Simulations conducted here serve the purpose of field- testing equipment and procedures before they are utilized at the more remote simulation sites. [17] [18] NASA is conducting its own simulations under a program caled the Desert Research And Technology Studies (RATS). The Desert RATS brings together equipment and 21 personnel from across NASA and its contractors to test them in an integrated field environment. In 2005 a series of tests were conducted combing two subjects working in presurized space suits and a manned rover. This combination alowed for testing procedures that wil be needed for two person EVA?s on either the moon or Mars. This data wil alow engineer to beter timeline the tasks involved. Recharging stations were also used to increase the amount of time that could be spent in the field. This is likely to be a common occurrence in future exploration. Insights were gained into how the stations should be designed including the need for the station to support the subjects to reduce fatigue. The 2005 Desert RATS also was the first time since Apollo that two suited crew members worked simultaneously. [19] Backpack Loading The cariage of external loads has been looked at for subjects in Earth gravity. The results found in the literature do not sem to form a solid conclusion about the best place to locate the center of gravity of the pack. It sems intuitive that the load should be placed as close to the subject?s back as posible to minimize the disturbance from their normal unloaded posture. What is not intuitive is whether that load should be placed high in the pack or low in the back or even somewhere in betwen. Research in the literature has shown both high and low CG packs to have advantages. The high CG pack minimizes the amount the subject must lean forward to bring the combined CG of themselves and the pack over their fet. The low CG pack tends to reduce the amount sway sen in a standing subject as compared to a high CG. This is interpreted as a sign of 22 greater stability. The downside of the low CG pack is that the subjects tend to lean such that the load is over the front half of the foot increasing the chance of injury to the foot. [20] 23 Chapter 3: Partial Gravity System Design & Development The first step in designing a partial gravity simulation is to decide the method that wil be used from those listed in the previous chapter. At the Space Systems Laboratory (SL) this was an easy decision. Parabolic flight is too costly and aces to the proper aircraft is very limited. The two suspension techniques would have required the development of a complex simulation system, which is unnecesary because the SL is located at a water imersion facility. The Neutral Buoyancy Research Facility (NBRF) is a 25-foot deep 50-foot diameter pool used for human factors and space robotics research, as shown in Figure 9. For this reason water imersion was the chosen method of simulation. Figure 9 - The NBRF Dive Tank [21] The human body has a density similar to that of water; because of this, balast must be added to test subjects so that they have the correct apparent weight for the desired gravity level. The total balast required is easily determined by taking a subject?s Earth weight and multiplying it by the desired gravity level. Adding this amount of balast to the subject would cause a scale to read correctly underwater, however the balast cannot be 24 added just anywhere. It needs to be correctly distributed on the subject so that their CG is maintained and body parts balasted proportionaly. The corect distribution is given body segment parameters. This is a set of data that shows the percent of the total mas that is contributed by each body part or segment. Distributing the balast acording to body segment parameters yields a subject with the correct center of gravity and who wil react correctly when moving, bending, or performing other tasks. These parameters are diferent for each individual, but their measurement is outside the scope of this thesis. Instead a generic set of parameters published by de Leva are being used. [22] Utilizing every segment presented in this paper would generate an extremely acurate simulation, but would be completely impractical. Table 1 shows the breakdown of body segments. The two segments chosen for this research were the trunk and the thigh. These two segments acount for 43.46% and 14.16% respectively. For Lunar gravity al of the other segments require approximately 2 pounds or les. This was considered low enough such that adding more segments would not significantly add to the acuracy of the simulation. Table 1 - Body Segment Parameters [2] Body Segment % of Body Mas Lunar Balast for 175 lb subject Mars Balast for 175 lb subject Head & Neck 6.94 2.02 4.5 Trunk 43.46 12.68 28.52 Upper Arm 2.71 0.79 1.78 Forearm 1.62 0.47 1.06 Hand 0.61 0.18 0.40 Thigh 14.16 4.13 9.29 Shank 4.3 1.26 2.84 Fot 1.37 0.40 0.90 Once the segments were chosen the total mas of the subject stil had to be acounted for by adding extra balast to the chosen segments. The amount to add to each chosen segment was calculated by dividing the percentage contribution of each segment by the 25 total contributed by the trunk and thighs. For example the trunk and thighs acount for 57.62% of the body mas while the trunk alone contributes 43.46% therefore the trunk acounts for 75% of the simulated segments. For a 175 lb subject 24 lbs are placed on the trunk and 8 lbs are placed on the thighs. Now that the necesary amount of balast has been calculated a way to atach it to the subject must be designed. The system must be designed to satisfy two important but competing objectives. The first objective is to maximize safety. Having a balast system that is easy to release in the event of an emergency satisfies this objective. The second objective is that the balast is secure. If the balast is loosely sliding on the subject or spontaneously jetisoning this wil be detrimental to the simulation?s acuracy. The design of the system developed for this thesis can be broken into two segments: the balast system and the balast release mechanism. Ballast System Development The balast system is the system that is physicaly worn by the subject and it supports the balast release mechanism. During the development of the balast system two basic concepts were considered: a jumpsuit and a harnes. Jumpsuit The jumpsuit was envisioned as a full body suit with pockets sewn in to hold balast in the appropriate locations. The jumpsuit would have been created from a wetsuit, which is designed to be close fiting. The jumpsuit concept has several advantages that made it an atractive option. The primary advantage was the security with which the pockets would 26 be held to the subject due to the form fiting nature of the suit. This also is the primary drawback of the concept. The form-fiting suit would be dificult to adjust to fit a range of body sizes. A solution to this problem would be to make multiple suits, but this would be prohibitive for cost and time reasons. The other disadvantages of the jumpsuit include the inability to adjust the pockets and its inherent buoyancy. The pockets would be sewn into the suit, which would preclude the ability to either resize or move the pockets relative to each other. Resizing of the pockets is important because if the pockets were too big the balast would slide exerting a destabilizing force that could impact the data. If the pockets were on the smal side this would limit either the mas of the subject or the gravity level that the system could handle. The use of a wetsuit as the jumpsuit would add buoyancy to the subject due to the material they are made from. This would require the addition of more balast to compensate increasing the mas of the subject-system combination. Harness The harnes concept resembles a full-body harnes of the type that would be found in rock climbing or fal-arest gear. Like either of these systems the harnes is adjustable to fit a wide range of subjects. The harnes contains six adjustments to ensure a proper fit to the subject. There are thre adjustments in the upper harnes and thre in the lower harnes. The upper harnes has an adjustment on each shoulder to raise and lower the torso balast evenly. There is also a chest band that is tightened to ensure the torso balast is held close to the body. The lower harnes adjusts to the diameter of the subject?s legs and has a waist adjustment. Another benefit of the harnes comes from the fact that it is modeled after a rock climbing harnes. This alows the subject to be lifted by the 27 harnes, an advantage that wil be shown during the next chapter?s discussion of emergency procedures. The third and final major advantage of the harnes is due to its lack of inherent buoyancy. This reduces the balast required when compared to the jumpsuit concept alowing for les mas to be caried by the subject. The completed harnes is shown in Figure 10. Figure 10 - Completed Harnes The large amount of adjustability has also proven to be a detriment in some cases. Some of the adjustments have a tendency to crep over time, which loosens the harnes. The crep sems to be due primarily to the type of adjustment hardware chosen. Each of the adjustments uses a quick adjustment buckle similar to those found in skydiving harneses. These were chosen for their ease of adjustability and proven security in their primary application. During testing it was discovered that if the extra webbing 28 protruding from the atachment is not held tight to the subject the fitings tend to loosen. The crep is combated through careful donning of the harnes and vigilance of the test director in keeping the fre ends of the webbing secure. To help hold down the webbing ends elastic loops were added to the harnes through which the webbing is fed. Ballast Atachment Development The balast atachment is atached to the harnes and is what physicaly holds the balast in place on the subject. During the development of the balast atachment two concepts were tested and evaluated: a hard plate system and a pocket system. Hard Plate System The hard plate system was the first concept that was developed and tested. It consisted of a rigid metal plate to which weights were bolted shown in Figure 11. The plate had a tab on one end that was inserted into a webbing slot. The opposing end of the plate had a hole though which a clevis pin was inserted. Through the clevis pin was a hitch pin holding the plate in place. The plate and atachment mechanism are shown in Figure 12. When the hitch pin is pulled gravity causes the plate to slide off of the clevis pin, the tab slides out of the slot and the plate fals fre. This system was envisioned as a way to rigidly atach the balast to the subject while stil alowing it to be easily removed in the event of an emergency. Once the concept was developed it was tested using a demonstration test rig as wel as a human subject. 29 Figure 1 - Balast Atached to Plate Figure 12 - Balast Plate Atached to Restraint The demonstration test rig consisted of a metal backing plate to which webbing was atached containing the slot and clevis pin as sen in Figure 12. The test rig was used underwater using two diferent balast levels. The first test used a two-pound weight atached to the plate. The purpose of this test was to validate the basic concept and to ensure that the system behaved in a predictable manner before increasing the balast. The 30 test was conducted by clamping the backing plate to a mockup that was already residing in the dive tank. After the backing plate was secured the balast plate was instaled and the test diver floated above while the plate was released. This setup kept the test diver out of harms way during the test sequence. During the test the plate was easy to instal and jetison and followed a predictable path once released. The one problem discovered during this test was that the plate followed a ?faling leaf? patern once released and would strike a subject in the legs if they were wearing this release system. This was thought to be due to the low weight being used and the high surface area of the plate. For this reason the release was tested with a heavier weight. For the second test sesion the plate was modified to hold one ten-pound weight. The ten-pound weight was tested in the same manner as the two-pound weight. The test confirmed that the ?faling leaf? patern was due to the low weight used during the previous test. The plate fel in a manner that was both predictable and safe for the test diver. The main problem uncovered during the test was that it was dificult to instal the plate if the distance betwen the slot and the clevis pin was not precisely right. This problem was addresed by adding an adjustment betwen the clevis pin and the slot as shown in Figure 13. Figure 13 - Hard Plate Retainer with Quick Adjustment Buckle 31 Figure 14 - Partial Gravity Test with Hard Plate System After the demonstration rig showed that the concept was feasible, the hard plate system was integrated into the harnes described earlier. The integrated system is shown in Figure 14. A test dive was conducted to se how the hard plate system functioned on a test subject. The dive was conducted with the subject and the lead diver utilizing Full Face Masks (FM). The FMs alowed for constant verbal communication betwen the subject and lead diver alowing them to work though any problems that were encountered during the test. The test dive did not go as smoothly as the dives with the demonstration rig. Two major problems and one nuisance contributed to the dificulties encountered during the test. The problem was that the plates were not as easy to atach as they were with the demonstration rig. This was most pronounced on the thigh balast and was due primarily to the curvature of the body. This forced the safety divers to have to bend the plates as they inserted them, which was made dificult by the rigid weight that was bolted 32 to the plate. The plates were also prone to slipping out of the slots, which appeared to be a combination of the curve of the plate and the narownes of the slots. The slots were made to fit the tab tightly so that the plates would be held firmly. Instead the narow slots prevented the tabs from fully inserting. The nuisance found during the test was the sharp edges on the plates. An atempt to mitigate this was made using plastic tubing around the edges, but this tubing did not remain in place during the test. For these reasons and the fact that a diferent set of plates would have to be made for each test subject it was decided to discard the hard plate system. Pocket System Once it was discovered that the hard plate system was not feasible atention was turned to designing a soft pocket system. This system would use pockets containing lead shot pouches in the same location as the plates. This system would have some advantages over the hard plate system, such as ease of balast adjustment and the use of soft weights themselves. Through the use of diferent sized shot pouches the amount of balast can be quickly varied for subjects of diferent mases. The soft weights themselves are also safer in that they wil not hurt if they land on a subject?s foot; they simply conform to the shape. The major concern with the pockets is that the balast wil slide around upseting the stability of the test subject. This problem is managed by keeping the pockets close to the size of the balast that wil be in them. Two diferent release concepts were conceived for the pockets: a Velcro release and a ripcord release. The two concepts were the same in al respects except at the bottom of the pocket where the balast would be released during an emergency. The balast is loaded into the top of the pocket and this opening is sealed using a Velcro flap. 33 The ripcord pocket was conceived as having the bottom front corner form a hinge. Through this hinge would be inserted either a rod or a cable holding the bottom flap in place. When the cable or rod was pulled fre, gravity would cause the bottom of the pocket to fal away and the balast would be released. The rod and cable each have advantages and disadvantages and a final decision would have been based on trial runs. The rod would have likely held the load easily, but the length required for the torso pocket could have prevented the subject from completely extracting it in one motion. The cable on the other hand would not have suffered from this problem, but substitutes the problem of possibly sagging under the load of the balast. This would create a large pull force. The concept was considered in case the Velcro pocket proved to have either too high of pull force or insufficient holding power to restrain the balast. The Velcro pocket was the primary pocket design and was the only one tested. The Velcro release consists of two flaps, one on the front of the pocket and one that includes the bottom side of the pocket. The upper flap covers the bottom flap, so that when the upper flap is pulled up the Velcro disengages and the balast is released. Figure 15 shows the pocket with the flap closed and the flap open; the D-ring is used as a handle. In the open photo you can se the balast weight starting to exit the pocket. 34 Figure 15 - Pocket Closed (left) and Open 35 Chapter 4: Procedures Development An important part of a system such as the one discussed in the previous chapter is a complete set of procedures for handling both nominal and off-nominal conditions. The nominal procedures serve to keep al divers on the same page which increases productivity and safety. The emergency procedures are in place so that in the event of an off-nominal condition everyone involved conducts a coordinated response. The development and testing of those procedures wil be discussed in this chapter. Nominal Operations This section wil discuss the nominal dive operation of the SL as wel as the procedures specific to the research being discussed. Al dives at the SL require thre personnel: a lead diver, a deck chief, and a second diver. Anyone who is a diver at the SL must met certain basic requirements. Among these requirements are SCUBA certification, first aid & CPR training, and training as an oxygen provider. A physical is also required. After these requirements are met a checkout dive is conducted testing the diver?s basic dive skils. Once this checkout is complete they may work as a diver in the lab. Once a diver is cleared to work in the dive tank they are alowed to participate in the deck chief course. The deck chief is responsible for the safety of everyone involved in a dive. They are the final word on al safety related maters. They monitor the dive from the surface through the use of cameras placed in the dive tank. They also have the ability to communicate with the divers through the use of an intercom and underwater speaker. The deck chief course is designed to teach the divers acident management. During the course acident scenarios are conducted and the student working as the deck chief must respond with the 36 appropriate action. The lead diver is a SL diver with 30 hours of experience diving at the lab in a variety of test conditions. They are responsible for keeping an eye on the divers in the water as wel as ensuring that al divers check their presure gauges. During test runs of the partial gravity simulation system the subject is added to the team. The subject is required to be a diver cleared to dive at the SL, so they are already familiar with the rules and procedures of the facility. The two other divers are safety divers ready to respond to any trouble the subject may have. If needed to satisfy the objectives of the dive a utility diver can be added to handle tasks such as logistics (retrieving needed equipment from the surface) and documentation (photographic and videographic). Each dive begins with a briefing of the test objectives. This informs the deck chief of what they should expect to se throughout the dive, so that they can more easily recognize off nominal situations. The briefing also informs each diver of the specific role they wil be playing in the dive and outlines their responsibilities. The briefing is the time to answer any questions that any of the personnel may have. If anyone is not completely comfortable with the tasks and procedures outlined the plan is reworked or the task may even be removed from the plan. Once everyone has agred on the dive plan the divers enter the water and gear up. The subject dons the balast harnes with no balast instaled before entering the water. Once everyone is ready the divers descend to the bottom of the dive tank. The balast is caried down using a basket and a lift bag. The lift bag contains air that ofsets the weight of the balast, so the diver can descend at a normal pace. Once at the bottom the lead diver ensures that everyone is comfortable, after this the divers prepare to start the test. The subject removes their 37 scuba unit and switches to the hookah rig. The hookah rig is a regulator atached to a SCUBA cylinder by a 25-foot long hose shown in Figure 16. The rig supplies the subject with air while not encumbering them during the trial. Figure 16 - Hokah Rig The subject?s fins are also removed at this time. At this point the subject is relying on the safety divers, as it would be very dificult for the subject to swim to the surface without fins. With the subject breathing comfortably from the hookah rig the safety divers place the appropriate amount of balast in the pockets of the harnes. The test run may now be conducted. Once the test run is over this proces is reversed. The balast is removed from the pockets using the quick release mechanism. The subject replaces their fins and scuba unit and transfers from the hookah to breathing from the SCUBA cylinder on their back. Once this transition is complete al of the equipment used in the test is gathered and the dive team ascends to the surface. At the surface al of the equipment used is removed from the dive tank and placed on the deck. The divers exit the water and the debriefing is conducted. During the debriefing information is collected on what went wel and what can be improved for the next test as wel as the subject?s impresions and observations during the test. 38 Emergency Procedures In any emergency it is important to have a pre-defined response, so that al team embers involved wil know their roles and what response to expect from others. To this end a list of the possible failure modes and responses to these failures has been developed and wil be discussed in this section. There are two major factors that afect the response used for an emergency with the partial gravity system. They are whether or not the subject is able to asist in the response and whether or not the balast has been released. The combination of these two factors results in four categories each of which has its own method for extracting the subject from the dive tank as sen in Table 2. Table 2 - Subject Extraction Breakdown Ballast Released Ballast Atached Subject Asisting Normal Extraction Ladder Extraction Subject Unable to Asist Safety Divers Swim Subject Crane Extraction Normal Extraction This extraction covers two main families of emergency: subject discomfort and loss of breathing gas. The first family is very broad, but at anytime if the subject fels uncomfortable the test is ended and the normal end of test extraction procedure is caried out. The second family, loss of breathing gas, requires imediate action. The first step is providing the subject with an alternate source of air. This can occur in one of two ways. Either through a safety divers alternate air source or through a Spare Air bottle. 39 Each of the safety divers has a second regulator that can be donated to the subject in an emergency. The subject is also wearing a 3.0 cubic foot tank with integrated regulator on their harnes, which wil provide enough air until the subject is either given a safety diver?s alternate or the tank they descended on. Once breathing gas is restored the balast can be released, the subject is transfered to the tank they descended on, and a normal extraction is used. Ladder Extraction If during either of the two scenarios in the previous section the balast cannot be released a ladder extraction is used. Atached to the side of the dive tank is a ladder that reaches from the bottom of the tank to the surface shown in Figure 17. The subject can use this ladder to climb out of the dive tank. If the subject is not wearing their air supply during the extraction the safety diver wil be able to cary it while monitoring the subject as they climb. Figure 17 - Lader Extraction 40 Safety Diver Extraction This extraction would be used in the event that the subjects were unable to asist in their own extraction. This would be most likely due to a medical emergency. During this extraction the safety divers would first remove the balast from the harnes. This lightens the subject so that it is possible to swim them to the surface. Then the two safety divers would swim the subject to the surface as shown in Figure 18. As the hose is 25 fet long, it is possible to reach the surface without carying the hookah rig up with the subject. Figure 18 - Safety Diver Extraction Crane Extraction If the diver is incapacitated and the balast is unable to be removed, the subject may be extracted by using the overhead crane. This is required, as it has been shown in testing to be extremely dificult for the safety divers to swim a fully balasted subject to the surface. Even if the safety divers succeded the subject would have no means of flotation to 41 remain at the surface. The front of the harnes has a ring such as those found on fal- arest harneses. This ring is connected to the crane and the subject is lifted to the surface. The crane and atachment point can be sen in Figure 19. Figure 19 - Crane Extraction 42 Chapter 5: Experimental Procedures & Hardware This chapter wil describe the procedures and hardware used during data collection. The experiment consisted of two phases: the orientation phase and the data collection phase. The orientation phase consisted of one dive to familiarize the test subject with the hardware and the emergency procedures. The data collection phase took place over two or thre dives per subject. Orientation Dive The orientation phase consisted of one dive for each of the subjects. This dive served to familiarize the subject with the hardware and the emergency procedures. The dive was conducted in acordance with the nominal procedure for descent and ascent outlined in the previous chapter. Once the subject had removed their SCUBA unit and was comfortably breathing off of the hookah their neutral buoyancy was tested. For this they stood on the bottom of the dive tank without balast and breathed normaly. If they are approximately neutral they wil ascend as they inhale and descend as they exhale. The buoyancy of the test subject changes while they breathe due to their chest expanding and displacing more water when they inhale, with the opposite happening as they exhale. Weights can be removed or added to a belt around the subject?s waist until the desired efect is achieved. Once the subject was neutral the proper amount of balast was added to the pockets of the harnes. For al of the orientation dives the proper amount of balast for lunar gravity was used. The subject was then alowed to walk around and aclimatize to the partial gravity environment. Once the subject felt comfortable the subject and a safety diver jetisoned the balast, so the subject could get a fel for how the pockets 43 operated. Once the releases were reset and the balast was replaced in the pockets the subject practiced bailing out to their onboard alternate air supply, the Spare Air. With a fel for how the regulator breathes on the Spare Air bottle the subject practiced receiving the octopus regulator of one of the safety divers. With a mastery of the techniques required in a loss of breathing gas emergency the subject was ready to practice the ladder extraction. This test gave the subject a sense of how much traction and balance they had on the exit ladder. One safety diver monitored them while the other diver caried the hookah rig towards the surface. The crane and swiming extractions are not practiced as they asume the subject is incapacitated and therefore require no action on the part of the test subject. With these scenarios completed the subject was given further time air supply and time permiting to become further acquainted with the sensation of partial gravity. After this final familiarization period the dive was terminated and the nominal ascent procedures were completed. Data Collection Dives Data was collected on the stability of the subjects standing on various slopes under diferent loading conditions. Besides the balast system, four major components were used in the data collection proces: a ramp, a pusher, a backpack, and a video camera. The components are used to collect data by disturbing the test subject while they are standing at rest. The results of the disturbance are recorded by the video camera for post- dive analysis. 44 Ramp The ramp shown in Figure 20 can be adjusted to approximately ten, fiften, or twenty degres. The maximum angle of the ramp was set at twenty degres due to proposed OSHA regulation 1910.26. This regulation cals for a railing on any ramp with an angle greater than 20 degres. Since this ramp would not have a railing this was deemed an appropriate maximum slope to test. The ramp?s center section is covered with strips of non-skid tape as it was found during early testing that the ramp?s surface did not provide enough consistent traction for the experiment. The plate is reinforced with two bars that run the length of the ramp. The reinforcement reduces the amount of flex in the ramp so that the subject does not bounce during the trials. Figure 20 - Subject on ramp set at 20 degres Pusher The purpose of the pusher shown in Figure 21 is to provide a consistent disturbance to the subject. The pusher consists of a thre main components: a length of two-inch square rapid prototyping extrusion beam, the pushing asembly and the latch. The extrusion acts as the handle with the latch and pushing asembly atached to it. The pushing asembly 45 consists of a plastic rod, a spring, the pushing surface, and a guide bolt. The plastic rod is held in the center of the extrusion using by the guide bolt at the back end and a thin plate atached to the front of the extrusion. The plastic rod slides through the thin plate imparting the disturbance to the test subject. The latch consists of an extrusion joining plate into which a notch has been cut. The head of the bolt on the pushing asembly engages the notch holding the pushing asembly in the ready position. When the test diver releases the latch, the spring forces the pushing asembly forward imparting the disturbance to the test subject. Figure 21 - Pusher Two diferent springs were tested for the pusher. The first spring alowed the test diver to easily compres the pusher and latch it without outside asistance. However this spring did not appear to provide enough force. For this reason a stifer spring was substituted. This second spring provided enough force to the subjects, but proved to be nearly impossible for the test diver alone to compres. To solve this problem the C- clamp shown in Figure 22 was used. This alowed the test diver to compres and latch the pusher at his own pace. 46 Figure 2 - Pusher in C-clamp Simulation PLSS Backpack The backpack sen in Figure 23 was used to simulate a Portable Life Support System (PLS) such as those used on current spacesuits. The mas and CG of the backpack can be adjusted using the two extrusion beams shown in the left of the photo. The harnes and tubular frame of the backpack were originaly part of an external frame backpack. The harnes consists of two padded adjustable shoulder straps and a padded adjustable waist belt with quick release buckle. The configuration shown weighs seven pounds and was used in the lunar gravity simulation equating to a forty-two pound PLS. While this is significantly lighter than current PLS it is within the goal range of next generation designs. [23] The horizontal beam moves up and down to adjust the CG verticaly. To adjust the CG in the horizontal direction weights are atached to the horizontal beam and slid towards or away from the test subject. 47 Figure 23 - Simulation PLS Backpack The balast on the subject?s torso can stil be jetisoned with the backpack in place. The only requirement is that the safety divers ensure that the release D-rings are pulled out from the padded waistband. Video Camera A video camera was used to collect data during the experimental runs. The camera used was a Sony DCR-TRV 950 Digital Video Camera Recorder. The data was recorded onto sixty-minute Mini-DV casetes. The camera was housed in an Ikelite Digital housing to protect it during the underwater trials. A mount was built and atached to the housing alowing it to be clamped around a mockup that was in the dive tank. This provided a stable base and ensured the camera was the same distance from the ramp during al trials. The camera, housing, and mount are shown in Figure 24. 48 Figure 24 - Camera in Housing Experimental Procedures Thre diferent weighting configurations were completed with each subject over the course of two dives per subject. Al thre configurations were caried out in lunar gravity, so that the pockets were able to hold the appropriate balast for the full range of subjects. The two lightest subjects were put through thre weight configurations in Mars gravity for comparison. To acommodate simulating heavier subjects in Martian gravity larger pockets need to be constructed. The current pocket size was chosen to hold lunar balast for the heaviest subject while limiting how much the balast would be able to shift during tests of the lightest subject. The thre weighting configurations for the Lunar tests were as follows: no backpack, seven pound backpack with high CG, and seven pound backpack with low CG. The backpack CG was located 4.3 inches from the subject?s back in both trials. For the high CG configuration the CG was 13.3 inches above the middle of the waistband of the backpack. The low CG location placed the CG 7.7 inches above the middle of the backpack waistband. For the Mars gravity trials the thre 49 weighting configurations were no backpack, seventen-pound backpack with high CG, and seventen-pound backpack with low CG. The backpack trials had approximately the same CG in both the Lunar and Mars cases and the Mars backpack weight was chosen to simulate backpack of equivalent mas to the Lunar trials. These weighting conditions represent only a very smal fraction of the data that can be collected once the system and data collection procedures are perfected. The no backpack trial was conducted first with al subjects while the order of the backpack trials was not set. Each trial contained six diferent orientations that the subjects were placed in while data was collected. Two of these were on a zero degre slope in which the subjects were pushed from the front and from the back. The remaining four orientations were on a twenty-degre slope where the subjects were pushed from the front and the back while facing up the slope of the ramp as wel as down the slope. In each orientation the subjects were pushed at least thre times so that an average response could be found. The orientations are labeled by the push direction and the orientation of the subject. The two zero slope orientations are labeled as Front and Back. During the Front trials the subjects were pushed from the front and the Back trials pushed the subjects from behind. The twenty-degre slope trials are labeled with two terms. The first describes the direction the subject was facing either Up the ramp or Down the ramp. The second term is the direction they were pushed either Up or Down. For example a Up Up trial would have the subject facing uphil and the push would be from the back pushing them farther uphil. During each push the subjects were videotaped. To provide a reference point to measure displacements the subjects wore four smal white targets. These targets were placed on 50 the upper arm, the waist, and one on each ankle as can be sen on the subject in Figure 25. Figure 25 - Diver Adjusting Balast The videos were viewed in iMovie, a video editing program that alowed the start and end of each push to be found. The start of the push was defined as the point when the pusher was sen to release. The end was marked as the time at which the subject?s maximum displacement was reached. Stil images were extracted at the start and end of every push for the first round of analysis. The second round of analysis involved selecting several images betwen the start of the trial and the moment the subject?s foot left the ground. The images were analyzed using MATLAB, which alowed for several parameters of each push to be measured. These parameters include displacement of the targets and angle of the subject?s torso in each photo. The targets were located using the ginput command, which alowed the author to click on the targets and receive an x-y coordinate for each. 51 The video camera was run continuously throughout each testing sequence. This alowed data to be collected about the operation of each test in addition to the stability data. This was done to study the eficiency of the current testing procedures. And if the procedures are not eficient, look for ways to improve them. The data collected for this analysis included the number of misfires of the pusher, the length of time to reset the pusher and the length of time to swap the targets. 52 Chapter 6: Results & Data Analysis Subjects Four subjects completed the lunar trials that were discussed in the previous chapter. Their weight ranged from 114 pounds to 176 pounds. Subjects 1 and 2 also completed trials at Mars gravity. The subjects consisted of two females, the lightest two subjects, and two males, the heaviest two subjects. Table 3 - Subject Information Subject Gender Earth Weight (lb) Height (in) 1 Female 130 61 2 Female 114 64 3 Male 176 74 4 ale 171 67 Quantitative Results Two rounds of analysis were conducted utilizing the video data collected during the experiments. The first round utilized two photos. One was taken at the start of the trial and the second came at the point the subject ceased to move away from the pusher. The data generated by analyzing these photos in MATLAB was tabulated and viewed to se if any paterns could be discerned. The first thing that was investigated was how closely the data from each push was to the others completed with the same orientation and weighting. This was acomplished by finding the range in each parameter measured for one of the subjects. The results of this analysis are shown in Table 4. The third column shows the average amount of variation in the parameter across al configurations. The last column shows what percentage the average variation is of the average value. 53 Table 4 - Data Variation for Subject 2 Parameter Minimum Variation Average Variation Maximum Variation Variation, as Percent of Avg Time (sec) 0.07 0.87 1.80 26.54 % Shoulder Displacement 0.30 23.87 65.21 39.64 % Waist Displacement 3.78 21.85 86.05 43.90 % Near Leg Displacement 3.46 30.06 69.12 59.29 % Far Leg Displacement 8.55 30.00 76.34 52.04 % Starting Torso Angle (deg) 1.18 8.87 20.02 15.21 % Ending Torso Angle (deg) 1.58 12.31 27.31 22.31 % As can be sen from the last column there is a large amount of variation in the displacement and angular data. This implies that either the tests that were conducted are not repeatable and do not give consistent data, or this is not the proper way to analyze the data. This large variation would make any conclusion distiled from this analysis suspect. Consistent with the large variations no clear paterns emerge from any of the measured parameters. The large variation and lack of consistent trends led to a second round of data analysis. In this round six photos of each trial were used. This was done to beter understand the time history of each subjects? response. The six photos were chosen at even intervals from the start of the trial to the point at which one of the subjects? fet left the ground. This spacing placed the photos closer together for the quicker trials in order to capture the subject?s faster response. This analysis only utilized two out of the four targets: the waist and the ankle closest to the camera. This was done, as the shoulder target did not always 54 track the movement of the body wel. If the subject moved their arms during a trial this would cause movement of the upper arm target that was not indicative of the motion of the rest of the body. With al six photos selected the subject?s angular change was graphed in a variety of combinations to look for trends. Each orientation was graphed for each subject with al backpack configurations to analyze backpack efects. Al subjects were graphed together in each orientation to se how the mas and height of the subjects afected the response. And finaly the Mars gravity results were graphed against that subject?s Lunar trials to isolate any variation due to gravity level. Early in the analysis, while viewing the data of one subject, interesting trends appeared to emerge. The Lunar and Mars trials appeared to follow the same trend of angular change, but in the Mars trials the subject lifted their foot sooner and after les angular change. Another trend was that the angular change in each trial appeared to be linear. It would be expected that the slope would continualy increase as the subject leaned farther due to the increasing moment arm. However after analyzing al trials of al subjects none of these trends held for al subjects. In addition to graphing the time history, the lift time and maximum angular rotation were also compared as was the slope of each trial. The slope was found by taking the point of maximum angular rotation in each trial and dividing by the time to reach that point. After this was done for each trial, the trials with the same experimental conditions were averaged. No definitive conclusion was possible from these graphs either due to the wide spread in some of the data sets. It is dificult to pin down the cause of the variations because there are numerous sources of eror that are apparent upon qualitative asesment of the videos. 55 Experimental Problems & Solutions Although a quantitative analysis yielded no clear conclusions, an analysis of the experimental erors wil serve to improve the next generation of research into this field. Several of the experimental problems were created due to the unforesen reactions of the human test subjects. The next largest contributor of experimental problems came through the use of a portable disturbing force operated by a human test diver. After completing the trials and viewing the video the reactions that disrupt the data can be identified. Once identified, countermeasures can be developed to combat the disruptions. Although most are instinctive reactions the test subjects can overcome them through conscious efort. One of the most obvious problems sen in some of the trials was the subject utilizing their hands for propulsion and stability as shown in Figure 26. This is a natural human reaction to use every means at their disposal to regain balance. It is however a detriment to the acuracy of the simulation due to the high density of water as compared to a planetary atmosphere. Figure 26 - Hand Propulsion 56 This problem is not likely to be eliminated completely as balance is hardwired into the human brain. However the efects can be mitigated by having the subjects clasp their hands in front of them as sen in Figure 27. This occurred on some of the trials and semed to be quite efective in reducing arm otion. Figure 27 - Hand Clasped The next problem encountered was caused by the subjects? reaction to return to the start point for the next trial as quickly as possible. This caused the subject to lose traction in some cases while trying to run back to the start position. The main problem is that the subject?s maximum displacement appears to be reduced as they try to fight their way back to the start. This skews the results related to angle and displacement. For this reason foot lift was used as the finish for the second round of analysis. This would not be that large of a problem if the subjects slowed their movement the same way every time, but this is impossible. One of two problems is usualy responsible for the inconsistent nature of the movement. The subjects tended to either slip in betwen the grip tape causing them to fal back towards the start point or were inconsistent with the amount of 57 force applied to return to the start. This tendency also prevents the subject from achieving a stable posture at the end of the trial. This makes it dificult for the data analyst to pin down the end of the trial on the video, disrupting the time measurement. To solve this problem the subjects should be instructed to maintain the ending posture until instructed otherwise by the test diver. Future experiments should also cover the entire ramp with grip tape to provide a uniform amount of traction. The last problem created by the subjects was the inconsistent manner in which they moved after being pushed. During some trials the subject hopped backwards almost imediately after being pushed causing litle rotational motion. In other cases the subject would pivot as far as they could and then took a large step to catch themselves. And stil other times the subject would pivot a litle and then take several steps. Any or al of these reactions could be applied to the same weighting and directional conditions. This is a very tough problem to solve. It is likely that this problem is impossible to solve and can only be mitigated by discussing the problem with the subjects and reminding them to try and use the same movement technique every time. It may also be possible to atack this problem by using a disturbance that does not require the subject to move their fet, such as a weaker spring on the pusher. The test diver operating the pusher generated several anomalies as wel that need to be addresed. Since the pusher was a handheld device it had to be aligned by the operator and relied on sight and fel to atempt to repeat its placement in each trial. This caused numerous erors due to poor contact and misalignment of the pusher with the direction of 58 travel. The largest problem in this category was placing the pusher as close to the subject as possible without imparting a separate disturbing force with the operator?s arms. If the pusher was too far from the subject, some energy was lost as the push plate traveled across the gap. If the pusher was too close to the subject extra energy was imparted to the subject before the latch was released. The operator is also not placing the pusher in the same orientation every time. This causes some of the energy of the spring to be expended in directions other than the one intended. The mobile pusher is also a problem as it moves in the opposite direction of the subject reducing the force. Since the same diver was used to impart each push their mas was constant, but it is impossible to ensure that the he was standing in the exact same manner every time. This causes a variation in the amount of energy imparted to the test diver, which afects the amount imparted to the subject. A solution to these problems is to replace the pusher operator with a test stand that places the pusher in the same position every time. This solution does however add its own set of problems. Now the subject must be maneuvered to place them in the same relative position to the pusher. As inacurate as it may have been the pusher operator performs this compensation automaticaly. Another problem is that a reset mechanism must be fited to the test stand. Currently the pusher is placed in a c-clamp, reset, and then removed before the trial. The test stand would have to alow for the spring to be compresed and the mechanism moved out of the say so it does not interfere with the release of the pusher. This mechanism wil trade inacuracy for complexity. The targets themselves are also a source of eror. The shoulder target in particular had a habit of shifting down the subjects arm. This was caused by gravity and the narowing of 59 the arm from the shoulder to the elbow. The shoulder target also moves if the subject waves their arms. This causes a false impresion of the movement of the subject?s body and was the reason for excluding this target from the second analysis. The other targets were not always placed in the exact same location, but their position did not shift much from trial to trial. Their eror was created by human divers instaling them. These erors could be reduced by rigidly ataching the targets to the harnes system. The divers would stil have to ensure that the harnes was worn in the same way every time, but this appeared to be the case during the experiment already. Data Analysis Problems & Solutions Once the data was collected including al of the asociated problems mentioned in the previous section, it had to be analyzed. This proces used two diferent programs, iMovie and MATLAB, and added its own series of erors to the data. The first step in the data analysis was to pick out each individual push from the video. This was done by playing the video until the pusher was placed next to the subject?s torso. Once the pusher was in place the video was advanced frame by frame until the pusher was sen to release. This point on the video was almost always obvious and thus the selection was acurate. With the beginning of the push identified the point at which the push ended was the next priority. This was dificult to find due to the lack of a definitive end pose. A subjective cut off was often made betwen what was reaction to the push and what was the subject atempting to return to position. Solutions for this problem were discused in the previous section. The next step in the analysis required stil photos to be selected from the video clip for analysis in MATLAB. Selection of these photos depends on the parameters that are most important to what information is desired. For the first analysis 60 photos were chosen at the beginning and end of push. These two photos were chosen because they would show the subject?s maximum displacement and it was a time eficient way of selecting photos. Neither of these two photos typicaly depicts the point of maximum rotation. A MATLAB script was writen that read in the two photos and asked the user to select the four targets worn by the subject. As imprecise as this method sems, it generates results that are repeatable. Table 5 is the raw MATLAB data from ten atempts to pick out the targets in Figure 28. As can be sen the data shows that on many occasions the same numbers are generated in diferent trials. Trial 5 does show one anomaly in the Leg 2 data. This eror is believed to be caused by a delay in MATLAB. After the point was clicked the mouse was moved while MATLAB was procesing which may have skewed the value. Table 5 - Repeatability Test Data X Cordinate Trial Shoulder Midle Leg 1 Leg 2 1 383 385 375 375 2 383 385 375 375 3 383 385 375 375 4 383 385 375 375 5 383 385 37 381 6 381 385 374 375 7 383 385 375 375 8 381 385 375 375 9 381 385 375 375 10 383 385 375 375 Y Cordinate Trial Shoulder Midle Leg 1 Leg 2 1 127 215 451 451 2 129 217 451 451 3 127 217 451 451 4 129 215 451 451 5 129 217 451 127 6 127 217 451 450 7 127 215 450 451 8 127 215 451 451 61 9 129 215 450 451 10 129 215 451 451 Figure 28 - Photo Used in Repeatability Test The second round of analysis utilized a series of photos to capture the time history of the subjects? motion. The point picked for the end of he trial in this analysis was the lifting of the subjects? foot. This was chosen in an atempt to reduce the uncertainties due to the subject either not holding in an end pose or rushing back to the start point. The foot lift as an end point for the trial into a problem due to the fact that sometimes the subjects stepped to catch themselves and in some cases they hopped to catch themselves. When the subject steps the data shows a consistent increase in the angle of the subject, however if the subject hops they roll onto the bal of their foot moving their ankle in relation to their hip. This creates an artificial drop in the angular change that is graphed. A possible solution to this problem is to define the end of the trial as the moment the subjects begin a large rapid motion to catch themselves, either stepping or preparing to jump. This introduces the problem of deciding when a large motion has begun which would add erors to the analysis. 62 Chapter 7: Conclusions The current method of collecting data clearly needs to be improved before the next round of data collection is atempted. These improvements cover a wide range of topics from the equipment used in the tests, to the analysis tools, and even the instructions given to the subjects. Equipment Improvements There is always room for improvement when building the next generation of any system. The major improvements for this system include the adjustments on the harnes and the pusher latch. One area that should be looked at, as a possible avenue is a fixed pushing device. The adjustments on the harnes are currently a quick adjustment buckle. These buckles are great for quickly sizing the harnes, but do tend to crep over time. Replacing them with buckles where the webbing doubles back on itself should eliminate this problem. The latch on the pusher needs to be improved so that the strength of the operator has no bearing whether the pusher misfires or not. Lastly in an atempt to reduce the uncertainty due to the operator, a fixed mount should be considered. Procedural Improvements The main improvement to the procedure should be in the form of instructions to the subjects to reduce the erors inherent to a human experiment. These instructions should include that the subjects hands should be clasped in front of them during al trials to reduce hand waving. Clasping the subject?s hand in front may afect their neutral body posture, but this may be necesary to reduce the amount the subject uses the water drag to asist in the recovery of their undisturbed posture. Next, the subjects should be instructed 63 to remain in an end pose until the test diver clears them to move. This wil improve the ease with which a trial can be identified in the post tape analysis. Analysis Improvements The analysis of the data can be improved through a diferent selection of trial end point. If consistent criteria can be developed and applied to define the start of gross motion this wil likely be the best condition. Further experimentation wil be required to develop the criteria necesary to identify the proper time. Alternate Experimental Method Up to this point the suggestions have focused on improving the current experiment. However this may not be the best approach. As Akin?s Law #11 states: ?Sometimes, the fastest way to get to the end is to throw everything out and start over.? [24] The current method of disturbing the subjects? posture is force based. The pusher utilizes the same force for every subject at every gravity level. This means that the smaler subjects receive a larger aceleration and the lower the gravity level the larger the aceleration. This can cause problems comparing trials betwen subjects and gravity levels. A possible solution to this problem is to replace a constant force disturbance with a constant displacement disturbance. This could be implemented in the form of floor that moves horizontaly a specified distance. This type of experiment would be analogous to pulling a rug out from under someone. The constant displacement would be consistent for every permutation of the experiment. 64 Another way to achieve a constant disturbance would be to utilize a plate that drops slightly under the subject. The plate could either drop and maintain its orientation or be designed to pivot such that the orientation of the plate changes. This plate could be instrumented with a force-torque sensor. The sensor would show how the subject reacts to the disturbance and would quantify changes in the subject?s stance that may not be discernable on the video. 65 Chapter 8: Future Work The research presented represents only a fraction of the possible avenues that can be explored. Balasted partial gravity systems alow for research into many areas of planetary EVA. Numerous experiments can be completed in the shirtsleve environment as was done in this study. These include rover/habitat ingres/egres studies, suitport design, and load handling. This research can also be combined with other work being done at the SL including the MX-2 space suit analogue and surface robotics. Shirtsleve Experiments In addition, to filing in more PLS CG data points in experiments similar to the one conducted here, there are several other avenues that can be explored. Studies can be conducted utilizing simulated habitats and rovers. These studies wil help to identify the optimum size for hatches. The goal is to design a hatch that is easy to climb through while designing it as smal as possible to reduce mas. Suitports as an entry point for rear-entry suits can be studied in a manner similar to the rover and habitat studies. Lastly diferent methods for lifting loads in partial gravity environments can be investigated. To validate any results gained through further shirtsleve testing the same studies should be caried out at full Earth gravity both underwater and on land. This wil help to isolate the efects that are due to the change in gravity level and those that are due to the drag of water. Earth gravity studies wil be dificult to implement. Implementation in the underwater environment wil require placing a large amount of balast on the test subject. This wil greatly increase the mas of the test subject adding to the inertial efects on the data. It wil also be dificult to place the mas on the subject such that it is both secure 66 and comfortable for the test subject. Lastly an extremely reliable balast release wil be necesary in order to ensure the subject?s safe extraction in the event of an emergency. Earth gravity testing out of the water environment wil be dificult as wel. Padding or other safety precautions wil be required in order to make the subject fel comfortable with being place off balance. If the subject does not fel comfortable there could be a considerable diference in the subject?s reaction as compared to the underwater trials. MX-2 Space Suit Analogue The SL is testing a neutral buoyancy space suit analog known as the MX-2 shown in Figure 29. This second generation design is a low cost platform that is being used to demonstrate technologies that can be incorporated into next generation space suits. Currently the MX-2 has only been used to simulate microgravity EVA, but future upgrades could alow the MX-2 to simulate partial gravity as wel. This wil alow for data to be compared with the shirtsleve tests to understand which efects are due to the presure suit and which are inherent to the partial gravity environment. 67 Figure 29 - MX-2 Space Suit Analog The main obstacle that must be overcome when integrating partial gravity capabilities into the MX-2 wil be the atachment of the large amount of balast required. Due to the fact that the suit is full of air it has a large amount of inherent buoyancy. During current trials this buoyancy is overcome by the addition of lead weights to achieve neutral buoyancy. Approximately 250 lbs of lead is required in addition to the weight of the suit subject. To this more balast must be added to simulate a properly weighted suit in the chosen gravity field. Current generation planetary prototype suits such as the Mk II and the I-Suit have Earth weights of 120 lb. and 84 lb. respectively. This translates into 20 lb. and 14 lb. for a lunar gravity simulation. [19] This balast is in addition to the amount needed to simulate the weight of the suit subject. For a 175 lb subject on the Moon the total balast required to simulate an I-Suit would be approximately 293 lb. 68 Human-Robot Interaction The SL has a long history of human-robot interaction going back to the Beam Asembly Teleoperator (BAT). This research has continued with the latest SL dexterous robot Ranger shown in Figure 30 working with the MX-2. The SL is also conducting research in planetary robotics including astronaut asistants. With the design and construction of a rover that could be used in the underwater environment, studies could be conducted into planetary human-robot interaction. Studies could be conducted to investigate the best roles for the rovers as wel as optimal methods for their implementation. Figure 30 - Ranger & MX-2 Working Together 69 Appendix A: Pusher Experiment Data Individual test run data for al subjects at both Lunar and Martian gravity is displayed below. Graphs show the angular change at each point in the trial. Graphs are separated by backpack loading and subject orientation. 70 71 72 73 74 75 Al backpack loading conditions are shown on each graph for Subject 1. Graphs show the angular change at each point in the trial. Graphs are separated by subject orientation and gravity level. 76 77 78 Both Lunar and Martian gravity levels are shown on each graph for Subject 1. Graphs show the angular change at each point in the trial. Graphs are separated by subject orientation and backpack loading. 79 80 81 82 Al backpack loading conditions are shown on each graph for Subject 2. Graphs show the angular change at each point in the trial. Graphs are separated by subject orientation and gravity level. 83 84 85 Both Lunar and Martian gravity levels are shown on each graph for Subject 2. Graphs show the angular change at each point in the trial. Graphs are separated by subject orientation and backpack loading. 86 87 88 89 Al backpack loading conditions are shown on each graph for Subject 3. Graphs show the angular change at each point in the trial. Graphs are separated by subject orientation. 90 91 Al backpack loading conditions are shown on each graph for Subject 4. Graphs show the angular change at each point in the trial. Graphs are separated by subject orientation. 92 93 Trial Legend Trial 1 to 6 ? No Backpack Trial 7 to 12 ? High CG Backpack Trial 13 to 18 ? Low CG Backpack In each range the orientations are in the following order: Front Back Up Up Up Down Down Down Down Up The following graphs show the average lift time for each configuration. The eror bars represent the maximum and minimum lift times. Eror bars were not included on graphs when they obscured the data points. 94 95 96 The following graphs show the average angular change for each configuration. The eror bars represent the maximum and minimum angular change. Eror bars were not included on graphs when they obscured the data points. 97 98 99 The following graphs compare the average lift time to the average angular change for the two subjects, which conducted trials and Lunar and Mars gravity. 100 101 The following graphs show the angular change divided by the lift time for each trial. Once these values were calculated they were averaged for each loading condition. 102 103 104 Appendix B: MATLAB Code %John Mularski %MS Thesis Research %6/7/2007 %Computes angular change before subject lifts foot %Points must be clicked in the following order: %1) Middle of Body %2) Leg Closest to Camera clc clear all pic = {'0 front 1 start.jpeg','0 front 1 lift.jpeg','0 front 2 start.jpeg','0 front 2 lift.jpeg',... '0 front 3 start.jpeg','0 front 3 lift.jpeg','0 back 1 start.jpeg','0 back 1 lift.jpeg',... '0 back 2 start.jpeg','0 back 2 lift.jpeg','0 back 3 start.jpeg','0 back 3 lift.jpeg',... '20 up up 1 start.jpeg','20 up up 1 lift.jpeg','20 up up 2 start.jpeg','20 up up 2 lift.jpeg',... '20 up up 3 start.jpeg','20 up up 3 lift.jpeg','20 up down 1 start.jpeg','20 up down 1 lift.jpeg',... '20 up down 2 start.jpeg','20 up down 2 lift.jpeg','20 up down 3 start.jpeg','20 up down 3 lift.jpeg',... '20 up down 4 start.jpeg','20 up down 4 lift.jpeg',... '20 down down 1 start.jpeg','20 down down 1 lift.jpeg',... '20 down down 2 start.jpeg','20 down down 2 lift.jpeg',... '20 down down 3 start.jpeg','20 down down 3 lift.jpeg',... '20 down up 1 start.jpeg','20 down up 1 lift.jpeg','20 down up 2 start.jpeg','20 down up 2 lift.jpeg',... '20 down up 3 start.jpeg','20 down up 3 lift.jpeg'}; m = size(pic); q = m(2); for i = 1:q c = imread(pic{i}); figure(1) image(c) title(pic{i}) [x1,y1] = ginput(2); 105 x1 = [528.5-x1(1); 528.5-x1(2)]; % d = imread(pic{2*i}); % figure(2) % image(d) % title(pic{2*i}) % [x2,y2] = ginput(2); % % x2 = [528.5-x2(1); 528.5-x2(2)]; angle1_rad = atan((x1(2) - x1(1))/(y1(1) - y1(2))); % angle2_rad = atan((x2(2) - x2(1))/(y2(1) - y2(2))); angle1 = angle1_rad*180/pi; % angle2 = angle2_rad*180/pi; data(i,:) = [x1(1) y1(1) x1(2) y1(2) angle1]; % data(2*i,:) = [x2(1) y2(1) x2(2) y2(2) angle2]; close all clear x1 y1 angle1 x2 y2 angle2 angle1_rad angle2_rad end csvwrite('PGS Data AK Lunar.csv',data) 106 %John Mularski %MS Thesis Research %6/13/2007 %Computes angular change before subject lifts foot %Points must be clicked in the following order: %1) Middle of Body %2) Leg Closest to Camera clc clear all %pic = {'f11.jpeg' 'f12.jpeg' 'f13.jpeg' 'f14.jpeg' 'f21.jpeg' 'f22.jpeg' 'f23.jpeg' 'f24.jpeg' 'f31.jpeg' 'f32.jpeg' 'f33.jpeg' 'f34.jpeg'}; %pic = {'b11.jpeg' 'b12.jpeg' 'b13.jpeg' 'b14.jpeg' 'b21.jpeg' 'b22.jpeg' 'b23.jpeg' 'b24.jpeg' 'b31.jpeg' 'b32.jpeg' 'b33.jpeg' 'b34.jpeg'}; %pic = {'uu11.jpeg' 'uu12.jpeg' 'uu13.jpeg' 'uu14.jpeg' 'uu21.jpeg' 'uu22.jpeg' 'uu23.jpeg' 'uu24.jpeg' 'uu31.jpeg' 'uu32.jpeg' 'uu33.jpeg' 'uu34.jpeg'}; %pic = {'ud11.jpeg' 'ud12.jpeg' 'ud13.jpeg' 'ud14.jpeg' 'ud21.jpeg' 'ud22.jpeg' 'ud23.jpeg' 'ud24.jpeg' 'ud31.jpeg' 'ud32.jpeg' 'ud33.jpeg' 'ud34.jpeg' 'ud41.jpeg' 'ud42.jpeg' 'ud43.jpeg' 'ud44.jpeg'}; %pic = {'dd11.jpeg' 'dd12.jpeg' 'dd13.jpeg' 'dd14.jpeg' 'dd21.jpeg' 'dd22.jpeg' 'dd23.jpeg' 'dd24.jpeg' 'dd31.jpeg' 'dd32.jpeg' 'dd33.jpeg' 'dd34.jpeg'}; pic = {'du11.jpeg' 'du12.jpeg' 'du13.jpeg' 'du14.jpeg' 'du21.jpeg' 'du22.jpeg' 'du23.jpeg' 'du24.jpeg' 'du31.jpeg' 'du32.jpeg' 'du33.jpeg' 'du34.jpeg'}; m= size(pic); q = m(2); for i = 1:q c = imread(pic{i}); figure(1) image(c) title(pic{i}) [x1,y1] = ginput(2); x1 = [528.5-x1(1); 528.5-x1(2)]; angle1_rad = atan((x1(2) - x1(1))/(y1(1) - y1(2))); angle1 = angle1_rad*180/pi; 107 data(i,:) = [x1(1) y1(1) x1(2) y1(2) angle1]; close all clear x1 y1 angle1 angle1_rad end csvwrite('PGS Data AK middle.csv',data) 108 %John Mularski %MS Thesis Research %5/19/2007 %Computes translation and rotation distances for Partial Gravity Photos %Points must be clicked in the following order: %1) Shoulder %2) Middle of Body %3) Leg Closest to Camera %4) Leg Farthest from Camera clc c = imread('20 down up 3 start.jpeg'); figure(1) image(c) [x1,y1] = ginput(4); d = imread('20 down up 3 end.jpeg'); figure(2) image(d) [x2,y2] = ginput(4); shoulder_dist = sqrt((x1(1) - x2(1))^2 + (y1(1) - y2(1))^2) mid_dist = sqrt((x1(2) - x2(2))^2 + (y1(2) - y2(2))^2) front_leg_dist = sqrt((x1(3) - x2(3))^2 + (y1(3) - y2(3))^2) back_leg_dist = sqrt((x1(4) - x2(4))^2 + (y1(4) - y2(4))^2) angle1_rad = atan((abs(y1(1) - y1(2)))/(abs(x1(1) - x1(2)))); angle2_rad = atan((abs(y2(1) - y2(2)))/(abs(x2(1) - x2(2)))); angle1 = angle1_rad*180/pi angle2 = angle2_rad*180/pi angle_change = abs(angle1 - angle2); close all 109 Appendix C: IRB Paperwork IRB Aplication 0. Title: Water Imersion Balasted Partial Gravity Simulation for Lunar and Martian EVA Simulation 1. Abstract: This experiment wil simulate the gravity conditions found on Mars and the Moon, and ases the ability of humans to operate and perform routine tasks in these partial gravity environments. Subjects wil wear a weight harnes that wil load them to the appropriate percentage of their weight on Earth. The experiment wil be conducted in the Neutral Buoyancy Research Facility (NBRF) based at the Space Systems Lab (SL). The NBRF is a 25-foot deep 367,000-galon water tank used for space simulation research. This research wil sek to develop an information base from which hardware can be designed for extravehicular activities (EVAs) in reduced gravity conditions. This wil include hatchway design, load-carying devices, and mobility aids. The tests wil be conducted through the use of SCUBA. The subjects wil utilize a hookah diving rig to alow unencumbered movement. Should the primary air supply cease functioning, subjects wil also be equipped with a 3 ft 3 Spare-Air bottle. This wil provide approximately 2 ? 3 minutes of breathing time while safety divers execute emergency procedures. Depending on the severity of the situation this wil include the use of an alternate air source, jetisoning balast, and subject extraction from the tank. Al diving operations wil be performed under the auspices of the University of Maryland Diving Control Board, and in acordance with the UMd Diving Safety anual. Subject involvement wil be completely voluntary, informed consent wil be obtained before the start of the experiment, and al subject data wil be kept confidential. 2. Subject Selection: The subjects wil be SL certified divers. As a condition of diving at the University of Maryland, al of the subjects have current physical exams and medical histories that demonstrate that they met the established criteria to dive. As part of the diver certification proces, they have already been tested for scuba diving skils and trained in required subjects, including CPR, first aid, and emergency oxygen administration. Subjects wil be recruited by an email sent to al active SL divers. Participation in the experiment wil be completely voluntary and any subject can withdraw at anytime for any reason. 3. Procedures: SL Facilities & Procedures Al experimental sesions wil be conducted at the NBRF. Sesions wil be conducted with safety as the primary focus. To this end SL personnel wil monitor al sesions both underwater and on the deck. There are two porthole cameras located in the tank, one in the north and one in the south. These two cameras are adjustable in both position and zoom. If additional views are needed, 110 underwater cameras can be mounted and adjusted by the divers. Information can be relayed to the divers from surface personnel through the use of an underwater speaker located in the tank. In case of emergency the divers can be verbaly recaled, or a siren can be sounded to bring them to the surface. The NBRF is equipped to deal with most emergencies that can be expected during a dive. Supplies include a first aid kit, emergency oxygen supply, backboard, and AED. During al diving operations at the SL at least thre personnel are required: a deck chief, a lead diver, and a second diver. The deck chief has overal control of al aspects of a dive and has control in the event of an emergency. Al deck chiefs are certified NBRF divers and are trained in first aid, CPR, oxygen administration, and acident management. The lead diver monitors al operations underwater. It is the lead diver?s responsibility to monitor air supplies of the other divers and to ensure that al operations are conducted safely. For these sesions the lead diver wil also be equipped with an Aga Divator full-face mask (FM) to alow direct two-way communications with the surface. The lead diver wil be able to continuously report on the status of the test subject. The second diver wil act as a safety diver during these sesions. This diver wil be ready at al times to provide the subject with an alternate air source and begin emergency extraction. FM Equipped Diver 111 Test Equipment The test subjects wil utilize a hookah diving rig to alow them unencumbered movement during the test sesions. The hookah rig is a standard SCUBA diving regulator atached to the cylinder by a long hose. The hose from the regulator wil be atached to the subject?s shoulder harnes. This prevents hose tension from pulling the regulator out of the subject?s mouth. During the tests the air cylinder wil rest on the bottom of the tank. If continuous fedback from the subject is required during a test the hookah can also be used with the FM. If the fedback is not required for a particular sesion, a standard regulator wil be used to eliminate the buoyancy caused by the FM. The test subjects wil descend using a normal SCUBA setup and switch to the hookah once at the bottom. In addition to the hookah, the subjects wil be equipped with a Spare Air cylinder. The Spare Air is a 3 ft 3 cylinder equipped with an integrated regulator. This wil provide the diver with approximately two minutes of air in an emergency. To alow the subjects to be balasted to simulate partial gravity, they wil wear a full body harnes. Based on standard models of body segment parameters, balast wil be atached to the harnes at the torso and upper legs to replicate the weight of the various body segments for partial gravity situations. For example, for an average 170 lb test subject, the total balast would have to equal 28 lbs for lunar- equivalent body loading. To maintain appropriate mas distribution, this would be atached as 5 lbs of balast to each thigh, and 18 lbs on the torso. Similarly for Mars, 10 lbs would be atached to each thigh and 45 lbs to the torso, for a total balast mas of 65 lbs. The balast weights wil be atached to quick release plates. One end of the plate has been formed into a tab that fits into a slot on the harnes. The other end of the plate has a hole into which a clevis pin is inserted. Through the hole at the end of the clevis pin a retaining pin is inserted. Atached to the retaining pin is a cable, which when pulled wil alow the balast to drop fre. This ensures that the balast can be quickly ditched in the case of an emergency. The harnes also serves as a means for quickly extracting the subject from the water in an emergency. 112 Ballast harnes made in-house based on this design with ballast added as indicated 113 Sample ballast plate with 10 lb weight Sample plate installed on test rig 114 Plate shown after quick release activated (Shown without weight for purposes of ilustration) Emergency Procedures Primary Air Supply Interuption If the subject?s hookah air supply fails for any reason, a ?bailout? to an alternate air source wil be performed. The subject has at least thre alternate air supplies to choose from. The first is the Spare Air cylinder atached to their harnes, which supplies them with several minutes of breathing air. This is an ?interim? supply, capable of providing air during an emergency ascent, or as a bridge to an alternate air source. The second source is an ?octopus? (safety) regulator offered by a safety diver. As part of the test protocols, safety divers are required to stay within close range of the test subject, and to be capable of handing them the spare regulator within ten seconds following an emergency. Al scuba divers are practiced in the skil of switching from one regulator from their training courses. Actualy, any diver underwater can share their air supply through secondary regulators or ?buddy-breathing?, but the safety diver is positioned to be the first responder if a problem occurs. The third air source is the standard tank/buoyancy compensator/regulator set-up, which the test subject wears while descending to the test site and which is laid on the tank floor beside the test apparatus. Once alternative air sources have been supplied and the subject is stabilized, the balast wil be removed and the subject wil nominaly transfer back to the standard SCUBA rig. This wil then be used to ascend to the surface. Emergency Requiring Imediate Extraction If an emergency occurs that requires imediate extraction, the safety divers wil ditch the subject?s balast using the quick release mechanisms. After the balast has been removed one safety diver wil imediately asist the subject to the surface. The hookah hose is long enough to reach the surface without adjustment 115 under normal circumstances. If for any reason the hose does not reach, the second safety diver wil cary the air cylinder supplying the hookah from the bottom of the tank to a point where the hose wil reach. Once at the surface any necesary medical action wil be taken and the deck chief wil be informed of the situation. Balast Release Failure A balast release failure by itself is not an emergency. As long as the subject is breathing there is plenty of time to troubleshoot the problem. If during the troubleshooting the subject or safety divers run low on air, additional cylinders can be sent from the surface. Due to having a compresor on site, a virtualy unlimited supply of air is available. Once the balast release problem has been resolved, a normal ascent can be conducted. If for any reason the balast release problem is not resolved, a number of resolution techniques are available. The subject can remove the entire harnes system with the weights atached. Alternately, even at the largest Mars balast loads two divers would be able to swim the subject to the surface, where they could stand on the donning platform with their head out of the water while the weights are removed. A further alternative is to climb the aces ladder to the surface, which runs the entire distance from the botom of the tank to the top. Finaly, if al other alternatives have failed, the subject can imediately extracted from the water using the overhead crane. Test Overview Subjects wil participate in multiple sesions the first of which wil be a familiarization sesion. First, the consent form wil be presented to the subjects to be signed before the test goes further. Next, during this sesion the subjects wil be weighed to determine the amount of balast needed. Then the weight harnes wil be fited to the subject and al emergency procedures wil be explained. The subject and safety divers wil then descend on SCUBA to the bottom of the tank. Once at the bottom the subject wil remove their fins, switch to the hookah, and balast wil be atached. The subject wil be alowed to become comfortable with the balast before any experiments are conducted. This wil be followed by a rehearsal of the emergency procedures. Subsequent sesions wil be used for the data collection. Each of these subsequent sesions wil focus on a particular activity. These sesions wil include backpack stability, ease of ingres/egres, and package transport. Data for each sesion wil be collected by videotape and subject debrief. Backpack Stability During these sesions weights wil be placed in diferent location on the subjects back to simulate a space suit backpack. The location as wel as the amount of weight wil be varied during the sesion. The subject wil be asked to perform simple tasks with each backpack configuration. These wil include climbing slopes and ladders, bending to pick up rocks, recovery from a prone position, and walking. 116 Ingres/Egres These sesions are being conducted to investigate the optimum size and shape of a hatchway on a presurized rover or habitat. The test apparatus wil consist of an adjustable frame that wil simulate the hatchway. The subject wil move through hatch from one side to the other. The habitat/rover wil not be simulated, so the subject wil not be entering an overhead environment. Package Transport The subjects wil transport packages representing planetary surface experiments betwen two points. This wil be done using several diferent methods to include baskets, atached handles, carts, and simply lifting the object from the botom. 4. Risks and Benefits: The main risks of the experiment are those inherent to SCUBA diving. These include drowning, arterial gas embolism, decompresion sicknes, and barotrauma to the ears. Since the subjects are al divers they are trained to deal with the risks. The balast system wil add some risks unique from diving, namely the subject?s inability to ascend while wearing the weights. To mitigate this risk, the weights wil be mounted on quick release mechanisms and safety divers wil be present at al times to asist the subjects. 5. Confidentiality: Data collected during debriefings wil not be labeled with the subjects? name. The debriefings wil be labeled with a code that alows the principal and student investigator to identify the subject. This is necesary to alow the debrief comments to be linked to the videos. Al debriefings wil be stored in electronic form on pasword-protected computers. Videos of each sesion wil be stored on tapes at the SL. The tapes wil become part of the SL video archive. The tapes wil be labeled with the same code as the debriefings. The subject wil only be identifiable if recognized from the video. Subjects? identities wil not be revealed in any publication of this research. 6. Information and Consent Forms: Al subjects wil be supplied with printed consent forms to be signed before the commencement of experimental sesions. The consent form wil include information on the purpose, procedure, and risks involved with the experiment. As divers at the SL al of the subjects have already signed waivers informing them of the risks inherent to SCUBA diving. When the consent form is provided to the subjects they wil be given the opportunity to ask questions. The consent form wil only be presented in English, as al subjects wil be fluent English speakers. 7. Conflict of Interest: No private sector company is involved in this research. No financial or employment conflict of interest is presented by this research. 8. HIPA Compliance: No HIPA protected health information wil be used in this experiment. 9. Research Outside of the United States: Not Applicable 117 10. Research Involving Prisoners: Not Applicable CONSENT FORM Project Title Water Imersion Balasted Partial Gravity Simulation for Lunar and Martian EVA Simulation Why is this research being done? This is a research project being conducted by Dr. David Akin and John Mularski of the Space Systems Lab (SSL) at the University of Maryland, Colege Park. We are inviting you to participate in this research project because you are a diver certified at the SL. The purpose of this research project is to evaluate hardware designed for Lunar and Martian EVA. This information wil be used to design hardware to improve the productivity of astronauts on Lunar and Martian EVA. What wil I be asked to do? The procedures involved in the first session wil begin with being weighed and fited into the harness. After al safety procedures have been explained you wil then descend on SCUBA to the botom of the tank to begin the test. The first session wil be used simply to become comfortable with the harness as well as the emergency procedures. After practicing the emergency procedures the first session wil be completed. Subsequent sessions wil be divided between backpack stability, ingress/egress, and package transport. Backpack Stability During these sessions weights wil be placed in diferent location on your back to simulate a space suit backpack. The location as well as the amount of weight wil be varied during the session. You wil be asked to perform simple tasks with each backpack configuration. These wil include climbing slopes and laders, bending to pick-up rocks, recovery from a prone position, and walking. Ingress/Egress These sessions are being conducted to investigate the optimum size and shape of a hatchway on a pressurized rover or habitat. The test aparatus wil consist of an adjustable frame that wil simulate the hatchway. You wil move through hatch from one side to the other. The habitat/rover wil not be simulated, so there wil always be direct access to the surface in case of emergency. Package Transport You wil transport packages representing planetary surface experiments between two points. This wil be done using several diferent methods to include baskets, atached handles, carts, and simply lifting the object from the botom. The subject may end any session at any time for any reason by displaying the ascend signal (thumbs up). This wil terminate the test; the safety divers wil then remove the balast and asist the subject to the surface. 118 Project Title Water Imersion Balasted Partial Gravity Simulation for Lunar and Martian EVA Simulation What wil I be asked to do? Al of the sessions wil be video taped for later analysis and subjective coments wil be gathered during a debriefing after each session. Al videotapes wil be kept in the archives of the Space Systems Lab and may be used for documentation of test activities. Each session wil last aproximately 1 hour in-water. The study wil have a total of 4 in-water sessions with briefings and debriefings. The total time comitment should be aproximately 8 hours. What about confidentiality? We wil do our best to keep your personal information confidential. To help protect your confidentiality, (1) your name wil not be included on the surveys and other colected data; (2) a code wil be placed on the survey and other colected data; (3) through the use of an identification key, the researcher wil be able to link your survey to your identity; and (4) only the researcher wil have access to the identification key. If we write a report or article about this research project, your identity wil be protected to the maximum extent posible. Your information may be shared with representatives of the University of Maryland, Colege Park or governmental authorities if you or someone else is in danger or if we are required to do so by law. What are the risks of this research? There may be some risks from participating in this research study. The main risks of the experiment are those inherent to SCUBA diving. These include drowning, arterial gas embolism, decompression sickness, and barotrauma to the ears. The balast system wil ad some risks unique from diving, namely the inability to ascend while wearing the weights. To mitigate this risk, the weights wil be mounted on quick release mechanisms and safety divers wil be present at al times to After any major ilness, injury, or medical condition requiring hospitalization for 24 hours or longer subjects must be cleared before returning to diving. If the ilness, injury or condition is pressure-related, then the clearance must come from a physician trained in diving medicine. What are the benefits of this research? This research is not designed to help you personaly, but the results may help the investigator learn more about ways to design hardware for planetary exploration. We hope that, in the future, other people might benefit from this study through improved understanding of design limitations placed on this equipment by astronauts? movement in reduced gravity. 119 Project Title Water Imersion Balasted Partial Gravity Simulation for Lunar and Martian EVA Simulation Do I have to be in this research? May I stop participating at any time? Your participation in this research is completely voluntary. You may chose not to take part at al. If you decide to participate in this research, you may stop participating at any time for any reason. If you decide not to participate in this study or if you stop participating at any time, you wil not be penalized or lose any benefits to which you otherwise qualify. The subject?s desire to end one session, does not disqualify them in anyway from participating in subsequent sessions. Is any medical treatment available if I am injured? The University of Maryland does not provide any medical, hospitalization or other insurance for participants in this research study, nor wil the University of Maryland provide any medical treatment or compensation for any injury sustained as a result of participation in this research study, except as required by law. What if I have questions? This research is being conducted by Dr. David Akin and John Mularski of the Aerospace Engineering Department at the University of Maryland, Colege Park. If you have any questions about the research study itself, please contact: Dr. David Akin University of Maryland Building 382 Rom 210D Colege Park, MD 20742 (email) dakin@sl.umd.edu (telephone) 301-405-138 or John Mularski University of Maryland Building 382 Rom 10C Colege Park, MD 20742 (email) mularski@sl.umd.edu (telephone) 301-405-7353 If you have questions about your rights as a research subject or wish to report a research-related injury, please contact: Institutional Review Board Office University of Maryland, Colege Park, Maryland, 20742 (e-mail) irb@deans.umd.edu (telephone) 301-405-0678 This research has been reviewed according to the University of Maryland, Colege Park IRB procedures for research involving human subjects. 120 Project Title Water Imersion Balasted Partial Gravity Simulation for Lunar and Martian EVA Simulation Statement of Age of Subject and Consent [Please note: Parental consent always needed for minors.] Your signature indicates that: you are at least 18 years of age;, the research has been explained to you; your questions have been fuly answered; you agree to be videotaped during the sessions; and you freely and voluntarily chose to participate in this research project. NAME OF SUBJECT SIGNATURE OF SUBJECT Signature and Date DATE 121 Bibliography [1] University of Minnesota. ?AEM Students to Participate in Microgravity Experiments.? Aerospace Engineering & Mechanics Department. http:/ww.aem.umn.edu/info/update/1999-00/micrograv.html [2] Wiliams, R. K. and Bilica, L. W. ?Johnson Space Center Reduced Gravity Program and 1992 Highlights.? 31 st Aerospace Sciences Meting & Exhibit, 1993. [3] NASA Technical Reports Server. ?Astronaut Alan Shepard with Modular Equipment Transporter aboard KC-135.? ID S70-53479. [4] NASA. ?Apollo 12 Technical Crew Debriefing.? Houston, Texas, 1969. [5] Newman, D. J. ?Human Locomotion and Energetics in Simulated Partial Gravity.? Ph.D. dis., Masachusets Institute of Technology, 1992. [6] NASA Technical Reports Server. ?Astronaut Alan Bean participates in lunar surface simulation.? ID S69-56059. [7] Grifin, T. M., Tolani, N. A., and Kram, R. ?Walking in Simulated Reduced Gravity: Mechanical Energy Fluctuations and Exchange.? Journal of Applied Physiology 86, no. 1 (1999): 383 ? 390. [8] Car, C. E. ?The Bioenergetics of Walking and Running in Space Suits.? Sc.D. dis., Masachusets Institute of Technology, 2005. [9] Kennedy, K. J. ?JSC-CTSD Advanced Habitation Design Capabilities.? Space 2006. San Jose, California. Sept. 2006. [10] NASA Technical Reports Server. ?Reduced Gravity Walking Simulator.? ID GPN- 2000-001782. [11] Spady, Jr., A. A. and Krasnow, W. D. ?Exploratory Study of Man?s Self- Locomotion Capabilities with a Space Suit in Lunar Gravity.? NASA Technical Note, 1966. [12] Wickman, L. A. and Luna, B. ?Locomotion while Load-Carying in Reduced Gravities.? Aerospace Medical Asociation Annual Scientific Meting. Anaheim, CA. 1995. [13] Todd, B. and Reagan, M. ?The NEMO Undersea Analog: Another Type of Dep Space Exploration.? Space 2006. Long Beach, CA. 2006. 122 [14] NOA. ?Aquarius Undersea Laboratory.? http:/ww.uncw.edu/aquarius/archive/1993_1996/aq_exterior2_high.jpg. [15] NASA Technical Reports Server. ?Apollo 14 Crew Training.? ID GPN-2000- 000962. [16] NASA Technical Reports Server. ?Astronauts Young and Duke study rock formations on simulated lunar traverse.? ID S71-59355. [17] The Mars Society. ?The Mars Society ? Flashline Arctic Research Station.? http:/ww.marsociety.org/arctic/index.asp. [18] The Mars Society. ?The Mars Society: Mars Desert Research Station.? http:/ww.marsociety.org/DRS/index.asp. [19] Ross, A. J., Kosmo, J. J., Janoiko, B. A., Bernard, C., Splawn, K., and Eppler, D. B. ?Desert Research and Technology Studies 2005 Report.? ICES 2006. Norfolk, VA. 2006. [20] Knapik, J., Harman, E., and Reynolds, K. ?Load Cariage using Packs: A Review of Physiological, Biomechanical and Medical Aspects.? Applied Ergonomics. Vol 27. No. 3. pgs. 207 ? 216. 1996. [21] Space Systems Laboratory. ?Facilities at the SL.? htp:/sl/facilities/index.shtml. [22] de Leva, P., ?Adjustments to Zatsiorsky-Seluyanov?s Segment Inertia Parameters.? Journal of Biomechanics 29, no. 9 (1996): 1223 ? 1230. [23] Larson, W. J., and Pranke, L. K. Human Spaceflight: Mision Analysis and Design. 1999. pg. 709. McGraw-Hil. [24] Akin, D., ?Akin?s Laws of Spacecraft Design.? http:/spacecraft.sl.umd.edu/akins_laws.html