- Email: [email protected]
Pressure sensing for in-suit measurement of space suited biomechanics
Acta Astronautica 115 (2015) 218–225
Contents lists available at ScienceDirect
Acta Astronautica journal homepage: www.elsevier.com/locate/actaastro
Pressure sensing for in-suit measurement of space suited biomechanics Allison P. Anderson a,n, Dava J. Newman b a b
Massachusetts Institute of Technology, 77 Massachusetts Ave., 37-217, Cambridge, MA 02139, United States Massachusetts Institute of Technology, 77 Massachusetts Ave., 33-319, Cambridge, MA 02139, United States
a r t i c l e in f o
abstract
Article history: Received 20 January 2015 Received in revised form 15 April 2015 Accepted 18 May 2015 Available online 27 May 2015
Extravehicular Activity (EVA) is a critical component of human spaceflight, but working in gaspressurized space suits causes fatigue, excessive energy expenditure, and injury. Relatively little is known about how the astronaut moves and interacts with the space suit, and what factors lead to injury. A wearable pressure sensing system to quantitatively measure areas on the body's surface that the space suit impacts during dynamic EVA movement is developed. The system is used to characterize human–suit interaction in the NASA Mark III space suit. Three experienced subjects perform a series of upper body movements: 3 isolated joint movements and 2 functional tasks. Movements are repeated 12 times each and in-suit pressure responses are evaluated both by quantifying peak pressure and full profile responses. Results: Sequential sensor activation allows subjects to be indexed inside the space suit during complicated motions to better understand suited biomechanics. Subjectively, subjects generally feel they are consistent for all movements. However, using a nonparametric H-test, 54% of movements are found to be biomechanically inconsistent (po0.05). This experiment provides the first “window” inside the space suit to evaluate contact pressures and sequential indexing of the person inside the suit for realistic EVA movement. It cannot be extrapolated how changes in contact pressure would affect a subject's propensity for injury as injuries accumulate over long time scales. However, changes in pressure may be due to alterations in biomechanical strategies or fatigue, both of which could be precursors for injury and discomfort. & 2015 IAA. Published by Elsevier Ltd. All rights reserved.
Keywords: Space suit Pressure sensing Astronaut injury Human–suit interaction Arm biomechanics
1. Introduction Gas pressurized space suits cause injuries and significantly increase metabolic expenditure [1–14]. The current U. S. space suit, the extravehicular mobility unit (EMU), causes a variety of musculoskeletal injuries, including skin and surface injuries, where astronauts impact and rub against
Abbreviations: EVA, Extravehicular Activity; EMU, Extravehicular Mobility Unit; MG, Movement group n Correspondence to: 77 Massachusetts Ave., 37-217, Cambridge, MA 02139, United States. Tel.: þ1 417 388 0621. E-mail addresses: [email protected] (A.P. Anderson), [email protected] (D.J. Newman). http://dx.doi.org/10.1016/j.actaastro.2015.05.024 0094-5765/& 2015 IAA. Published by Elsevier Ltd. All rights reserved.
the suit as they bend the garment [7,15]. Although no EVArelated injury has prevented successful completion of mission objectives, there have been several instances when the EVA was nearly terminated due to suit discomfort [6,16]. Crew health, comfort, and safety are negatively affected by the human–suit impact. As of 2012, a total of 25 shoulder surgeries were performed on astronauts, most attributed to training in the spacesuit underwater [15]. Space suited motion is different from normal biomechanical movement. The suit is pressurized with gas to 29.6 kPa (4.3 psi) making it stiff in the vacuum of space. Additional rigidity comes both from changes in suit volume and fabric stiffness as the joints bend [17–21]. The suit is primarily composed of fabric pieces called “soft goods”, mounted to a
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
hard central torso. The soft goods are connected with metal rotational bearings and sizing rings. Programming, or planes through which the suit can move due to the angle of bearings [22], further limits mobility and makes movement unnatural. To perform EVA, astronauts must learn to change their biomechanical movement strategies, rather than attempting to move as they do naturally, or unsuited [23]. The difference between how a person moves relative to the suit has not been quantified. Previous studies used a variety of techniques, such as photogrammetry, motion capture, and ergonomic strength measurement to evaluate suited performance [4,17–19,24–28]. Results, however, are difficult to compare because methodologies and measurements inherently vary. Also, these methods cannot evaluate suited human biomechanics because they measure performance from the outside of the suit, characterizing the human and space suit as a whole. Recent work has focused on determining body joint angles within the suit [29–31], but knowing the nature and location of body-suit contact is needed to evaluate suited biomechanics. A pressure–sensing system specifically designed for dynamic movement inside the environment of the space suit is developed, and an experimental evaluation of human–suit interaction on the upper arm is presented. Human–space suit interaction is evaluated by assessing body placement within the suit and consistency of movement. We hypothesize that subjects with experience working in the space suit will perform motion tasks with consistent movement strategies. Subject experience is determined by previous time spent performing tests inside the suit. Movement strategies are defined by peak pressures averaged over trials or full time averaged pressure profiles. Consistency of movement is an important metric revealing fatigue or changes in biomechanical strategies, both of which could be precursors to EVA injury. Additionally, subjective feedback is evaluated as a proxy for measuring human–suit interaction using the techniques proposed here.
219
2. Methods The Polipo (Octopus), seen in Fig. 1, is the system of 12 sensors developed as part of this research effort for the pressure regime expected to be measured on the body under the space suit soft goods. The sensors are 2.5 cm in diameter and cover the pressure range from 0–100 kPa with approximately 1 kPa resolution [32–34]. These sensors are placed on the left arm in a way that targets anticipated hot spots of body–suit contact points, and secondarily for uniform coverage. Anticipated hot spots were determined through discussion with subject matter experts. The Polipo is integrated into a conformal wearable garment with placement targets to which the sensors attach. This experiment is performed on three subjects at NASA's Johnson Space Center in the Advanced Space Suit Laboratory. The test is performed in the Mark III space suit, shown in Fig. 1. Suit pressure is maintained at 29.6 kPa (4.3 psi). All subjects are space suit engineers, and have extensive experience performing similar motions inside the Mark III during experiments. Subjects were briefed on any potential hazards associated with the experiment. Informed consent was obtained. The protocol is approved both by the MIT Committee on the use of Humans as Experimental Subjects and the NASA Johnson Space Center's Institutional Review Board. Subject 1 is left handed, while Subjects 2 and 3 are right handed. This does not effect in the way subjects perform the experiment, since the subjects performed the motions with both arms simultaneously. However, it is not possible to determine if handedness affects the pressure measurements since many variables may contribute to differences across subjects. Subjects are asked to perform a series of upper body motions. The protocol consists of 12 repetitions of 5 motions, described in detail in Fig. 2. For each motion, the 12 repetitions are further subdivided into 3 movement groups (MG) of 4 repetitions each. This is done to evaluate subject fatigue
Fig. 1. Experimental hardware. (A) The Polipo pressure sensing system is a network of 12 sensors designed for a 0–100 kPa pressure range during dynamic upper body movement. (B) The Mark III space suit was evaluated in a human subject experiment to evaluate contact pressures on the person's body while moving.
220
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
or potential change of biomechanical strategies over the course of the test period. After each MG, the subject rests for 5 min and qualitative information is gathered. Perceived rate of exertion is evaluated by the test operator. Effort and perceived consistency of movement are evaluated by the subject on a 5-point Likert scale. Contact location and the nature of the contact (i.e. sharp, broad, etc.) are recorded using a body schematic and discomfort is rated on a 10 point Likert scale as described by the subject. A representative schematic of the test protocol is shown in Fig. 3. Static data taken prior to donning the suit and after pressurization is used to establish baseline loading of the suit on the body in the subject's relaxed standing posture while not resting in the suit's donning stand. The wiring of several wearable sensors began to deteriorate over the course of the experiment, causing erroneous jumps in the measured pressure. Therefore, this data is excluded from analysis. A total of 495 pressure profiles are evaluated. The peak pressure for each repetition within the MG is recorded. When the peak pressure is collected for all 12 repetitions, a Kruskal–Wallis H-test, equivalent to a nonparametric single factor ANOVA, is performed to evaluate the subject's consistency between MGs. The test is significant when the peak pressures within (at least) one group are significantly different than the others. Note that when data are highly variable within the MG itself, the H-test would not yield a
significant result. Therefore, it is not a measure of consistency, but rather inconsistency. 3. Results Due to the limited number of subjects, general conclusions about space suited biomechanics cannot be drawn. Rather, data from each subject is analyzed individually to present the first quantitative pressure measurements recorded inside a space suit. The pressure data is used to index a subject inside the space suit to determine when contact occurs and the nature of that contact. Quantitative evaluation of motion is useful in assessing the variability associated with suited biomechanics and subjective reporting of space suit contact, which might lead to injury. 3.1. Temporal data analysis Fig. 4 shows our first ‘look inside the suit’ plotting the pressure profiles of 6 sensors during the cross body reach motion for Subject 1. The data are normalized by time and plotted by mean and standard deviation of all repetitions within MG 1 to show the activation and relative pressure amplitudes over the course of the movement. Standard deviations show higher variability for loading certain sensors than others. Subject 1 initially begins in a resting
Fig. 2. Movement tasks performed by each subject. Three isolated joint tasks are performed: Elbow Flexion/Extension, Shoulder Flexion/Extension, and Shoulder Abduction/Adduction. Two functional tasks were performed: Cross Body Reach and Overhead Hammering. Subjects were given very specific instructions on how to perform the isolated joint tasks, while subjects were given way-point markers to meet and allowed to develop their own biomechanical strategies for the functional tasks. Subject suited in Mark III suit is shown.
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
Fig. 3. Experimental test protocol for a single subject. Subjects are given time to train each of the 5 movements inside the space suit. Subjective information is taken on comfort and pressure hot spots. The 5 movements are performed in 3 groups with subjective information taken after each group. The order is counterbalanced within the group and randomized between subjects. Each of the movements is repeated 4 times each within a movement group. Sensor pressure profiles over time are recorded for analysis.
221
posture. Only Sensor 10 is loaded due to the back of the arm resting on the space suit joint. The first 10–30% of the movement is associated with moving the arm from the neutral position across the body to the opposite hip. Sensor 1 is highly loaded during the portion of the movement since it is on the inside of the wrist and the subject is moving his arm across the body. Next he raises his arm to chest level on the opposite side. Sensor 11, located on the upper bicep, is the only sensor activated during this period due to the weight of the space suit resting on his arm. Between 40– 60% of the movement the subject sweeps his arm in front of his body at chest level in front of his shoulder. Sensor 2 is slightly loaded over this period, while Sensor 5 becomes loaded half way through this section. As Subject 1 crossed the body, he performs a shoulder rotation to prepare for the next phase of movement, the elbow bend. This rotation causes Sensor 5 to be loaded on the inside of the forearm near the elbow. Between 60–80% of the movement an elbow flexion/extension is performed to touch the helmet near his ear. As expected, this loads Sensors 1, 2, 5, and 6, all the lower arm and elbow sensors. Finally, he extends his elbow and lowers his arm to the neutral position in the last 90–100% of the movement. This re-loads Sensor 10 since it is loaded in the neutral posture as the subject's arm rests on
Fig. 4. Temporal activation of sensors for cross body reach. Results are for Subject 1, movement group 1. Response is averaged over each repetition and normalized by time. Sensors are shown with most distal on top of figure and proximal at the bottom. The Y axis is consistent for all sensors to compare relative pressure magnitudes.
222
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
the suit. Sensor 11 is also slightly loaded as the subject brings his arm down causing contact with the bicep due to the weight of the suit arm. This sequence of arm–suit interaction and loading is consistent for Subject 1 over all three MGs. The capability to discern movement patterns as a subject performs a task shows in great detail how the person's body shifts within the suit and moves the space suit as he or she attempts to complete a movement. The pressure sensors and wearable garment technology developed in this study provides a novel capability to measure body–suit interactions.
3.2. Movement consistency Example pressure profiles are shown in Figs. 5 and 6 to provide a sense of inter- and intra-subject consistency, but do not reflect all motions and profile responses. Profiles are plotted by MG and are normalized by time of the movement. The peak pressure of each of the 12 repetitions is used to calculate the mean (m) and standard deviation of the peak pressures. Fig. 5 shows pressure response profiles for all 3 subjects during elbow flexion/extension over the upper forearm, distal of the elbow crease. Subject 1 produces consistent pressures during MG 1, but varies widely in MG 2 producing the highest pressure seen for all subjects during this movement over this area of the body. The sensor wiring broke inside the suit prior to MG 3, so statistical analysis was not performed across all MGs for this motion. However, the subject's inconsistency is apparent. Within each of the three MGs, Subject 2 exhibits a steady decrease in peak pressure over the four repetitions. Overall, the peak pressures were lower in MG 1, exhibiting marginal inconsistency (H¼5.55, p¼0.06). Subject 3 produces similar pressure profiles with each repetition, although the peaks in MG 2 are slightly elevated compared to the other two MGs (H¼5.38, p¼ 0.06).
In MG 3, subject 3 also exhibits contact pressure between repetitions, likely due to a change in the positioning of his arm, altering the overall shape of the response profile. Fig. 6 gives additional examples to highlight subject movement consistency. Data for Subject 1 is collected over the wrist during shoulder flexion/extension. The subject's movements are inconsistent (H¼8.8, p¼0.01). During MG 1, the subject exhibits a steady increase in peak pressures, while in MG 2 there is a steady decrease. During MG 3, the peak pressures are consistent, but are 25% of the magnitudes seen in previous repetitions, indicating a change in movement. By contrast, data from Subject 1 over the upper forearm shows consistent peak pressures during shoulder abduction/adduction, but also shows a change in movement strategies between MGs. In the first two repetitions, Subject 3 exhibits a double peak response, indicating he loaded the sensor in both abduction and adduction. However, this pattern is not seen again for subsequent repetitions. The repetitions in MG 2 are similar in shape and peak, followed by decreasing pressure peaks during MG 3 and a sharpening of the profile for the final two repetitions. Overall, there were 34 complete sets of 12 pressure profiles for which the H-test is calculated. Of these, 54% tests are found to be inconsistent (po0.05), with 66% of tests with po0.1. A summary of statistical evaluate by subject is shown in Table 1. As shown, however, additional variability is apparently when qualitatively analyzing the profiles.
3.3. Subjective feedback In general all subjects felt they were consistent in their movements. The tasks that subjects rated as least consistent are shoulder abduction/adduction and cross body reach. The reasons given are the programming of the suit and working against the rotation of the bearings to keep the movements isolated to a single joint. These are also the tasks subjects
Fig. 5. Pressure contact over the upper forearm for all subjects during elbow flexion/extension. Data shows changes both within and across movement groups for all subjects. Average peak pressures for Subjects 2 and 3 are similar, although no direct conclusions may be drawn. Symbol † indicates marginal significance, po 0.10.
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
223
Fig. 6. Example data demonstrating pressure profile changes. (A) Subject 1 performing shoulder flexion/extension. Data shown over the wrist exhibits statistically significant change across movement groups. Additional variation may be seen within movement groups. (B) Subject 3 performing shoulder abduction/adduction shows a change in biomechanical movement strategies transitioning from a double peak response, to a single broad response, finishing in sharp peak response. Maximum peak pressures are not inconsistent. The subject also rated this task as the most fatiguing, potentially causing the change in pressure profiles. Symbol n indicates significance, p o 0.05.
Table 1 Consistency of movement by subject and group totals. A large percentage of movements were inconsistent for all subjects, with Subject 2 as the most inconsistent. Subject 1
# %
Subject 2
Subject 3
Combined
Total
p o 0.05
p o 0.1
Total
p o0.05
p o0.1
Total
po 0.05
po 0.1
Total
p o0.05
po 0.1
19
11 58
13 68
8
5 63
6 75
8
3 38
4 50
35
19 54
23 66
rated as requiring the greatest level of effort (However overall they felt the tasks require a “Reasonable Effort”, 3/5 on a Likert scale, or less). The data shown in Fig. 6 is for Subject 3 during shoulder abduction/adduction. After MG 1 he rates shoulder abduction/adduction the least consistent of the movement tasks and the most fatiguing. This may be evidence of the slow, deleterious effect of fatigue on biomechanics beyond the subject's reasonable ability to identify it as such. He rates his movements as ‘Very Consistent’ for MG 2 and 3 and said he did not explicitly change movement strategies, although a change in movement strategy finds evidence in the data. The utility of subjective feedback as compared to quantitative data varies by subject. Subject 2 gives subjective comments that matched quantitative pressure readings by anatomical locations. He accurately notes changes in his own movement within a MG, which is reflected in the data. However, Subject 1 does not provide the same quality of feedback. In several motions, he does not note contact against his arm where the sensors detected pressures. In general, his feedback is less descriptive than the other subjects and he does not note many instances of contact variation over the course of the experiment although many of his motions are statistically inconsistent. Despite using a body graphic aid to assist in describing contact locations, subjects felt it was difficult to describe the location and nature of the contact since they are, “feeling contact all over,” which is not supported by the pressure data.
4. Discussion This research, to the authors knowledge, is the first published work to use an untethered wearable pressure sensing system inside a pressurized space suit to measure the human–suit contact during movement. Space suit evaluation is traditionally measured treating the human and space suit as a system, evaluating gross metrics of performance [4,25,26,35–38]. Previously, no technology has allowed their separation. This system is the first to specifically target the interface between the person and space suit at the body's surface to overcome this limitation, allowing us to move beyond external visual measures, such as motion capture and photogrammetry. Due to the distributed nature of the pressure sensing system, each sensor provides unique information about human movement while suited. Temporal activation of sensors show how a subject can be indexed in the suit in a way internally measured joint angles cannot [29–31]. The pressure profiles shows patterns for complicated dynamic movements in a sequence consistent with sensor placement on the body, a capability not previously possible. We hypothesize that subjects with experience working in the space suit would perform motion tasks with consistent movement strategies. Each of the subjects tested are experienced working in the Mark III space suit. Subjects rate themselves as consistent in their movement and rarely note changes in movement strategies. Comparing against quantitative pressure profiles, however, gives a much
224
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
greater sense of the variability of movement and minor changes in loading on the body. Over 50% of the movements statistically analyzed are found to be inconsistent, and qualitatively analyzing the remaining profiles reveals an even greater degree of inconsistency. It cannot be extrapolated from these results how changes in consistency would affect a subject's propensity for injury since longer time scales and repetitions also affect injury propensity. However, it is clear that inconsistencies in movement are present in many motions both within and across MGs and trials. Based on the data presented herein, the hypothesis is rejected, experienced subjects are not found to be consistent in movement. These results, however, cannot be extrapolated to all subjects performing motions inside the space suit. This data represents three experienced subjects, but further results will differ due to differences in suit fit, subject movement strategies, and motions performed. Adding quantitative information to subjective feedback gives more precision to space suit assessment. Using internal pressure sensing to get a clearer understanding of human–suit interaction and suited biomechanics will be particularly important as EVA objectives move toward planetary exploration, where subjects will not be under ideal conditions performing simple movements. Rather, the nature of space suit testing will be focused on comfort, fit, and performance for complicated EVA tasks, which can be further assessed with this pressure measurement capability. It may be desirable for future studies to concentrate sensors in a smaller area of the limb to provide a pressure map with higher fidelity. Future work will use pressure sensing in a variety of conditions with more subjects to understand the differences between human and suit performance, potentially providing insight into the causal mechanisms of injury.
5. Conclusion Human–suit interaction is characterized for three subjects performing a series of upper body EVA movements while suited in NASA's Mark III space suit. Evaluating the temporal activation and magnitude of sensors allows the subject to be indexed in the suit to better understand suited biomechanics. Additional results show that subjects, although experienced working in the suit, do not perform suited motions consistently. Inconsistency of movement may be due to fatigue or change in biomechanical strategy. In either case, the changes may be early indicators of astronaut discomfort or potentially injury-producing over many additional repetitions. Despite detailed subjective feedback, subjects are unable to identify many points of contact between their bodies and the space suit. These experiments provide valuable insight into how space-suited motions occur, how consistent subjects are, the limitations of relying on subjective feedback, and how discomfort and fatigue may build up over time while working in the suit. An initial baseline of human–suit interaction is established and will guide future work to optimize sensor design, influence space suit design, improve biomechanical modeling, and ultimately prevent injuries that occur inside the space suit.
Acknowledgments The authors would like to acknowledge our colleagues at Dainese S.p.a., particularly Stefano Zanotto and Marcello Bencini. We would also like to acknowledge our collaborators at the Wyss Institute, Prof. Rob Wood, Prof. Yigit Menguc, Yong-Lae Park, Nicholas Lesniewski-Laas, and Daniel Vogt. We acknowledge our co-investigators Alexandra Hilbert and Pierre Bertrand. Finally, our colleagues at NASA for their support and collaboration, Shane McFarland and Amy Ross. This project is funded through NASA Grant NNX12AC09G, “Spacesuit Trauma Countermeasure System for Intravehicular and Extravehicular Activities”. Additional support provided by the National Science Foundation Graduate Research Fellowship Program, the MIT Portugal Program, and the Harvard Wyss Institute. The authors have filed a patent extension on an existing patent for the pressure sensing system used in this experiment. References [1] J. Hochstein, Astronaut Total Injury Database and Finger/Hand Injuries During EVA Training and Tasks, International Space University, Strausborg, FR, 2008. [2] J.A. Jones, R.B. Hoffman, D.A. Buckland, C.M. Harvey, C.K. Bowen, C.E. Hudy, S. Strauss, J. Novak, M.L. Gernhardt, The use of an extended ventilation tube as a countermasure for EVA-associated upper extremity medical issues, Acta Astronaut. 63 (2008) 763–768. [3] D. Longnecker, et al., in: I.o. Medicine (Ed.), Review of NASA's Longitudinal Study of Astronaut Health, National Academies Press, Washington, D.C., 2004. [4] D.A. Morgan, R. Wilmington, A. Pandya, J. Maida, K. Demel, Comparison of Extravehicular Mobility Unit (EMU) Suited and Unsuited Isolated Joint Strength Measurements, Johnson Space Center, Houston, TX, 1996. [5] R. Opperman, Astronaut Extravehicular Activity - Safety, Injury, and Countermeasures & Orbital Collisions & Space Debris - Incidence, Impact, and International Policy, Aeronautics and Astronautics, Technology Policy Program, Massachusetts Institute of Technology, Cambridge, MA, 2010, 183. [6] R.A. Scheuring, C.H. Mathers, J.A. Jones, M.L. Wear, Musculoskeletal injuries and minor trauma in space: incidence and injury mechanisms in U.S. astronauts, Aviat. Space Environ. Med. 80 (2009) 117–124. [7] S. Strauss, Extravehicular Mobility Unit Training Suit Symptom Study Report, Johnson Space Center, Houston, TX, 2004. [8] S. Viegas, J. Jones, S. Strauss, J. Clark, Physical demands and injuries to the upper extremity associated with the space program, J. Hand Surg. 29 (2004) 7. [9] D.R. Williams, B.J. Johnson, EMU Shoulder Injury Tiger Team Report, in: Houston, TX, 2003, p. 104. [10] C. Carr, Bioenergetics of Walking and Running in Space Suits, Aerospace and Astronautics, Massachusetts Institute of Technology, Cambridge, 2005. [11] C. Carr, D. Newman, Charachterization of a lower-body exoskeleton for simulation of space-suited locomotion, Acta Astronaut. 62 (2008) 308–323. [12] A. Rader, C. Carr, D. Newman, Loping: a strategy for reduced gravity locomotion?, in: Proceedings of the International Conference on Environmental Systems, Society for Automotive Engineers, 2007. [13] D. Newman, H. Alexander, Human locomotion and workload for simulated lunar and martian environments, Acta Astronaut. 29 (1993) 613–620. [14] D. Newman, H. Alexander, B. Webbon, Energetics and mechanics for partial gravity locomotion, Aviat. Space Environ. Med. 65 (1994) 815–823. [15] R. Scheuring, P. McCullouch, M. VanBaalen, C. Minard, R. Watson, S. Bowen, T. Blatt, Shoulder Injuries in US Astronauts Related to EVA Suit Design, Aerospace Medical Association, Atlanta, GA, 2012. [16] R. Scheuring, J. Jones, J. Novak, J. Polk, D. Billis, J. Schmid, J. Duncan, J. Davis, The apollo medical operations project: recommendations to improve crew health and performance for future exploration missions and lunar surface operations, Acta Astronaut. 2008 (2008) 8.
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
[17] P. Schmidt, An Investigation of Space Suit Mobility with Applications to EVA Operations, Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, MA, 2001, 254. [18] B. Holschuh, J. Waldie, J. Hoffman, D. Newman, Characterization of structural, volume, and pressure components to space suit joint rigidity, in: Proceedings of the International Conference on Environmental Systems, Society of Automotive Engineers, Savannah, GA, 2009. [19] D. Parry, J.L. Curry, D. Hanson, G. Towle, A Study of Techniques and Equipment for the Evaluation of Extravehicular Protective Garments, Hamilton Standard, Dayton, OH, 1966, 427. [20] I. Abramov, A. Stoklitsky, A. Barer, S. Filipenkov, Essential aspects of space suit operating pressure trade-off, in: Proceedings of the International Conference on Environmental Systems, Society of Automotive Engineers, Friedrichshafen, Germany, 1994. [21] P. Schmidt, D. Newman, E. Hodgson, Modeling space suit mobility: applications to design and operations, in: Proceedings of the International Conference on Environmental Systems, Society of Automotive Engineers, 2001. [22] M. Cowley, S. Margerum, L. Harvill, S. Rajulu, Model for predicting the performance of planetary suit hip bearing designs, in: V.G. Duffy (Ed.), Advances in Human Factors and Ergonomics, CRC Press, Houston, TX, 2012. [23] M. Gast, S. Moore, A glimpse from the inside of a space suit: whats it really like to train for an EVA? Acta Astronaut. 2011 (2010) 9. [24] A. Reinhardt, J. Magistad, AX-5 space suit reliability model, in: Proceedings of the International Conference on Environmental Systems, SAE, Williamsburg, VA, 1990, pp. 1057–1065. [25] J. Matty, L. Aitchison, A method for and issues associated with the determination of space suit joint requirements, in: Proceedings of the International Conference on Environmental Systems, SAE International Berlin, Germany, 2009. [26] D. Valish, K. Eversley, Space suit joint torque measurement method validation, in: Proceedings of the International Conference on Environmental Systems, American Institute of Aeronautics and Astronautics, San Diego, CA, 2012, p. 14. [27] F. Meyen, B. Holschuh, R. Kobrick, S. Jacobs, D. Newman, Robotic joint torque testing: a critical tool in the development of pressure suit mobility elements, in: Proceedings of the International Conference on Environmental Systems, AIAA, Portland, OR, 2011. [28] M. Greenisen, Effect of STS Space Suit on Astronaut Dominant Upper Limb EVA Work Performance, University of Houston, Houston, TX, 1986, 8. [29] M. Di Capua, D. Akin, Body pose measurement system (BPMS): an inertial motion capture system for biomechanics analysis and robot control from within a pressure suit, in: Proceedings of the International Conference on Environmental Systems, American Institute of Aeronautics and Astronautics, San Diego, 2012. [30] R. Kobrick, C. Carr, F. Meyen, A. Domingues, D. Newman, Using inertial measurement units for measuring spacesuit mobility and work envelope capability for intravehicular and extravehicular activties, in: Proceedings of the International Astronautical Congress, International Astronautical Federation, Naples, Italy, 2012, p. 9. [31] P. Bertrand, A. Anderson, A. Hilbert, D. Newman, Characterization of spacesuit kinematics and human–suit interactions, in: Proceedings of the International Conference on Environmental Systems, Texas Tech University, Tuscon, AZ, 2014.
225
[32] A. Anderson, A. Hilbert, P. Bertrand, S. McFarland, D. Newman, Insuit sensor systems for characterizing human–space suit interaction in: Proceedings of the International Conference on Environmental Systems, Texas Tech University, Tuscon, AZ, 2014. [33] A. Anderson, Understanding Human–Space Suit Interaction to Prevent Injury During Extravehicular Activity, Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, MA, 2014, 200. [34] Y.-L. Park, B.-R. Chen, R. Wood, Design and fabrication of soft artificial skin using embedded microchannels and liquid conductors, IEEE Sens. J. 12 (2012). [35] M.A. Jaramillo, B. Angermiller, R. Morency, S. Rajulu, Refinement of Optimal Work Enevelope for Extravehicular Activity Suit Operations, in: A.a.B. Facility (Ed.), Johnson Space Center, Houston, TX, 2008. [36] L. Aitchison, A comparison of methods for assessing space suit joint ranges of motion, in: Proceedings of the International Conference on Environmental Systems, American Institute of Aeronautics and Astronautics, San Diego, CA, 2012, p. 12. [37] J. Norcross, K. Clowers, T. Clark, L. Harvill, R. Morency, L. Stroud, L. Desantis, J. Vos, M. Gernhardt, Metabolic Costs and Biomechanics of Level Ambulation in a Planetary Suit, Johnson Space Center, Houston, TX, 2010, 74. [38] J. Norcross, L. Lee, K. Clowers, R. Morency, L. Desantis, J. Witt, J. Jones, J. Vos, M. Gernhardt, Feasibility of Performing a Suited 10km Ambulation on the Moon - Final Report of the EVA Walkback Test (EWT), Johnson Space Center, Houston, TX, 2009, 48. Dr. Allison Anderson received a Bachelor of Science degree at the University of Southern California in astronautics engineering in 2007. She received the Master of Science degree in astronautics engineering in 2011, the Master of Science degree in 2011, and the Doctor of Philosophy in 2014 from the Massachusetts Institute of Technology. She is currently a Post-doctoral Research Fellow at Dartmouth College Geisel School of Medicine as a National Space Biomedical Research Institute First Award fellow. Her research interests focus on biomedical device development for human space flight applications. Prof. Dava Newman received the Bachelor of Science degree in aerospace engineering from the University of Notre Dame in 1986, the Masters of Science degrees in aeronautics and astronautics and technology and policy in 1989, and the Ph.D. degree from the Massachusetts Institute of Technology (MIT) in aerospace biomedical engineering in 1992. She is currently a Professor in the Department of Aeronautics and Astronautics and Engineering Systems at MIT and affiliate faculty in the Harvard-MIT Health Sciences and Technology Program. She leads research efforts in extravehicular activity (EVA) including advanced space suit design, musculoskeletal modeling, biomechanics and energetics, humanrobotic cooperation, and biomedical devices.
Contents lists available at ScienceDirect
Acta Astronautica journal homepage: www.elsevier.com/locate/actaastro
Pressure sensing for in-suit measurement of space suited biomechanics Allison P. Anderson a,n, Dava J. Newman b a b
Massachusetts Institute of Technology, 77 Massachusetts Ave., 37-217, Cambridge, MA 02139, United States Massachusetts Institute of Technology, 77 Massachusetts Ave., 33-319, Cambridge, MA 02139, United States
a r t i c l e in f o
abstract
Article history: Received 20 January 2015 Received in revised form 15 April 2015 Accepted 18 May 2015 Available online 27 May 2015
Extravehicular Activity (EVA) is a critical component of human spaceflight, but working in gaspressurized space suits causes fatigue, excessive energy expenditure, and injury. Relatively little is known about how the astronaut moves and interacts with the space suit, and what factors lead to injury. A wearable pressure sensing system to quantitatively measure areas on the body's surface that the space suit impacts during dynamic EVA movement is developed. The system is used to characterize human–suit interaction in the NASA Mark III space suit. Three experienced subjects perform a series of upper body movements: 3 isolated joint movements and 2 functional tasks. Movements are repeated 12 times each and in-suit pressure responses are evaluated both by quantifying peak pressure and full profile responses. Results: Sequential sensor activation allows subjects to be indexed inside the space suit during complicated motions to better understand suited biomechanics. Subjectively, subjects generally feel they are consistent for all movements. However, using a nonparametric H-test, 54% of movements are found to be biomechanically inconsistent (po0.05). This experiment provides the first “window” inside the space suit to evaluate contact pressures and sequential indexing of the person inside the suit for realistic EVA movement. It cannot be extrapolated how changes in contact pressure would affect a subject's propensity for injury as injuries accumulate over long time scales. However, changes in pressure may be due to alterations in biomechanical strategies or fatigue, both of which could be precursors for injury and discomfort. & 2015 IAA. Published by Elsevier Ltd. All rights reserved.
Keywords: Space suit Pressure sensing Astronaut injury Human–suit interaction Arm biomechanics
1. Introduction Gas pressurized space suits cause injuries and significantly increase metabolic expenditure [1–14]. The current U. S. space suit, the extravehicular mobility unit (EMU), causes a variety of musculoskeletal injuries, including skin and surface injuries, where astronauts impact and rub against
Abbreviations: EVA, Extravehicular Activity; EMU, Extravehicular Mobility Unit; MG, Movement group n Correspondence to: 77 Massachusetts Ave., 37-217, Cambridge, MA 02139, United States. Tel.: þ1 417 388 0621. E-mail addresses: [email protected] (A.P. Anderson), [email protected] (D.J. Newman). http://dx.doi.org/10.1016/j.actaastro.2015.05.024 0094-5765/& 2015 IAA. Published by Elsevier Ltd. All rights reserved.
the suit as they bend the garment [7,15]. Although no EVArelated injury has prevented successful completion of mission objectives, there have been several instances when the EVA was nearly terminated due to suit discomfort [6,16]. Crew health, comfort, and safety are negatively affected by the human–suit impact. As of 2012, a total of 25 shoulder surgeries were performed on astronauts, most attributed to training in the spacesuit underwater [15]. Space suited motion is different from normal biomechanical movement. The suit is pressurized with gas to 29.6 kPa (4.3 psi) making it stiff in the vacuum of space. Additional rigidity comes both from changes in suit volume and fabric stiffness as the joints bend [17–21]. The suit is primarily composed of fabric pieces called “soft goods”, mounted to a
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
hard central torso. The soft goods are connected with metal rotational bearings and sizing rings. Programming, or planes through which the suit can move due to the angle of bearings [22], further limits mobility and makes movement unnatural. To perform EVA, astronauts must learn to change their biomechanical movement strategies, rather than attempting to move as they do naturally, or unsuited [23]. The difference between how a person moves relative to the suit has not been quantified. Previous studies used a variety of techniques, such as photogrammetry, motion capture, and ergonomic strength measurement to evaluate suited performance [4,17–19,24–28]. Results, however, are difficult to compare because methodologies and measurements inherently vary. Also, these methods cannot evaluate suited human biomechanics because they measure performance from the outside of the suit, characterizing the human and space suit as a whole. Recent work has focused on determining body joint angles within the suit [29–31], but knowing the nature and location of body-suit contact is needed to evaluate suited biomechanics. A pressure–sensing system specifically designed for dynamic movement inside the environment of the space suit is developed, and an experimental evaluation of human–suit interaction on the upper arm is presented. Human–space suit interaction is evaluated by assessing body placement within the suit and consistency of movement. We hypothesize that subjects with experience working in the space suit will perform motion tasks with consistent movement strategies. Subject experience is determined by previous time spent performing tests inside the suit. Movement strategies are defined by peak pressures averaged over trials or full time averaged pressure profiles. Consistency of movement is an important metric revealing fatigue or changes in biomechanical strategies, both of which could be precursors to EVA injury. Additionally, subjective feedback is evaluated as a proxy for measuring human–suit interaction using the techniques proposed here.
219
2. Methods The Polipo (Octopus), seen in Fig. 1, is the system of 12 sensors developed as part of this research effort for the pressure regime expected to be measured on the body under the space suit soft goods. The sensors are 2.5 cm in diameter and cover the pressure range from 0–100 kPa with approximately 1 kPa resolution [32–34]. These sensors are placed on the left arm in a way that targets anticipated hot spots of body–suit contact points, and secondarily for uniform coverage. Anticipated hot spots were determined through discussion with subject matter experts. The Polipo is integrated into a conformal wearable garment with placement targets to which the sensors attach. This experiment is performed on three subjects at NASA's Johnson Space Center in the Advanced Space Suit Laboratory. The test is performed in the Mark III space suit, shown in Fig. 1. Suit pressure is maintained at 29.6 kPa (4.3 psi). All subjects are space suit engineers, and have extensive experience performing similar motions inside the Mark III during experiments. Subjects were briefed on any potential hazards associated with the experiment. Informed consent was obtained. The protocol is approved both by the MIT Committee on the use of Humans as Experimental Subjects and the NASA Johnson Space Center's Institutional Review Board. Subject 1 is left handed, while Subjects 2 and 3 are right handed. This does not effect in the way subjects perform the experiment, since the subjects performed the motions with both arms simultaneously. However, it is not possible to determine if handedness affects the pressure measurements since many variables may contribute to differences across subjects. Subjects are asked to perform a series of upper body motions. The protocol consists of 12 repetitions of 5 motions, described in detail in Fig. 2. For each motion, the 12 repetitions are further subdivided into 3 movement groups (MG) of 4 repetitions each. This is done to evaluate subject fatigue
Fig. 1. Experimental hardware. (A) The Polipo pressure sensing system is a network of 12 sensors designed for a 0–100 kPa pressure range during dynamic upper body movement. (B) The Mark III space suit was evaluated in a human subject experiment to evaluate contact pressures on the person's body while moving.
220
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
or potential change of biomechanical strategies over the course of the test period. After each MG, the subject rests for 5 min and qualitative information is gathered. Perceived rate of exertion is evaluated by the test operator. Effort and perceived consistency of movement are evaluated by the subject on a 5-point Likert scale. Contact location and the nature of the contact (i.e. sharp, broad, etc.) are recorded using a body schematic and discomfort is rated on a 10 point Likert scale as described by the subject. A representative schematic of the test protocol is shown in Fig. 3. Static data taken prior to donning the suit and after pressurization is used to establish baseline loading of the suit on the body in the subject's relaxed standing posture while not resting in the suit's donning stand. The wiring of several wearable sensors began to deteriorate over the course of the experiment, causing erroneous jumps in the measured pressure. Therefore, this data is excluded from analysis. A total of 495 pressure profiles are evaluated. The peak pressure for each repetition within the MG is recorded. When the peak pressure is collected for all 12 repetitions, a Kruskal–Wallis H-test, equivalent to a nonparametric single factor ANOVA, is performed to evaluate the subject's consistency between MGs. The test is significant when the peak pressures within (at least) one group are significantly different than the others. Note that when data are highly variable within the MG itself, the H-test would not yield a
significant result. Therefore, it is not a measure of consistency, but rather inconsistency. 3. Results Due to the limited number of subjects, general conclusions about space suited biomechanics cannot be drawn. Rather, data from each subject is analyzed individually to present the first quantitative pressure measurements recorded inside a space suit. The pressure data is used to index a subject inside the space suit to determine when contact occurs and the nature of that contact. Quantitative evaluation of motion is useful in assessing the variability associated with suited biomechanics and subjective reporting of space suit contact, which might lead to injury. 3.1. Temporal data analysis Fig. 4 shows our first ‘look inside the suit’ plotting the pressure profiles of 6 sensors during the cross body reach motion for Subject 1. The data are normalized by time and plotted by mean and standard deviation of all repetitions within MG 1 to show the activation and relative pressure amplitudes over the course of the movement. Standard deviations show higher variability for loading certain sensors than others. Subject 1 initially begins in a resting
Fig. 2. Movement tasks performed by each subject. Three isolated joint tasks are performed: Elbow Flexion/Extension, Shoulder Flexion/Extension, and Shoulder Abduction/Adduction. Two functional tasks were performed: Cross Body Reach and Overhead Hammering. Subjects were given very specific instructions on how to perform the isolated joint tasks, while subjects were given way-point markers to meet and allowed to develop their own biomechanical strategies for the functional tasks. Subject suited in Mark III suit is shown.
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
Fig. 3. Experimental test protocol for a single subject. Subjects are given time to train each of the 5 movements inside the space suit. Subjective information is taken on comfort and pressure hot spots. The 5 movements are performed in 3 groups with subjective information taken after each group. The order is counterbalanced within the group and randomized between subjects. Each of the movements is repeated 4 times each within a movement group. Sensor pressure profiles over time are recorded for analysis.
221
posture. Only Sensor 10 is loaded due to the back of the arm resting on the space suit joint. The first 10–30% of the movement is associated with moving the arm from the neutral position across the body to the opposite hip. Sensor 1 is highly loaded during the portion of the movement since it is on the inside of the wrist and the subject is moving his arm across the body. Next he raises his arm to chest level on the opposite side. Sensor 11, located on the upper bicep, is the only sensor activated during this period due to the weight of the space suit resting on his arm. Between 40– 60% of the movement the subject sweeps his arm in front of his body at chest level in front of his shoulder. Sensor 2 is slightly loaded over this period, while Sensor 5 becomes loaded half way through this section. As Subject 1 crossed the body, he performs a shoulder rotation to prepare for the next phase of movement, the elbow bend. This rotation causes Sensor 5 to be loaded on the inside of the forearm near the elbow. Between 60–80% of the movement an elbow flexion/extension is performed to touch the helmet near his ear. As expected, this loads Sensors 1, 2, 5, and 6, all the lower arm and elbow sensors. Finally, he extends his elbow and lowers his arm to the neutral position in the last 90–100% of the movement. This re-loads Sensor 10 since it is loaded in the neutral posture as the subject's arm rests on
Fig. 4. Temporal activation of sensors for cross body reach. Results are for Subject 1, movement group 1. Response is averaged over each repetition and normalized by time. Sensors are shown with most distal on top of figure and proximal at the bottom. The Y axis is consistent for all sensors to compare relative pressure magnitudes.
222
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
the suit. Sensor 11 is also slightly loaded as the subject brings his arm down causing contact with the bicep due to the weight of the suit arm. This sequence of arm–suit interaction and loading is consistent for Subject 1 over all three MGs. The capability to discern movement patterns as a subject performs a task shows in great detail how the person's body shifts within the suit and moves the space suit as he or she attempts to complete a movement. The pressure sensors and wearable garment technology developed in this study provides a novel capability to measure body–suit interactions.
3.2. Movement consistency Example pressure profiles are shown in Figs. 5 and 6 to provide a sense of inter- and intra-subject consistency, but do not reflect all motions and profile responses. Profiles are plotted by MG and are normalized by time of the movement. The peak pressure of each of the 12 repetitions is used to calculate the mean (m) and standard deviation of the peak pressures. Fig. 5 shows pressure response profiles for all 3 subjects during elbow flexion/extension over the upper forearm, distal of the elbow crease. Subject 1 produces consistent pressures during MG 1, but varies widely in MG 2 producing the highest pressure seen for all subjects during this movement over this area of the body. The sensor wiring broke inside the suit prior to MG 3, so statistical analysis was not performed across all MGs for this motion. However, the subject's inconsistency is apparent. Within each of the three MGs, Subject 2 exhibits a steady decrease in peak pressure over the four repetitions. Overall, the peak pressures were lower in MG 1, exhibiting marginal inconsistency (H¼5.55, p¼0.06). Subject 3 produces similar pressure profiles with each repetition, although the peaks in MG 2 are slightly elevated compared to the other two MGs (H¼5.38, p¼ 0.06).
In MG 3, subject 3 also exhibits contact pressure between repetitions, likely due to a change in the positioning of his arm, altering the overall shape of the response profile. Fig. 6 gives additional examples to highlight subject movement consistency. Data for Subject 1 is collected over the wrist during shoulder flexion/extension. The subject's movements are inconsistent (H¼8.8, p¼0.01). During MG 1, the subject exhibits a steady increase in peak pressures, while in MG 2 there is a steady decrease. During MG 3, the peak pressures are consistent, but are 25% of the magnitudes seen in previous repetitions, indicating a change in movement. By contrast, data from Subject 1 over the upper forearm shows consistent peak pressures during shoulder abduction/adduction, but also shows a change in movement strategies between MGs. In the first two repetitions, Subject 3 exhibits a double peak response, indicating he loaded the sensor in both abduction and adduction. However, this pattern is not seen again for subsequent repetitions. The repetitions in MG 2 are similar in shape and peak, followed by decreasing pressure peaks during MG 3 and a sharpening of the profile for the final two repetitions. Overall, there were 34 complete sets of 12 pressure profiles for which the H-test is calculated. Of these, 54% tests are found to be inconsistent (po0.05), with 66% of tests with po0.1. A summary of statistical evaluate by subject is shown in Table 1. As shown, however, additional variability is apparently when qualitatively analyzing the profiles.
3.3. Subjective feedback In general all subjects felt they were consistent in their movements. The tasks that subjects rated as least consistent are shoulder abduction/adduction and cross body reach. The reasons given are the programming of the suit and working against the rotation of the bearings to keep the movements isolated to a single joint. These are also the tasks subjects
Fig. 5. Pressure contact over the upper forearm for all subjects during elbow flexion/extension. Data shows changes both within and across movement groups for all subjects. Average peak pressures for Subjects 2 and 3 are similar, although no direct conclusions may be drawn. Symbol † indicates marginal significance, po 0.10.
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
223
Fig. 6. Example data demonstrating pressure profile changes. (A) Subject 1 performing shoulder flexion/extension. Data shown over the wrist exhibits statistically significant change across movement groups. Additional variation may be seen within movement groups. (B) Subject 3 performing shoulder abduction/adduction shows a change in biomechanical movement strategies transitioning from a double peak response, to a single broad response, finishing in sharp peak response. Maximum peak pressures are not inconsistent. The subject also rated this task as the most fatiguing, potentially causing the change in pressure profiles. Symbol n indicates significance, p o 0.05.
Table 1 Consistency of movement by subject and group totals. A large percentage of movements were inconsistent for all subjects, with Subject 2 as the most inconsistent. Subject 1
# %
Subject 2
Subject 3
Combined
Total
p o 0.05
p o 0.1
Total
p o0.05
p o0.1
Total
po 0.05
po 0.1
Total
p o0.05
po 0.1
19
11 58
13 68
8
5 63
6 75
8
3 38
4 50
35
19 54
23 66
rated as requiring the greatest level of effort (However overall they felt the tasks require a “Reasonable Effort”, 3/5 on a Likert scale, or less). The data shown in Fig. 6 is for Subject 3 during shoulder abduction/adduction. After MG 1 he rates shoulder abduction/adduction the least consistent of the movement tasks and the most fatiguing. This may be evidence of the slow, deleterious effect of fatigue on biomechanics beyond the subject's reasonable ability to identify it as such. He rates his movements as ‘Very Consistent’ for MG 2 and 3 and said he did not explicitly change movement strategies, although a change in movement strategy finds evidence in the data. The utility of subjective feedback as compared to quantitative data varies by subject. Subject 2 gives subjective comments that matched quantitative pressure readings by anatomical locations. He accurately notes changes in his own movement within a MG, which is reflected in the data. However, Subject 1 does not provide the same quality of feedback. In several motions, he does not note contact against his arm where the sensors detected pressures. In general, his feedback is less descriptive than the other subjects and he does not note many instances of contact variation over the course of the experiment although many of his motions are statistically inconsistent. Despite using a body graphic aid to assist in describing contact locations, subjects felt it was difficult to describe the location and nature of the contact since they are, “feeling contact all over,” which is not supported by the pressure data.
4. Discussion This research, to the authors knowledge, is the first published work to use an untethered wearable pressure sensing system inside a pressurized space suit to measure the human–suit contact during movement. Space suit evaluation is traditionally measured treating the human and space suit as a system, evaluating gross metrics of performance [4,25,26,35–38]. Previously, no technology has allowed their separation. This system is the first to specifically target the interface between the person and space suit at the body's surface to overcome this limitation, allowing us to move beyond external visual measures, such as motion capture and photogrammetry. Due to the distributed nature of the pressure sensing system, each sensor provides unique information about human movement while suited. Temporal activation of sensors show how a subject can be indexed in the suit in a way internally measured joint angles cannot [29–31]. The pressure profiles shows patterns for complicated dynamic movements in a sequence consistent with sensor placement on the body, a capability not previously possible. We hypothesize that subjects with experience working in the space suit would perform motion tasks with consistent movement strategies. Each of the subjects tested are experienced working in the Mark III space suit. Subjects rate themselves as consistent in their movement and rarely note changes in movement strategies. Comparing against quantitative pressure profiles, however, gives a much
224
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
greater sense of the variability of movement and minor changes in loading on the body. Over 50% of the movements statistically analyzed are found to be inconsistent, and qualitatively analyzing the remaining profiles reveals an even greater degree of inconsistency. It cannot be extrapolated from these results how changes in consistency would affect a subject's propensity for injury since longer time scales and repetitions also affect injury propensity. However, it is clear that inconsistencies in movement are present in many motions both within and across MGs and trials. Based on the data presented herein, the hypothesis is rejected, experienced subjects are not found to be consistent in movement. These results, however, cannot be extrapolated to all subjects performing motions inside the space suit. This data represents three experienced subjects, but further results will differ due to differences in suit fit, subject movement strategies, and motions performed. Adding quantitative information to subjective feedback gives more precision to space suit assessment. Using internal pressure sensing to get a clearer understanding of human–suit interaction and suited biomechanics will be particularly important as EVA objectives move toward planetary exploration, where subjects will not be under ideal conditions performing simple movements. Rather, the nature of space suit testing will be focused on comfort, fit, and performance for complicated EVA tasks, which can be further assessed with this pressure measurement capability. It may be desirable for future studies to concentrate sensors in a smaller area of the limb to provide a pressure map with higher fidelity. Future work will use pressure sensing in a variety of conditions with more subjects to understand the differences between human and suit performance, potentially providing insight into the causal mechanisms of injury.
5. Conclusion Human–suit interaction is characterized for three subjects performing a series of upper body EVA movements while suited in NASA's Mark III space suit. Evaluating the temporal activation and magnitude of sensors allows the subject to be indexed in the suit to better understand suited biomechanics. Additional results show that subjects, although experienced working in the suit, do not perform suited motions consistently. Inconsistency of movement may be due to fatigue or change in biomechanical strategy. In either case, the changes may be early indicators of astronaut discomfort or potentially injury-producing over many additional repetitions. Despite detailed subjective feedback, subjects are unable to identify many points of contact between their bodies and the space suit. These experiments provide valuable insight into how space-suited motions occur, how consistent subjects are, the limitations of relying on subjective feedback, and how discomfort and fatigue may build up over time while working in the suit. An initial baseline of human–suit interaction is established and will guide future work to optimize sensor design, influence space suit design, improve biomechanical modeling, and ultimately prevent injuries that occur inside the space suit.
Acknowledgments The authors would like to acknowledge our colleagues at Dainese S.p.a., particularly Stefano Zanotto and Marcello Bencini. We would also like to acknowledge our collaborators at the Wyss Institute, Prof. Rob Wood, Prof. Yigit Menguc, Yong-Lae Park, Nicholas Lesniewski-Laas, and Daniel Vogt. We acknowledge our co-investigators Alexandra Hilbert and Pierre Bertrand. Finally, our colleagues at NASA for their support and collaboration, Shane McFarland and Amy Ross. This project is funded through NASA Grant NNX12AC09G, “Spacesuit Trauma Countermeasure System for Intravehicular and Extravehicular Activities”. Additional support provided by the National Science Foundation Graduate Research Fellowship Program, the MIT Portugal Program, and the Harvard Wyss Institute. The authors have filed a patent extension on an existing patent for the pressure sensing system used in this experiment. References [1] J. Hochstein, Astronaut Total Injury Database and Finger/Hand Injuries During EVA Training and Tasks, International Space University, Strausborg, FR, 2008. [2] J.A. Jones, R.B. Hoffman, D.A. Buckland, C.M. Harvey, C.K. Bowen, C.E. Hudy, S. Strauss, J. Novak, M.L. Gernhardt, The use of an extended ventilation tube as a countermasure for EVA-associated upper extremity medical issues, Acta Astronaut. 63 (2008) 763–768. [3] D. Longnecker, et al., in: I.o. Medicine (Ed.), Review of NASA's Longitudinal Study of Astronaut Health, National Academies Press, Washington, D.C., 2004. [4] D.A. Morgan, R. Wilmington, A. Pandya, J. Maida, K. Demel, Comparison of Extravehicular Mobility Unit (EMU) Suited and Unsuited Isolated Joint Strength Measurements, Johnson Space Center, Houston, TX, 1996. [5] R. Opperman, Astronaut Extravehicular Activity - Safety, Injury, and Countermeasures & Orbital Collisions & Space Debris - Incidence, Impact, and International Policy, Aeronautics and Astronautics, Technology Policy Program, Massachusetts Institute of Technology, Cambridge, MA, 2010, 183. [6] R.A. Scheuring, C.H. Mathers, J.A. Jones, M.L. Wear, Musculoskeletal injuries and minor trauma in space: incidence and injury mechanisms in U.S. astronauts, Aviat. Space Environ. Med. 80 (2009) 117–124. [7] S. Strauss, Extravehicular Mobility Unit Training Suit Symptom Study Report, Johnson Space Center, Houston, TX, 2004. [8] S. Viegas, J. Jones, S. Strauss, J. Clark, Physical demands and injuries to the upper extremity associated with the space program, J. Hand Surg. 29 (2004) 7. [9] D.R. Williams, B.J. Johnson, EMU Shoulder Injury Tiger Team Report, in: Houston, TX, 2003, p. 104. [10] C. Carr, Bioenergetics of Walking and Running in Space Suits, Aerospace and Astronautics, Massachusetts Institute of Technology, Cambridge, 2005. [11] C. Carr, D. Newman, Charachterization of a lower-body exoskeleton for simulation of space-suited locomotion, Acta Astronaut. 62 (2008) 308–323. [12] A. Rader, C. Carr, D. Newman, Loping: a strategy for reduced gravity locomotion?, in: Proceedings of the International Conference on Environmental Systems, Society for Automotive Engineers, 2007. [13] D. Newman, H. Alexander, Human locomotion and workload for simulated lunar and martian environments, Acta Astronaut. 29 (1993) 613–620. [14] D. Newman, H. Alexander, B. Webbon, Energetics and mechanics for partial gravity locomotion, Aviat. Space Environ. Med. 65 (1994) 815–823. [15] R. Scheuring, P. McCullouch, M. VanBaalen, C. Minard, R. Watson, S. Bowen, T. Blatt, Shoulder Injuries in US Astronauts Related to EVA Suit Design, Aerospace Medical Association, Atlanta, GA, 2012. [16] R. Scheuring, J. Jones, J. Novak, J. Polk, D. Billis, J. Schmid, J. Duncan, J. Davis, The apollo medical operations project: recommendations to improve crew health and performance for future exploration missions and lunar surface operations, Acta Astronaut. 2008 (2008) 8.
A.P. Anderson, D.J. Newman / Acta Astronautica 115 (2015) 218–225
[17] P. Schmidt, An Investigation of Space Suit Mobility with Applications to EVA Operations, Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, MA, 2001, 254. [18] B. Holschuh, J. Waldie, J. Hoffman, D. Newman, Characterization of structural, volume, and pressure components to space suit joint rigidity, in: Proceedings of the International Conference on Environmental Systems, Society of Automotive Engineers, Savannah, GA, 2009. [19] D. Parry, J.L. Curry, D. Hanson, G. Towle, A Study of Techniques and Equipment for the Evaluation of Extravehicular Protective Garments, Hamilton Standard, Dayton, OH, 1966, 427. [20] I. Abramov, A. Stoklitsky, A. Barer, S. Filipenkov, Essential aspects of space suit operating pressure trade-off, in: Proceedings of the International Conference on Environmental Systems, Society of Automotive Engineers, Friedrichshafen, Germany, 1994. [21] P. Schmidt, D. Newman, E. Hodgson, Modeling space suit mobility: applications to design and operations, in: Proceedings of the International Conference on Environmental Systems, Society of Automotive Engineers, 2001. [22] M. Cowley, S. Margerum, L. Harvill, S. Rajulu, Model for predicting the performance of planetary suit hip bearing designs, in: V.G. Duffy (Ed.), Advances in Human Factors and Ergonomics, CRC Press, Houston, TX, 2012. [23] M. Gast, S. Moore, A glimpse from the inside of a space suit: whats it really like to train for an EVA? Acta Astronaut. 2011 (2010) 9. [24] A. Reinhardt, J. Magistad, AX-5 space suit reliability model, in: Proceedings of the International Conference on Environmental Systems, SAE, Williamsburg, VA, 1990, pp. 1057–1065. [25] J. Matty, L. Aitchison, A method for and issues associated with the determination of space suit joint requirements, in: Proceedings of the International Conference on Environmental Systems, SAE International Berlin, Germany, 2009. [26] D. Valish, K. Eversley, Space suit joint torque measurement method validation, in: Proceedings of the International Conference on Environmental Systems, American Institute of Aeronautics and Astronautics, San Diego, CA, 2012, p. 14. [27] F. Meyen, B. Holschuh, R. Kobrick, S. Jacobs, D. Newman, Robotic joint torque testing: a critical tool in the development of pressure suit mobility elements, in: Proceedings of the International Conference on Environmental Systems, AIAA, Portland, OR, 2011. [28] M. Greenisen, Effect of STS Space Suit on Astronaut Dominant Upper Limb EVA Work Performance, University of Houston, Houston, TX, 1986, 8. [29] M. Di Capua, D. Akin, Body pose measurement system (BPMS): an inertial motion capture system for biomechanics analysis and robot control from within a pressure suit, in: Proceedings of the International Conference on Environmental Systems, American Institute of Aeronautics and Astronautics, San Diego, 2012. [30] R. Kobrick, C. Carr, F. Meyen, A. Domingues, D. Newman, Using inertial measurement units for measuring spacesuit mobility and work envelope capability for intravehicular and extravehicular activties, in: Proceedings of the International Astronautical Congress, International Astronautical Federation, Naples, Italy, 2012, p. 9. [31] P. Bertrand, A. Anderson, A. Hilbert, D. Newman, Characterization of spacesuit kinematics and human–suit interactions, in: Proceedings of the International Conference on Environmental Systems, Texas Tech University, Tuscon, AZ, 2014.
225
[32] A. Anderson, A. Hilbert, P. Bertrand, S. McFarland, D. Newman, Insuit sensor systems for characterizing human–space suit interaction in: Proceedings of the International Conference on Environmental Systems, Texas Tech University, Tuscon, AZ, 2014. [33] A. Anderson, Understanding Human–Space Suit Interaction to Prevent Injury During Extravehicular Activity, Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, MA, 2014, 200. [34] Y.-L. Park, B.-R. Chen, R. Wood, Design and fabrication of soft artificial skin using embedded microchannels and liquid conductors, IEEE Sens. J. 12 (2012). [35] M.A. Jaramillo, B. Angermiller, R. Morency, S. Rajulu, Refinement of Optimal Work Enevelope for Extravehicular Activity Suit Operations, in: A.a.B. Facility (Ed.), Johnson Space Center, Houston, TX, 2008. [36] L. Aitchison, A comparison of methods for assessing space suit joint ranges of motion, in: Proceedings of the International Conference on Environmental Systems, American Institute of Aeronautics and Astronautics, San Diego, CA, 2012, p. 12. [37] J. Norcross, K. Clowers, T. Clark, L. Harvill, R. Morency, L. Stroud, L. Desantis, J. Vos, M. Gernhardt, Metabolic Costs and Biomechanics of Level Ambulation in a Planetary Suit, Johnson Space Center, Houston, TX, 2010, 74. [38] J. Norcross, L. Lee, K. Clowers, R. Morency, L. Desantis, J. Witt, J. Jones, J. Vos, M. Gernhardt, Feasibility of Performing a Suited 10km Ambulation on the Moon - Final Report of the EVA Walkback Test (EWT), Johnson Space Center, Houston, TX, 2009, 48. Dr. Allison Anderson received a Bachelor of Science degree at the University of Southern California in astronautics engineering in 2007. She received the Master of Science degree in astronautics engineering in 2011, the Master of Science degree in 2011, and the Doctor of Philosophy in 2014 from the Massachusetts Institute of Technology. She is currently a Post-doctoral Research Fellow at Dartmouth College Geisel School of Medicine as a National Space Biomedical Research Institute First Award fellow. Her research interests focus on biomedical device development for human space flight applications. Prof. Dava Newman received the Bachelor of Science degree in aerospace engineering from the University of Notre Dame in 1986, the Masters of Science degrees in aeronautics and astronautics and technology and policy in 1989, and the Ph.D. degree from the Massachusetts Institute of Technology (MIT) in aerospace biomedical engineering in 1992. She is currently a Professor in the Department of Aeronautics and Astronautics and Engineering Systems at MIT and affiliate faculty in the Harvard-MIT Health Sciences and Technology Program. She leads research efforts in extravehicular activity (EVA) including advanced space suit design, musculoskeletal modeling, biomechanics and energetics, humanrobotic cooperation, and biomedical devices.