Robotic prostheses have the potential to significantly improve mobility for people with lower-limb amputation. Humans exhibit complex responses to mechanical interactions with these devices, however, and computational models are not yet able to predict such responses meaningfully. Experiments therefore play a critical role in development, but have been limited by the use of productlike prototypes, each requiring years of development and specialized for a narrow range of functions. Here we describe a robotic ankle-foot prosthesis system that enables rapid exploration of a wide range of dynamical behaviors in experiments with human subjects. This emulator comprises powerful off-board motor and control hardware, a flexible Bowden cable tether, and a lightweight instrumented prosthesis, resulting in a combination of low mass worn by the human (0.96 kg) and high mechatronic performance compared to prior platforms. Benchtop tests demonstrated closedloop torque bandwidth of 17 Hz, peak torque of 175 Nm, and peak power of 1.0 kW. Tests with an anthropomorphic pendulum "leg" demonstrated low interference from the tether, less than 1 Nm about the hip. This combination of low worn mass, high bandwidth, high torque, and unrestricted movement makes the platform exceptionally versatile. To demonstrate suitability for human experiments, we performed preliminary tests in which a subject with unilateral transtibial amputation walked on a treadmill at 1.25 ms À1 while the prosthesis behaved in various ways. These tests revealed low torque tracking error (RMS error of 2.8 Nm) and the capacity to systematically vary work production or absorption across a broad range (from À5 to 21 J per step). These results support the use of robotic emulators during early stage assessment of proposed device functionalities and for scientific study of fundamental aspects of human-robot interaction. The design of simple, alternate end-effectors would enable studies at other joints or with additional degrees of freedom.

Individuals with unilateral below-knee amputation expend more energy than non-amputees during walking and exhibit reduced push-off work and increased hip work in the affected limb. Simple dynamic models of walking suggest a possible solution, predicting that increasing prosthetic ankle push-off should decrease leading limb collision, thereby reducing overall energy requirements. We conducted a rigorous experimental test of this idea wherein ankle-foot prosthesis push-off work was incrementally varied in isolation from one-half to two-times normal levels while subjects with simulated amputation walked on a treadmill at 1.25 m·s−1. Increased prosthesis push-off significantly reduced metabolic energy expenditure, with a 14% reduction at maximum prosthesis work. In contrast to model predictions, however, collision losses were unchanged, while hip work during swing initiation was decreased. This suggests that powered ankle push-off reduces walking effort primarily through other mechanisms, such as assisting leg swing, which would be better understood using more complete neuromuscular models.

Amputees using passive ankle-foot prostheses tend to expend more metabolic energy during walking than non-amputees, and reducing this cost has been a central motivation for the development of active ankle-foot prostheses. Increased push-off work at the end of stance has been proposed as a way to reduce metabolic energy use, but the effects of push-off work have not been tested in isolation. In this experiment, participants with unilateral transtibial amputation (N=6) walked on a treadmill at a constant speed while wearing a powered prosthesis emulator. The prosthesis delivered different levels of ankle push-off work across conditions, ranging from the value for passive prostheses to double the value for non-amputee walking, while all other prosthesis mechanics were held constant. Participants completed six acclimation sessions prior to a data collection in which metabolic rate, kinematics, kinetics, muscle activity and user satisfaction were recorded. Metabolic rate was not affected by net prosthesis work rate (p=0.5; R=0.007). Metabolic rate, gait mechanics and muscle activity varied widely across participants, but no participant had lower metabolic rate with higher levels of push-off work. User satisfaction was affected by push-off work (p=0.002), with participants preferring values of ankle push-off slightly higher than in non-amputee walking, possibly indicating other benefits. Restoring or augmenting ankle push-off work is not sufficient to improve energy economy for lower-limb amputees. Additional necessary conditions might include alternate timing or control, individualized tuning, or particular subject characteristics.

BackgroundRobotic ankle-foot prostheses that provide net positive push-off work can reduce the metabolic rate of walking for individuals with amputation, but benefits might be sensitive to push-off timing. Simple walking models suggest that preemptive push-off reduces center-of-mass work, possibly reducing metabolic rate. Studies with bilateral exoskeletons have found that push-off beginning before leading leg contact minimizes metabolic rate, but timing was not varied independently from push-off work, and the effects of push-off timing on biomechanics were not measured. Most lower-limb amputations are unilateral, which could also affect optimal timing. The goal of this study was to vary the timing of positive prosthesis push-off work in isolation and measure the effects on energetics, mechanics and muscle activity.MethodsWe tested 10 able-bodied participants walking on a treadmill at 1.25 m · s−1. Participants wore a tethered ankle-foot prosthesis emulator on one leg using a rigid boot adapter. We programmed the prosthesis to apply torque bursts that began between 46% and 56% of stride in different conditions. We iteratively adjusted torque magnitude to maintain constant net positive push-off work.ResultsWhen push-off began at or after leading leg contact, metabolic rate was about 10% lower than in a condition with Spring-like prosthesis behavior. When push-off began before leading leg contact, metabolic rate was not different from the Spring-like condition. Early push-off led to increased prosthesis-side vastus medialis and biceps femoris activity during push-off and increased variability in step length and prosthesis loading during push-off. Prosthesis push-off timing had no influence on intact-side leg center-of-mass collision work.ConclusionsProsthesis push-off timing, isolated from push-off work, strongly affected metabolic rate, with optimal timing at or after intact-side heel contact. Increased thigh muscle activation and increased human variability appear to have caused the lack of reduction in metabolic rate when push-off was provided too early. Optimal timing with respect to opposite heel contact was not different from normal walking, but the trends in metabolic rate and center-of-mass mechanics were not consistent with simple model predictions. Optimal push-off timing should also be characterized for individuals with amputation, since meaningful benefits might be realized with improved timing.Electronic supplementary materialThe online version of this article (doi:10.1186/s12984-015-0014-8) contains supplementary material, which is available to authorized users.

Biomechatronic devices show promise for restoring human performance, but development has been made inefficient by the need for specialized autonomous devices prior to testing benefits of proposed functionalities. This has severely limited exploration within and across intervention strategies. We have developed a laboratory testbed suitable for emulating and rapidly assessing wearable robot designs. The testbed is comprised of powerful off-board motor and control hardware, a flexible tether, and lightweight instrumented end-effectors worn by a person. We performed a series of benchtop tests to gauge mechatronic performance, and found significant improvements over prior candidate testbed platforms. In particular, this system has an unusual combination of low worn mass (less than 1 kg), high closed-loop torque bandwidth (17 Hz), and high peak torque (175 N·m), key to emulating specialized devices. We also performed walking trials to gauge dynamic torque control and versatility. Walking trials with a prosthesis end-effector demonstrated precise torque tracking (4 N·m RMS error), both in time and joint-angle space, and versatile mechanical behavior through systematic changes in high-level control law parameters. For example, we widely varied net ankle work (from -3 J to 9 J per step) using an impedance law relating joint angle and velocity to desired torque. These results suggest such testbeds could be used to emulate and evaluate novel assistive robot concepts prior to laborious product design.