Continuous A-Mode Ultrasound-Based Prediction of Transfemoral Amputee Prosthesis Kinematics Across Different Ambulation Tasks.

OBJECTIVE: Volitional control systems for powered prostheses require the detection of user intent to operate in real life scenarios. Ambulation mode classification has been proposed to address this issue. However, these approaches introduce discrete labels to the otherwise continuous task that is ambulation. An alternative approach is to provide users with direct, voluntary control of the powered prosthesis motion. Surface electromyography (EMG) sensors have been proposed for this task, but poor signal-to-noise ratios and crosstalk from neighboring muscles limit performance. B-mode ultrasound can address some of these issues at the cost of reduced clinical viability due to the substantial increase in size, weight, and cost. Thus, there is an unmet need for a lightweight, portable neural system that can effectively detect the movement intention of individuals with lower-limb amputation. METHODS: In this study, we show that a small and lightweight A-mode ultrasound system can continuously predict prosthesis joint kinematics in seven individuals with transfemoral amputation across different ambulation tasks. Features from the A-mode ultrasound signals were mapped to the user's prosthesis kinematics via an artificial neural network. RESULTS: Predictions on testing ambulation circuit trials resulted in a mean normalized RMSE across different ambulation modes of 8.7 ± 3.1%, 4.6 ± 2.5%, 7.2 ± 1.8%, and 4.6 ± 2.4% for knee position, knee velocity, ankle position, and ankle velocity, respectively. CONCLUSION AND SIGNIFICANCE: This study lays the foundation for future applications of A-mode ultrasound for volitional control of powered prostheses during a variety of daily ambulation tasks.

Can a powered knee-ankle prosthesis improve weight-bearing symmetry during stand-to-sit transitions in individuals with above-knee amputations?

BACKGROUND: After above-knee amputation, the missing biological knee and ankle are replaced with passive prosthetic devices. Passive prostheses are able to dissipate limited amounts of energy using resistive damper systems during "negative energy" tasks like sit-down. However, passive prosthetic knees are not able to provide high levels of resistance at the end of the sit-down movement when the knee is flexed, and users need the most support. Consequently, users are forced to over-compensate with their upper body, residual hip, and intact leg, and/or sit down with a ballistic and uncontrolled movement. Powered prostheses have the potential to solve this problem. Powered prosthetic joints are controlled by motors, which can produce higher levels of resistance at a larger range of joint positions than passive damper systems. Therefore, powered prostheses have the potential to make sitting down more controlled and less difficult for above-knee amputees, improving their functional mobility. METHODS: Ten individuals with above-knee amputations sat down using their prescribed passive prosthesis and a research powered knee-ankle prosthesis. Subjects performed three sit-downs with each prosthesis while we recorded joint angles, forces, and muscle activity from the intact quadricep muscle. Our main outcome measures were weight-bearing symmetry and muscle effort of the intact quadricep muscle. We performed paired t-tests on these outcome measures to test for significant differences between passive and powered prostheses. RESULTS: We found that the average weight-bearing symmetry improved by 42.1% when subjects sat down with the powered prosthesis compared to their passive prostheses. This difference was significant (p = 0.0012), and every subject's weight-bearing symmetry improved when using the powered prosthesis. Although the intact quadricep muscle contraction differed in shape, neither the integral nor the peak of the signal was significantly different between conditions (integral p > 0.01, peak p > 0.01). CONCLUSIONS: In this study, we found that a powered knee-ankle prosthesis significantly improved weight-bearing symmetry during sit-down compared to passive prostheses. However, we did not observe a corresponding decrease in intact-limb muscle effort. These results indicate that powered prosthetic devices have the potential to improve balance during sit-down for individuals with above-knee amputation and provide insight for future development of powered prosthetics.

A-Mode Ultrasound-Based Prediction of Transfemoral Amputee Prosthesis Walking Kinematics Via an Artificial Neural Network.

Lower-limb powered prostheses can provide users with volitional control of ambulation. To accomplish this goal, they require a sensing modality that reliably interprets user intention to move. Surface electromyography (EMG) has been previously proposed to measure muscle excitation and provide volitional control to upper- and lower-limb powered prosthesis users. Unfortunately, EMG suffers from a low signal to noise ratio and crosstalk between neighboring muscles, often limiting the performance of EMG-based controllers. Ultrasound has been shown to have better resolution and specificity than surface EMG. However, this technology has yet to be integrated into lower-limb prostheses. Here we show that A-mode ultrasound sensing can reliably predict the prosthesis walking kinematics of individuals with a transfemoral amputation. Ultrasound features from the residual limb of 9 transfemoral amputee subjects were recorded with A-mode ultrasound during walking with their passive prosthesis. The ultrasound features were mapped to joint kinematics through a regression neural network. Testing of the trained model against untrained kinematics from an altered walking speed show accurate predictions of knee position, knee velocity, ankle position, and ankle velocity, with a normalized RMSE of 9.0 ± 3.1%, 7.3 ± 1.6%, 8.3 ± 2.3%, and 10.0 ± 2.5% respectively. This ultrasound-based prediction suggests that A-mode ultrasound is a viable sensing technology for recognizing user intent. This study is the first necessary step towards implementation of volitional prosthesis controller based on A-mode ultrasound for individuals with transfemoral amputation.

Analysis and Validation of Sensitivity in Torque-Sensitive Actuators.

Across different fields within robotics, there is a great need for lightweight, efficient actuators with human-like performance. Linkage-based passive variable transmissions and torque-sensitive transmissions have emerged as promising solutions to meet this need by significantly increasing actuator efficiency and power density, but their modeling and analysis remain an open research topic. In this paper, we introduce the sensitivity between input displacement and output torque as a key metric to analyze the performance of these complex mechanisms in dynamic tasks. We present the analytical model of sensitivity in the context of two different torque-sensitive transmission designs, and used this sensitivity metric to analyze the differences in their performance. Experiments with these designs implemented within a powered knee prosthesis were conducted, and results validated the sensitivity model as well as its role in predicting actuators' dynamic performance. Together with other design methods, sensitivity analysis is a valuable tool for designers to systematically analyze and create transmission systems capable of human-like physical behavior.

Effect of Increasing Assistance From a Powered Prosthesis on Weight-Bearing Symmetry, Effort, and Speed During Stand-Up in Individuals With Above-Knee Amputation.

After above-knee amputation, the missing biological knee and ankle are commonly replaced with a passive prosthesis, which cannot provide net-positive energy to assist the user. During activities such as sit-to-stand, above-knee amputees must compensate for this lack of power using their upper body, intact limb, and residual limb, resulting in slower, less symmetric, and higher effort movements. Previous studies have shown that powered prostheses can improve symmetry and speed by providing positive assistive power. However, we still lack a systematic investigation of the effect of powered prosthesis assistance. Without this knowledge, researchers and clinicians have no framework for tuning powered prostheses to optimally assist users. Here we show that varying the assistive knee torque significantly affected weight-bearing symmetry, effort, and speed during the stand-up movement in eight above-knee amputees. Specifically, we observed improvements in the index of asymmetry of the vertical ground reaction force at the point approximating maximum vertical center of mass acceleration, the integral of the intact vastus medialis activation measured using electromyography, and the stand-up duration compared to the passive prosthesis. We saw significant improvements in all three metrics when subjects used the powered prosthesis compared to the passive prosthesis. We saw improvements in all three metrics with increasing assistive torque levels commanded by the powered prosthesis. We also observed increased weight-bearing asymmetry at the end of movement, and increased kinematic asymmetry with increasing assistance from the powered prosthesis. These results show that powered prostheses can improve functional mobility, potentially increasing quality of life for millions of people living with above-knee amputations.

Powered Hip Exoskeleton Reduces Residual Hip Effort Without Affecting Kinematics and Balance in Individuals With Above-Knee Amputations During Walking.

OBJECTIVE: A unilateral, lightweight powered hip exoskeleton has been shown to improve walking economy in individuals with above-knee amputations. However, the mechanism responsible for this improvement is unknown. In this study we assess the biomechanics of individuals with above-knee amputations walking with and without a unilateral, lightweight powered hip exoskeleton. We hypothesize that assisting the residual limb will reduce the net residual hip energy. METHODS: Eight individuals with above-knee amputations walked on a treadmill at 1 m/s with and without a unilateral powered hip exoskeleton. Flexion/extension assistance was provided to the residual hip. Motion capture and inverse dynamic analysis were performed to assess gait kinematics, kinetics, center of mass, and center of pressure. RESULTS: The net energy at the residual hip decreased from 0.05±0.04 J/kg without the exoskeleton to -0.01±0.05 J/kg with the exoskeleton (p = 0.026). The cumulative positive energy of the residual hip decreased on average by 18.2% with 95% confidence intervals (CI) (0.20 J/kg, 0.24 J/kg) and (0.16 J/kg, 0.20 J/kg) without and with the exoskeleton, respectively. During stance, the hip extension torque of the residual limb decreased on average by 37.5%, 95% CI (0.28 Nm/kg, 0.36 Nm/kg), (0.17 Nm/kg, 0.23 Nm/kg) without and with the exoskeleton, respectively. CONCLUSION: Powered hip exoskeleton assistance significantly reduced the net residual hip energy, with concentric energy being the main contributor to this change. We believe that the reduction in residual hip extension torque during early stance is the main contributor to this reduction. SIGNIFICANCE: This analysis shows that by assisting the residual hip, the exoskeleton significantly decreased the net hip energy produced by the residual limb, which may explain the improvements in walking economy previously observed.

Powered Knee and Ankle Prosthesis Control for Adaptive Ambulation at Variable Speeds, Inclines, and Uneven Terrains.

Ambulation in everyday life requires walking at variable speeds, variable inclines, and variable terrains. Powered prostheses aim to provide this adaptability through control of the actuated joints. Some powered prosthesis controllers can adapt to discrete changes in speed and incline but require manual tuning to determine the control parameters, leading to poor clinical viability. Other data-driven controllers can continuously adapt to changes in speed and incline but do so by imposing the same non-amputee gait patterns for all amputee subjects, which does not consider subjective preferences and differing clinical needs of users. Here, we present a controller for powered knee and ankle prostheses that can continuously adapt to different walking speeds, inclines, and uneven terrains without enforcing a specific prosthesis position, impedance, or torque. A virtual biarticular muscle connection determines the knee flexion torque, which changes with both speed and slope. Adaptation to inclines and uneven terrains is based solely on the global shank orientation. Continuously variable damping allows for speed adaptation. Minimum-jerk programming defines the prosthesis swing trajectory at variable cadences. Experiments with one individual with an above-knee amputation suggest that the proposed controller can effectively adapt to different walking speeds, inclines, and rough terrains.

A Unified Controller for Natural Ambulation on Stairs and Level Ground with a Powered Robotic Knee Prosthesis.

Powered lower-limb prostheses have the potential to improve amputee mobility by closely imitating the biomechanical function of the missing biological leg. To accomplish this goal, powered prostheses need controllers that can seamlessly adapt to the ambulation activity intended by the user. Most powered prosthesis control architectures address this issue by switching between specific controllers for each activity. This approach requires online classification of the intended ambulation activity. Unfortunately, any misclassification can cause the prosthesis to perform a different movement than the user expects, increasing the likelihood of falls and injuries. Therefore, classification approaches require near-perfect accuracy to be used safely in real life. In this paper, we propose a unified controller for powered knee prostheses which allows for walking, stair ascent, and stair descent without the need for explicit activity classification. Experiments with one individual with an above-knee amputation show that the proposed controller enables seamless transitions between activities. Moreover, transition between activities is possible while leading with either the sound-side or the prosthesis. A controller with these characteristics has the potential to improve amputee mobility.

Series-elastic actuator with two degree-of-freedom PID control improves torque control in a powered knee exoskeleton.

Powered exoskeletons need actuators that are lightweight, compact, and efficient while allowing for accurate torque control. To satisfy these requirements, researchers have proposed using series elastic actuators (SEAs). SEAs use a spring in series with rotary or linear actuators. The spring compliance, in conjunction with an appropriate control scheme, improves torque control, efficiency, output impedance, and disturbance rejection. However, springs add weight to the actuator and complexity to the control, which may have negative effects on the performance of the powered exoskeleton. Therefore, there is an unmet need for new SEA designs that are lighter and more efficient than available systems, as well as for control strategies that push the performance of SEA-based exoskeletons without requiring complex modeling and tuning. This article presents the design, development, and testing of a novel SEA with high force density for powered exoskeletons, as well as the use of a two degree-of-freedom (2DOF) PID system to improve output impedance and disturbance rejection. Benchtop testing results show reduced output impedance and damping values when using a 2DOF PID controller as compared to a 1DOF PID controller. Human experiments with three able-bodied subjects (N = 3) show improved torque tracking with reduced root-mean-square error by 45.2% and reduced peak error by 49.8% when using a 2DOF PID controller. Furthermore, EMG data shows a reduction in peak EMG value when using the exoskeleton in assistive mode compared to the exoskeleton operating in transparent mode.

Volitional EMG Control Enables Stair Climbing with a Robotic Powered Knee Prosthesis.

Existing controllers for robotic powered prostheses regulate the prosthesis speed, timing, and energy generation using predefined position or torque trajectories. This approach enables climbing stairs step-over-step. However, it does not provide amputees with direct volitional control of the robotic prosthesis, a functionality necessary to restore full mobility to the user. Here we show that proportional electromyographic (EMG) control of the prosthesis knee torque enables volitional control of a powered knee prosthesis during stair climbing. The proposed EMG controller continuously regulates knee torque based on activation of the residual hamstrings, measured using a single EMG electrode located within the socket. The EMG signal is mapped to a desired knee flexion/extension torque based on the prosthesis knee position, the residual limb position, and the interaction with the ground. As a result, the proposed EMG controller enabled an above-knee amputee to climb stairs at different speeds, while carrying additional loads, and even backwards. By enabling direct, volitional control of powered robotic knee prostheses, the proposed EMG controller has the potential to improve amputee mobility in the real world.

Ambulation Mode Classification of Individuals with Transfemoral Amputation through A-Mode Sonomyography and Convolutional Neural Networks.

Many people struggle with mobility impairments due to lower limb amputations. To participate in society, they need to be able to walk on a wide variety of terrains, such as stairs, ramps, and level ground. Current lower limb powered prostheses require different control strategies for varying ambulation modes, and use data from mechanical sensors within the prosthesis to determine which ambulation mode the user is in. However, it can be challenging to distinguish between ambulation modes. Efforts have been made to improve classification accuracy by adding electromyography information, but this requires a large number of sensors, has a low signal-to-noise ratio, and cannot distinguish between superficial and deep muscle activations. An alternative sensing modality, A-mode ultrasound, can detect and distinguish between changes in superficial and deep muscles. It has also shown promising results in upper limb gesture classification. Despite these advantages, A-mode ultrasound has yet to be employed for lower limb activity classification. Here we show that A- mode ultrasound can classify ambulation mode with comparable, and in some cases, superior accuracy to mechanical sensing. In this study, seven transfemoral amputee subjects walked on an ambulation circuit while wearing A-mode ultrasound transducers, IMU sensors, and their passive prosthesis. The circuit consisted of sitting, standing, level-ground walking, ramp ascent, ramp descent, stair ascent, and stair descent, and a spatial-temporal convolutional network was trained to continuously classify these seven activities. Offline continuous classification with A-mode ultrasound alone was able to achieve an accuracy of 91.8±3.4%, compared with 93.8±3.0%, when using kinematic data alone. Combined kinematic and ultrasound produced 95.8±2.3% accuracy. This suggests that A-mode ultrasound provides additional useful information about the user's gait beyond what is provided by mechanical sensors, and that it may be able to improve ambulation mode classification. By incorporating these sensors into powered prostheses, users may enjoy higher reliability for their prostheses, and more seamless transitions between ambulation modes.

A lightweight robotic leg prosthesis replicating the biomechanics of the knee, ankle, and toe joint.

Robotic leg prostheses promise to improve the mobility and quality of life of millions of individuals with lower-limb amputations by imitating the biomechanics of the missing biological leg. Unfortunately, existing powered prostheses are much heavier and bigger and have shorter battery life than conventional passive prostheses, severely limiting their clinical viability and utility in the daily life of amputees. Here, we present a robotic leg prosthesis that replicates the key biomechanical functions of the biological knee, ankle, and toe in the sagittal plane while matching the weight, size, and battery life of conventional microprocessor-controlled prostheses. The powered knee joint uses a unique torque-sensitive mechanism combining the benefits of elastic actuators with that of variable transmissions. A single actuator powers the ankle and toe joints through a compliant, underactuated mechanism. Because the biological toe dissipates energy while the biological ankle injects energy into the gait cycle, this underactuated system regenerates substantial mechanical energy and replicates the key biomechanical functions of the ankle/foot complex during walking. A compact prosthesis frame encloses all mechanical and electrical components for increased robustness and efficiency. Preclinical tests with three individuals with above-knee amputation show that the proposed robotic leg prosthesis allows for common ambulation activities with close to normative kinematics and kinetics. Using an optional passive mode, users can walk on level ground indefinitely without charging the battery, which has not been shown with any other powered or microprocessor-controlled prostheses. A prosthesis with these characteristics has the potential to improve real-world mobility in individuals with above-knee amputation.

Powered knee and ankle prostheses enable natural ambulation on level ground and stairs for individuals with bilateral above-knee amputation: a case study.

Ambulation with existing prostheses is extremely difficult for individuals with bilateral above-knee amputations. Commonly prescribed prostheses are passive devices that cannot replace the biomechanical functions of the missing biological legs. As a result, most individuals with bilateral above-knee amputations can only walk for short distances, have a high risk of falling, and are unable to ascend stairs with a natural gait pattern. Powered prostheses have the potential to address this issue by replicating the movements of the biological leg. Previous studies with individuals with bilateral above-knee amputations have shown that walking with powered prostheses is possible. However, stair ambulation requires different kinematics, kinetics, and power than walking. Therefore, it is not known whether powered prostheses can restore natural ambulation on stairs for bilateral above knee individuals. Here we show a case study with an individual with bilateral above-knee amputations using a pair of lightweight powered knee and ankle prostheses for walking and stair ambulation. The kinematic analysis shows that powered prostheses can restore natural leg movements, enabling the individual to walk and climb stairs using different gait patterns, such as step-over-step or step-by-step, one step or two steps at a time. The kinetic analysis shows that the powered prostheses can restore natural ankle push-off in walking and positive knee power generation in stair ascent, which are fundamental biomechanical functions of the missing biological legs. This case study is a first step towards enhancing functional mobility and quality of life for individuals with bilateral above-knee amputations through powered knee and ankle prostheses.

Knee Exoskeleton Reduces Muscle Effort and Improves Balance During Sit-to-Stand Transitions After Stroke: A Case Study.

After a stroke, the weight-bearing asymmetry often forces stroke survivors to compensate with overuse of the unaffected side muscles to stand up. Powered exoskeletons can address this problem by assisting the affected limb during sit-tostand transitions. However, there is currently no experimental evidence demonstrating the efficacy of this intervention with the target population. This study explores controlling a powered knee exoskeleton with EMG signals to assist a stroke patient during sit-to-stand transitions. Our results show decreased peak knee torques by 6.24% and 11.9% on their unaffected and affected sides, respectively, while wearing the exoskeleton. Additionally, the peak value of the EMG signal decreased by 29.3% and 21.9%, and the integrated EMG signal value decreased by 46.7% and 36.1% on their affected vastus medialis and lateralis while wearing the exoskeleton, respectively. Finally, our results indicate improved medial-lateral balance by 61.2%, 81.6%, and 70.0% based on the degree of asymmetry (DOA), the center of pressure (COP), and the center of mass (COM), respectively. These results support the efficacy of using powered exoskeletons for high-torque tasks such as sit-to-stand transitions with stroke survivors.

Assistive Powered Hip Exoskeleton Improves Self-Selected Walking Speed in One Individual with Hemiparesis: A Case Study.

Most individuals suffering a stroke have permanent weakness on one side of the body (hemiparesis) that reduces their ability to ambulate. Autonomous powered exoskeletons have been proposed as a possible solution to this problem. Studies with healthy subjects show that assistive powered exoskeletons have the potential to improve gait, for example, by reducing the metabolic cost of walking. However, only a handful of studies have been conducted with individuals with hemiparesis. Thus, the ability of autonomous exoskeletons to improve gait in this population remains largely unknown. In this study, we assess self-selected walking speed with and without an autonomous powered hip exoskeleton in one individual with hemiparesis walking on level ground. Results show that the proposed exoskeleton improves self-selected walking speed by ~30%. The biomechanical analysis suggest that the increased walking speed is the result of the powered hip exoskeleton enabling the subject to take longer strides on the hemiparetic side. This case study provides important information to inform future exoskeleton development and clinical study design.

Powered Knee and Ankle Prosthesis with Adaptive Control Enables Climbing Stairs with Different Stair Heights, Cadences, and Gait Patterns.

Powered prostheses can enable individuals with above-knee amputations to ascend stairs step-over-step. To accomplish this task, available stair ascent controllers impose a pre-defined joint impedance behavior or follow a pre-programmed position trajectory. These control approaches have proved successful in the laboratory. However, they are not robust to changes in stair height or cadence, which is essential for real-world ambulation. Here we present an adaptive stair ascent controller that enables individuals with above-knee amputations to climb stairs of varying stair heights at their preferred cadence and with their preferred gait pattern. We found that modulating the prosthesis knee and ankle position as a function of the user's thigh in swing provides toe clearance for varying stair heights. In stance, modulating the torque-angle relationship as a function of the prosthesis knee position at foot contact provides sufficient torque assistance for climbing stairs of different heights. Furthermore, the proposed controller enables individuals to climb stairs at their preferred cadence and gait pattern, such as step-by-step, step-over-step, and two-steps. The proposed adaptive stair controller may improve the robustness of powered prostheses to environmental and human variance, enabling powered prostheses to more easily move from the lab to the real-world.

Is there a trade-off between economy and task goal variability in transfemoral amputee gait?

BACKGROUND: Energy cost minimization has been widely accepted to regulate gait. Optimization principles have been frequently used to explain how individuals adapt their gait pattern. However, there have been rare attempts to account for the role of variability in this optimization process. Motor redundancy can enable individuals to perform tasks reliably while achieving energy optimization. However, we do not know how the non-goal-equivalent and goal-equivalent variability is regulated. In this study, we investigated how unilateral transfemoral amputees regulate step and stride variability based on the task to achieve energy economy. METHODS: Nine individuals with unilateral transfemoral amputation walked on a treadmill at speeds of 0.6, 0.8, 1.0, 1.2 and 1.4 m/s using their prescribed passive prostheses. We calculated the step-to-step and stride-to-stride variability and applied goal equivalent manifold (GEM) based control to decompose goal-equivalent and non-goal-equivalent manifold. To quantify the energy economy, the energy recovery rate (R) was calculated based on potential energy and kinetic energy. Comparisons were made between GEM variabilities and commonly used standard deviation measurements. A linear regression model was used to investigate the trade-off between R and GEM variabilities. RESULTS: Our analysis shows greater variability along the goal-equivalent manifold compared to the non-goal-equivalent manifold (p < 0.001). Moreover, our analysis shows lower energy recovery rate for amputee gait compared to nonamputee gait (at least 20% less at faster walking speed). We found a negative relationship between energy recovery rate and non-goal-equivalent variability. Compared to the standard deviation measurements, the variability decomposed using GEM reflected the preferred walking speed and the limitation of the passive prosthetic device. CONCLUSION: Individuals with amputation cleverly leverage task redundancy, regulating step and stride variability to the GEM. This result suggests that task redundancy enables unilateral amputees to benefit from motor variability in terms of energy economy. The differences observed between prosthetic step and intact step support the development of prosthetic limbs capable of enhancing positive work during the double support phase and of powered prosthesis controllers that allow for variability along the task space while minimizing variability that interferes with the task goal. This study provides a different perspective on amputee gait analysis and challenges the field to think differently about the role of variability.

Design of an Underactuated Powered Ankle and Toe Prosthesis.

Powered ankle/foot prostheses aim to replicate the biomechanical function of the missing biological limb. Biomechanical analysis shows that while the ankle injects positive energy into the gait cycle, the toe joint dissipates energy. Yet virtually all powered ankle/foot prostheses use custom ankle actuators in combination with carbon fiber foot springs to imitate the function of the missing ankle/foot complex. Here we introduce a powered ankle and toe prosthesis with an underactuated mechanism. The underactuated mechanism connects the toe and ankle joints, providing biomechanically accurate torque and enabling mechanical energy recovery during gait. The proposed powered ankle/toe prothesis is the first device to match the weight, size, and build height of microprocessor-controlled prostheses.

Performance of Sonomyographic and Electromyographic Sensing for Continuous Estimation of Joint Torque During Ambulation on Multiple Terrains.

Advances in powered assistive device technology, including the ability to provide net mechanical power to multiple joints within a single device, have the potential to dramatically improve the mobility and restore independence to their users. However, these devices rely on the ability of their users to continuously control multiple powered lower-limb joints simultaneously. Success of such approaches rely on robust sensing of user intent and accurate mapping to device control parameters. Here, we compare two non-invasive sensing modalities: surface electromyography and sonomyography, (i.e., ultrasound imaging of skeletal muscle), as inputs to Gaussian process regression models trained to estimate hip, knee and ankle joint moments during varying forms of ambulation. Experiments were performed with ten non-disabled individuals instrumented with surface electromyography and sonomyography sensors while completing trials of level, incline (10°) and decline (10°) walking. Results suggest sonomyography of muscles on the anterior and posterior thigh can be used to estimate hip, knee and ankle joint moments more accurately than surface electromyography. Furthermore, these results can be achieved by training Gaussian process regression models in a task-independent manner; i.e., incorporating features of level and ramp walking within the same predictive framework. These findings support the integration of sonomyographic and electromyographic sensing within powered assistive devices to continuously control joint torque.

Robust Torque Predictions From Electromyography Across Multiple Levels of Active Exoskeleton Assistance Despite Non-linear Reorganization of Locomotor Output.

Robotic exoskeletons can assist humans with walking by providing supplemental torque in proportion to the user's joint torque. Electromyographic (EMG) control algorithms can estimate a user's joint torque directly using real-time EMG recordings from the muscles that generate the torque. However, EMG signals change as a result of supplemental torque from an exoskeleton, resulting in unreliable estimates of the user's joint torque during active exoskeleton assistance. Here, we present an EMG control framework for robotic exoskeletons that provides consistent joint torque predictions across varying levels of assistance. Experiments with three healthy human participants showed that using diverse training data (from different levels of assistance) enables robust torque predictions, and that a convolutional neural network (CNN), but not a Kalman filter (KF), can capture the non-linear transformations in EMG due to exoskeleton assistance. With diverse training, the CNN could reliably predict joint torque from EMG during zero, low, medium, and high levels of exoskeleton assistance [root mean squared error (RMSE) below 0.096 N-m/kg]. In contrast, without diverse training, RMSE of the CNN ranged from 0.106 to 0.144 N-m/kg. RMSE of the KF ranged from 0.137 to 0.182 N-m/kg without diverse training, and did not improve with diverse training. When participant time is limited, training data should emphasize the highest levels of assistance first and utilize at least 35 full gait cycles for the CNN. The results presented here constitute an important step toward adaptive and robust human augmentation via robotic exoskeletons. This work also highlights the non-linear reorganization of locomotor output when using assistive exoskeletons; significant reductions in EMG activity were observed for the soleus and gastrocnemius, and a significant increase in EMG activity was observed for the erector spinae. Control algorithms that can accommodate spatiotemporal changes in muscle activity have broad implications for exoskeleton-based assistance and rehabilitation following neuromuscular injury.