These results indicate that translational motion tracking can be used to substantially enhance walking task prediction in leg prostheses without adding external sensing modalities. Our proposed algorithm can thus be used as a part of a task-adaptive and fully integrated prosthesis controller.

No abstract

The TF8 actuator is an untethered, lower-extremity powered-prostheses designed to replicate biological kinetic and kinematic function of ankles. An energy optimal hardware specification was found by kinematically clamping walking gait data to the dynamic model of a series elastic actuator (SEA). We searched for a minimal electrical energy configuration of motor, reduction ratio, and spring, subject to specified constraints and ultimately discretely available components. The outcome translated into a mechanical design that heavily weighted the importance of mechanical energy storage in springs. The resulting design is a moment-coupled cantilever-beam reaction-force SEA (RFSEA) that has a nominal torque rating of 85Nm, peak torque of 175Nm, 105 degree range of motion, and a hardware mass of 1.6kg.

We present an actuator designed for untethered, lower-extremity powered-prostheses that replicates biological kinetic and kinematic function of both human knees and ankles. An electric energy optimal hardware specification is defined by kinematically clamping walking gait data to the dynamic model of a series elastic actuator (SEA) and searching for motor, reduction ratio, and spring. The actuator is shown to achieve the required torque, angle, and velocity requirements for nominal walking conditions on level ground as well as varied terrain. The performance of the actuator is demonstrated on benchtop and as worn by a human subject with unilateral below knee amputation. The resulting design is a moment-coupled cantileverbeam reaction-force SEA (MC-RFSEA) that has a nominal torque rating of 85Nm, repeated peak torque of 175Nm, 105 o range of motion, and a hardware mass of 1.6kg. Preliminary results from level-ground walking with the actuator tested in an ankle configuration show an electric cost of transport of 0.053J/kg when walking at 1.5m/s.

This study describes the development and offline validation of a heuristic algorithm for accurate prediction of ground terrain in a lower limb prosthesis. This method is based on inference of the ground terrain geometry using estimation of prosthetic limb kinematics during gait with a single integrated inertial measurement unit. Methods: We asked five subjects with below-knee amputations to traverse level ground, stairs, and ramps using a high-range-of-motion powered prosthesis while internal sensor data were remotely logged. We used these data to develop two terrain prediction algorithms. The first employed a state-of-the-art machine learning approach, while the second was a directly tuned heuristic using thresholds on estimated prosthetic ankle joint translations and ground slope. We compared the performance of these algorithms using resubstitution error for the machine learning algorithm and overall error for the heuristic algorithm. Results: Our optimal machine learning algorithm attained a resubstitution error of 3.4% using 45 features, while our heuristic method attained an overall prediction error of 2.8% using only 5 features derived from estimation of ground slope and horizontal and vertical ankle joint displacement. Compared with pattern recognition, the heuristic performed better on each individual subject, and across both level and non-level strides. Conclusion and significance: These results demonstrate a method for heuristic prediction of ground terrain in a powered prosthesis. The method is more accurate, more interpretable, and less computationally expensive than state-of-the-art machine learning methods, and relies only on integrated prosthesis sensors. Finally, the method provides intuitively tunable thresholds to improve performance for specific walking conditions.