A magnetorheological knee prosthesis is presented that automatically adapts knee damping to the gait of the amputee using only local sensing of knee force, torque, and position. To assess the clinical effects of the user‐adaptive knee prosthesis, kinematic gait data were collected on four unilateral trans‐femoral amputees. Using the user‐adaptive knee and a conventional, non‐adaptive knee, gait kinematics were evaluated on both affected and unaffected sides. Results were compared to the kinematics of 12 age, weight and height matched normals. We find that the user‐adaptive knee successfully controls early stance damping, enabling amputee to undergo biologically‐realistic, early stance knee flexion. These results indicate that a user‐adaptive control scheme and local mechanical sensing are all that is required for amputees to walk with an increased level of biological realism compared to mechanically passive prosthetic systems.

User‐adaptive control of a magnetorheological prosthetic knee

Control systems and methods are disclosed for processing a time series of signals associated with the movement of a device associated with a limb. The time series of motion signals is filtered, such as thorough an autoregressive filter, and compared to stored data sets representing a limb-motion event and/or phase. In certain examples, a plurality of accelerometers generate the time series of motion signals based at least on acceleration measurements in three orthogonal directions and/or planes. The acceleration measurements may relate to the movement of an artificial limb, such as a prosthetic or orthotic device. Upon determining an event and/or phase of limb motion, the control system may trigger an actuator to appropriately adjust one or more prosthetic or orthotic joints.

Systems and methods for processing limb motion

Automatic prediction of gait events (e,g,, heel contact, flat foot, initiation of the swing, etc.) and corresponding profiles of the activations of muscles is important for real-time control of locomotion. This paper presents three supervised machine learning (ML) techniques for prediction of the activation patterns of muscles and sensory data, based on the history of sensory data, for walking assisted by a functional electrical stimulation (FES). Those ML's are: (1) a multilayer perceptron with Levenberg-Marquardt modification of backpropagation learning algorithm; (2) an adaptive-network-based fuzzy inference system (ANFIS); and (3) a combination of an entropy minimization type of inductive learning (IL) technique and a radial basis function (RBF) type of artificial neural network with orthogonal least squares learning algorithm. Here we show the prediction of the activation of the knee flexor muscles and the knee joint angle for seven consecutive strides based on the history of the knee joint angle and the ground reaction forces. The data used for training and testing of ML's was obtained from a simulation of walking assisted with an FES system. The ability of generating rules for an FES controller was selected as the most important criterion when comparing the ML's. Other criteria such as generalization of results, computational complexity, and learning rate mere also considered. The minimal number of rules and the most explicit and comprehensible rules were obtained by ANFIS. The best generalization was obtained by the IL and RBF network.

/pdf/three-machine-learning-techniques-for-automatic-g4i5v7hn0v.pdf

Three machine learning techniques for automatic determination of rules to control locomotion

An Active Ankle Foot Orthosis (AAFO) is provided where the impedance of an orthotic joint is modulated throughout the walking cycle to treat ankle foot gait pathology, such as drop foot gait. During controlled plantar flexion, a biomimetic torsional spring control is applied where orthotic joint stiffness is actively adjusted to minimize forefoot collisions with the ground. Throughout late stance, joint impedance is minimized so as not to impede powered plantar flexion movements, and during the swing phase, a torsional spring-damper (PD) control lifts the foot to provide toe clearance. To assess the clinical effects of variable-impedance control, kinetic and kinematic gait data were collected on two drop foot participants wearing the AAFO. It has been found that actively adjusting joint impedance reduces the occurrence of slap foot, allows greater powered plantar flexion, and provides for less kinematic difference during swing when compared to normals.

Active ankle foot orthosis

The present invention discloses an instrumented prosthetic foot for use with an actuated leg prosthesis controlled by a controller, the instrumented prosthetic foot comprising a connector to connect the instrumented prosthetic foot to the leg prosthesis, an ankle structure connected to the connector, a ground engaging member connected to the ankle, at least one sensor for detecting changes in weight distribution along the foot, and an interface for transmitting signals from the sensor to the controller.

G. Ishai

Papers

Adaptive microcomputer control of an artificial knee in level walking.

Evaluation of ak prostheses comparing conventional with adaptive knee control devices