Task-Based Knee Rehabilitation With Assist-as-Needed Control Strategy and Recovery Tracking System

This research aims to design and implement a novel task-based knee rehabilitation strategy through kinematic synthesis, assist-as-needed control strategy, and recovery tracking system. Experimental kinematic data collected through motion capture system are utilized as an input to the mechanism synthesis procedure. Parallel mechanisms with single degree-of-freedom are considered to generate the complex three-dimensional (3D) motions of the lower leg. An exact workspace synthesis approach is utilized, in which the implicit description of the workspace is made to be a function of the structural parameters of the serial chains of the parallel mechanism, making it easy to relate those parameters to the desired trajectory from the motion capture. The synthesis procedure resulted an exoskeleton which can guide the complex motion of the human knee without the need of mimicking the joint by the exoskeleton counterpart. This can potentially reduce the improper alignment problems arising due to the constantly varying axis of rotation of human joint, which is often very difficult to predict. An assist-as-needed control and recovery tracking strategy is outlined based on an electromyography (EMG) signals and force sensing resistors (FSRs) mounted on the user and exoskeleton, respectively. The EMG signal is captured from the user leg and FSRs are applied at the attachment area of the exoskeleton and the leg, this helps to get the amount of force applied by the exoskeleton to the leg as well as for the recovery tracking. The assist-as-needed controller eliminates the need of constant supervision, and hence saves time and reduces cost of the rehabilitation process. Similarly, the real-time progress tracking system will motivate and actively engage users