Long and short delay feedback on one-link nonlinear forearm with coactivation

Control strategies for a one-link model of the human forearm system are presented. Three attributes of the human forearm are implemented in the second order nonlinear model: neural transmission delays in the feedback paths, nonlinear behavior of the spindle reflex, and stiffness regulation through coactivation. Three feedback loops are present in the model: intrinsic feedback (undelayed) from the actuators, spindle feedback (delayed), and higher level controller feedback (delayed). The stability of the model is examined through computer simulation analysis. A method of speed control of the arm is presented that utilizes continuous transitions in the feedforward activation levels of the muscles. Two control strategies are utilized by the higher level controller: a proportional plus integral (PI) compensation strategy, and a fuzzy control strategy. Both strategies can endure long loop transmission delays without causing the system to become unstable. A comparison of the settling time of these two controllers in compensating for disturbances and loading errors is presented.<
>