A Hybrid Optimal Control Strategy for a Smart Prosthetic Hand

This paper presents a hybrid of a soft computing or control technique of adaptive neuro-fuzzy inference system (AN-FIS) and a hard computing or control technique of the hybrid finite-time linear quadratic optimal control for a two-fingered (thumb and index) prosthetic hand. In particular, the ANFIS is used for inverse kinematics, and the optimal control is used to minimize tracking error utilizing feedback linearized dynamics. The simulations of this hybrid controller, when compared with the proportional-integral-derivative (PID) controller showed enhanced performance. Work is underway to extend this methodology to a five-fingered, three-dimensional prosthetic hand.