Design of a Biomimetic Control System for Tendon-driven Prosthetic Hand

Prosthetic hands are frequently rejected due to frustrations when encountering disturbances or manipulating deformable objects. Humans developed a unique ability of compliant control through a complex neuro-musculo-skeletal system. We present the design and implementation of a biomimetic control system, which aims for duplicating physiologically realistic behavior of human motor control in a tendon-driven prosthetic hand. The system receives surface EMG as the source of command, which activates a simple muscle model to compute for the expected tension on the tendon. A torque motor establishes the expected tension with closed-loop adjustment from a neural circuitry simulated in software. An able-bodied subject and two fore-arm amputees participated in the study. We validated their abilities of force loading with the index finger, and power grasping with the tendon-drive prosthetic hand. Results showed the participants could establish expected force using the biomimetic control system. They were also able to grasp/release sample objects with various stiffnesses, sizes, and weights. This is an important step toward a full-fledged biomimetic control for prosthetic hand with hardware-in-loop.