A robotic arm design for stroke patients

Abstract

This paper presents a robotic arm for stroke patents. Two SEMG (Surface Electromyography) signals collected from bicipital muscle and triceps muscle of arm, which are used to control robotic arm. When patients want to flexile arm, the SEMG signal of bicipital muscle is larger than that of triceps muscle. On the other hand, when patient want to extend arm, the SEMG signal of triceps muscle is larger than that of bicipital muscle. The robotic arm's rotation direction is decided by the difference of two SEMG signals. The torque and speed of the robotic arm are controlled by the amplitudes of SEMG signals. The system is based DSP (Digital Signal Processor). H-bridge is used to drive DC motor. The experimental results show that the robotic arm can act correctly according to SEMG signals of human arm.