Research on the Recognition Method of Exoskeleton Rehabilitation Robot Based on Patient Intent Recognition

Abstract

With the development of exoskeleton rehabilitation robot technology, the use of such robots can assist patients in achieving basic motor functions such as normal walking, improve the physical condition of patients, and gradually restore the motor functions of the affected limbs. These robot systems are human-centered human-computer collaborative systems, whose primary task is to understand human movement intentions and move in coordination with the human body. The current research methods of human-computer interaction perception systems mainly involve bioelectrical signals and physical signals such as force/torque. This paper reviews the recognition methods of such robots and explores their intention recognition methods in depth. Focusing on the intention recognition technology of exoskeleton rehabilitation robots, this paper analyzes the principles, processes, advantages and disadvantages, and applicable scenarios of the Surface Electromyography (sEMG) and Inertial Measurement Unit (IMU) methods, providing theoretical references for optimized design and clinical applications. By evaluating the rehabilitation effects of each intention recognition method, this paper explains and evaluates their recognition principles, offering certain references for the optimized design and clinical practice of rehabilitation robots.