ADAPTIVE CONTROL METHOD AND SYSTEM FOR UPPER LIMB REHABILITATION ROBOT AND BASED ON GAME THEORY AND SEMG

Disclosed in the present invention are an adaptive control method and system for an upper limb rehabilitation robot and based on game theory and sEMG. The method comprises: firstly, designing a motion trajectory of a robot that is controlled within a training time during an operation process of a subject; on the basis of an sEMG-based back propagation neural network muscle force estimation model, establishing a nonlinear dynamic relationship between a surface electromyogram signal and a terminal force by means of constructing a three-layer neural network; analyzing a human-machine interaction system by using the principle of game theory, deducing the role of the robot, updating a control rate between the robot and the subject by using the Nash equilibrium, and determining adaptive weight factors of the robot and the subject; and finally, enabling the robot to adaptively adjust a training mode thereof according to a motion intention of the subject during the operation process and weight coefficients, which are