ROBOT EXTERNAL CONTACT FORCE ESTIMATION METHOD BASED ON ARTIFICIAL NEURAL NETWORK

A robot external contact force estimation method based on an artificial neural network, comprising performing modification processing on a robot dynamic model, and dividing the robot dynamic model into a linear model part and a nonlinear model part; performing approximation on the nonlinear part by using a radial basis function neural network, and constructing a radial basis function neural network-based adaptive observer to estimate the nonlinear model part and the unknown interference; feeding back the estimated nonlinear model part and unknown interference to a system model, and compensating for the corresponding part to obtain a linearized robot dynamic model; and when the external application force acts on a robot, according to the linearized robot dynamic model, constructing a subjective observer to estimate the external application force. According to the robot external contact force estimation method based on an artificial neural network, various interference influences are effectively suppressed, and