Obstacle avoidance and joint limit avoidance motion planning method for humanoid upper limb robot

Abstract

The application discloses a planning method for obstacle avoidance and joint limit avoidance of a humanoid upper limb robot, and belongs to the technical field of robots, and comprises the following steps: a mapping relationship between the end position and posture of the humanoid upper limb robot and the angle of each joint is established to obtain a velocity layer forward kinematics model; an estimation error of a pseudo-inverse matrix corresponding to a Jacobian matrix and an estimation error of a null space basis matrix are introduced; a hybrid error model is established based on the estimation errors; a recurrent neural network is constructed based on the hybrid error model; and the joint angle trajectory of the humanoid upper limb robot is calculated through a control law according to the matrix value obtained by the recurrent neural network. The application simultaneously considers the joint limit and obstacle constraints, and can improve the tracking accuracy and convergence speed of the robot.

Images