Research on singular robustness algorithm of robot inverse kinematics based on dynamic damping coefficient

In order to solve the singularity problem of the Jacobian matrix in the robotic inverse kinematics method, a new singular processing algorithm is proposed. Based on the damped least-square (DLS) method, the damping coefficients, by using the singular value decomposition (SVD) of the Jacobian matrix, are optimized from the fixed value of the traditional algorithm to the dynamic value adjusted with the Jacobian matrix. It is ensured that the angular velocity curve of each joint is continuous and smooth when the robot passes through the singularity, and the robustness of the inverse solution algorithm is enhanced. The algorithm is simulated on the Kawasaki BA006N robot, results show that the robot does not introduce the tracking error at the non-singular position. At the singular position, the joint motion of the robot is smooth and the singular robustness is better than that of the existing algorithm.