Inverse kinematic solution based on Lyapunov function for redundant and non-redundant robots

In this paper, we propose an iterative method using an optimization scheme (neural optimization network) to solve the inverse kinematic problem for redundant and nonredundant manipulators. The neural network is adapted in the direction of decreasing a Lyapunov function to move the end-effector to the desired position. This approach offers substantially better accuracy and avoids the computation of the Jacobian inverse and pseudoinverse Jacobian matrix. The application of this scheme to a 3 degrees of freedom redundant manipulator is demonstrated through simulation results