A Neural PID End-effector Controller for Robotic Manipulator with Joint Position Constraints

We present a novel efficient method to control the end- effector of robotic manipulators without solving the inverse kinematics problem in the presence of the hard bound on the joint space motion. The basic Proportional–Integral-Derivative (PID) controller is combined with a nonlinear law to track the position and control the joint move within physical limits. To provide the adaptive ability of such the controller, a neural control signal is integrated to compensate for the internal and external disturbances in the robot dynamic. Stability of the closed-loop system is guaranteed by Lyapunov constraints. The proposed controller is validated in intensive simulation results.