Time-jerk optimal trajectory planning for robotic manipulators

Abstract

This paper proposes a time and jerk optimal trajectory planning method to find an efficient and sufficiently smooth trajectory. The objective function of the proposed method consists of two parts: the execution time and the smoothness of the trajectory in the joint space. The latter part enables one to get a smoother trajectory and reduce the tracking error. Moreover, the kinematic constraints of the robot are also taken into consideration by setting boundary on the absolute value of kinematical constraints. A newly devised cubic spline is presented to ensure the starting and ending values of the velocity and acceleration controllable without requiring extra points. The method is tested on an EAST Tokamak inspection robot.