ON-LINE TRAJECTORY GENERATION CONSIDERING KINEMATIC MOTION CONSTRAINTS FOR ROBOT MANIPULATORS
Résumé
This paper aims to propose an on-line trajectory generation algorithm that is able to address not only constant but also time-variant kinematic motion constraints for multi-DOFs robot manipulators. By using a concatenation of cubic polynomials, the proposed method can provide a smooth trajectory that is synchronized and bounded in the robot kinematic motion constraints which are expressed as upper bounds on the absolute values of velocity, acceleration, and jerk. An additional decision tree will select intermediate motion profiles when the motion constraints are abruptly changed. Due to direct computation without optimization computation or randomized algorithms, the proposed solution requires only a short execution time. Simulations and experiments were conducted to verify the feasibility and effectiveness of this algorithm in smooth trajectory generation from arbitrary states of motion. With the proposed approach, robot motion can be limited by the kinematic motion constraints which will reduce manipulator wear and improve tracking accuracy and speed. The proposed algorithm can be used in real time due to the low computational complexity.