Polynomial-based smooth trajectory planning for fruit-picking robot manipulator

Abstract

Fruit-picking robot is an important form of intelligent agricultural machinery and has great impact in smart agriculture. Motion planning for its multi-Degree of Freedoms (DOFs) manipulator is significant to picking efficiency and fruit quality. However, it is difficult to plan appropriate trajectory for such a complex system. The purpose of this article is to generate smooth trajectories for fruit-picking robot manipulators using shortcuts that are constrained in velocity, acceleration and jerk. The proposed algorithm smooths the robot motion as a post process to trajectory planning. Given an input jerky path, the algorithm will first convert it into a fully synchronized trajectory, and then attempt to reduce the execution time of this trajectory as much as possible while retaining the kinematic motion constraints. The idea of the proposed method is to build shorter and collision-free shortcuts that replace the intermediate motion between two randomly-picked points on the trajectory. Experiments carried on a mobile fruit-picking robot demonstrate that this technique is able to generate smooth and collision-free trajectories for robot manipulators.