Optimal Trajectory Planning for A Manipulator Based on Multi-Objective Moth Swarm Algorithm

We propose an optimal trajectory planning method based on the improved multi-objective moth swarm algorithm (IMOMSA) to enhance the efficiency and stability of a hedge-trimming manipulator, with running time and joint impact as the optimization objectives. Based on the structure and operability of the robotic arm, we select and interpolate path points for fitting. Under kinematic constraints, IMOMSA performs time-impact multi-objective optimization, generating a Pareto optimal solution set, from which the optimal solution is selected using fuzzy membership functions. This algorithm integrates Pareto archiving and crowding distance mechanisms to extend single-objective algorithms while incorporating chaotic initialization and opposition-based learning (OBL) improvement strategies. We conduct two case studies to compare IMOMSA with classical multi-objective optimization algorithms. The diversity and convergence of the algorithm are evaluated using the ZDT test function set, and its effectiveness is validated in a practical trajectory planning problem. Simulation results show that the optimized running time is reduced by 11.8%, and the joint impact index decreases by 5.98% to 39.96%. The application of IMOMSA in trajectory planning substantially enhances the efficiency and stability of the hedge-trimming manipulator.