Optimal Trajectory Planning Method for Handling Robots Based on Multi-objective Particle Swarm Optimization Guided by Evolutionary Information

Abstract

This paper addresses the trajectory optimization and reliability challenges of 6-DOF handling robots by proposing a multi-objective particle swarm optimization method guided by evolutionary information (EIGMOPSO). The method optimizes trajectory planning in terms of time, energy consumption, and smoothness to enhance operational reliability and mechanical durability. To overcome the limitations of traditional MOPSO, a regionally dynamic stratification strategy based on evolutionary capability assessment is proposed, classifying the population into regions by evaluating fitness, diversity, and stability. A layered optimization mechanism dynamically adjusts exploration and exploitation processes, improving global search capability. Additionally, a dynamic two-stage archive maintenance strategy ensures high-quality solutions. Experimental results demonstrate that EIGMOPSO significantly improves operational efficiency, reduces mechanical wear and energy consumption, and enhances system maintainability, making it well-suited for handling robots in industrial environments.