关键词: 工业机器人, 轨迹规划, 多目标优化, 粒子群算法

Abstract: A comprehensive optimal trajectory planning method was proposed to improve the efficiency and impact resistance of robotic arms. Firstly, by establishing a 3-5-3 polynomial curve mathematical model, a joint motion trajectory with controllable endpoint motion parameters was constructed. Secondly, considering constraints such as joint position, velocity, and acceleration, the objective function was defined using the weighted coefficient method to achieve a comprehensive optimization of the action time, impact, and dexterity of the robotic arm. The dynamic weighting method was used in the design of the objective function to address the contradiction between joint velocity and impact. Finally, for the standard particle swarm optimization algorithm, the Latin hypercube sampling function was used to homogenize the population, and a random inertia weight update strategy was proposed to obtain an improved particle swarm algorithm. This algorithm was used to optimize the objective function and obtain the comprehensive optimal motion trajectory. Simulation and prototype experiments were conducted, and the experimental results demonstrate the feasibility of the proposed method.

Key words: industrial robot, trajectory planning, multi objective optimization, particle swarm optimization(PSO)