中国机械工程 ›› 2025, Vol. 36 ›› Issue (9): 2047-2056.DOI: 10.3969/j.issn.1004-132X.2025.09.017

• 智能制造 • 上一篇    

基于改进樽海鞘群算法的机械臂多目标轨迹规划研究

刘建林1(), 黄海松1,2(), 范青松1,3, 马驰1, 张浪浪1   

  1. 1.贵州大学现代制造技术教育部重点实验室, 贵阳, 550025
    2.贵州省装备制造数字化车间建模与仿真工程研究中心, 贵阳, 550025
    3.华中科技大学机械科学与工程学院, 武汉, 430074
  • 收稿日期:2024-07-25 出版日期:2025-09-25 发布日期:2025-10-15
  • 通讯作者: 黄海松
  • 作者简介:刘建林,男,1999年生,硕士研究生。研究方向为机械臂控制。E-mail:ljl2685177925@163.com
    黄海松*(通信作者),女,1977年生,教授。研究方向为工业机器人、智能制造等。E-mail:hshuang@gzu.edu.cn
  • 基金资助:
    国家自然科学基金(52165063);贵州省科技计划(黔科合支撑[2022],黔科合支撑[2023],黔科合支撑[2024],黔科合平台人才-CXTD[2023],黔科合平台人才-GCC[2022]);贵阳市科技计划(筑科合同[2023]13-11号)

Multi-objective Trajectory Planning of Manipulators Based on Improved SSA

Jianlin LIU1(), Haisong HUANG1,2(), Qingsong FAN1,3, Chi MA1, Langlang ZHANG1   

  1. 1.Key Laboratory of Advanced Manufacturing Technology,Ministry of Education,Guizhou University,Guiyang,550025
    2.Guizhou Province Equipment Manufacturing Digital Workshop Modeling and Simulation Engineering Research Center,Guiyang,550025
    3.School of Mechanical Science and Engineering,Huazhong University of Science and Technology,Wuhan,430074
  • Received:2024-07-25 Online:2025-09-25 Published:2025-10-15
  • Contact: Haisong HUANG

摘要:

提出了一种基于改进樽海鞘群算法(SSA)的机械臂多目标轨迹规划模型,以同时优化效率、能耗和冲击三个目标。利用人工势场法(APF)进行路径规划,得到机械臂抓取物料的最短、无碰撞路径,并提取关键运动序列,建立多目标函数。针对多目标樽海鞘群算法(MSSA)的初始种群多样性差、容易陷入局部最优以及在解集空间中收敛缓慢等问题,提出了一种改进的多目标樽海鞘群算法(LMSSA)。该算法结合logistic-sine混沌映射、小孔成像学习策略和黄金正弦开发策略来优化七阶B样条曲线的控制节点从而完成机械臂的多目标运动轨迹规划。搭建MATLAB-CoppeliaSim-UR16e实验平台,将轨迹规划模型应用于机械臂UR16e的实际抓取任务。实验结果表明,基于LMSSA算法的机械臂运动规划方法实现了机械臂准确、高效且节能的运动轨迹规划,并成功应用于实际操作场景中。

关键词: 轨迹规划, 多目标优化, 机械臂, 樽海鞘群算法

Abstract:

To optimize the three objectives of efficiency, energy consumption and impacts at the same time, a multi-objective trajectory planning model was proposed based on an improved SSA. Firstly, the artificial potential field method (APF) was used for path planning to obtain the shortest and collision-free path of the manipulator grasping the materials, and the key motion sequence was extracted to establish a multi-objective function. Then, aiming at the problems of multi-objective salp swarm algorithm (MSSA), such as poor diversity of initial population, easy to fall into local optimum and slow convergence in solution set space, an improved algorithm namely logistic-sine multi-objective salp swarm algorithm(LMSSA)was proposed. The algorithm combined logistic-sine chaotic mapping, pinhole imaging learning strategy and golden sine development strategy to optimize the control nodes of the seventh-order B-spline curve and complete the multi-objective motion trajectory planning of the robotic arms. Finally, the trajectory planning model was applied to the actual grasping tasks of the manipulator UR16e by building MATLAB-CoppeliaSim-UR16e experimental platform. Experimental results show that based on LMSSA, the manipulator motion planning method realizes the accurate, efficient and energy-saving motion trajectory planning of the manipulator, and is successfully applied to the actual operation scenes.

Key words: trajectory planning, multi-objective optimization, manipulator, salp swarm algorithm(SSA)

中图分类号: