首页|期刊导航|四川轻化工大学学报(自然科学版)|基于群体智能算法的机械臂轨迹规划

基于群体智能算法的机械臂轨迹规划OA

Trajectory Planning of Robotic Arms Based on Swarm Intelligence Algorithms

中文摘要英文摘要

为了解决鲸鱼优化算法(WOA)易困入局部最优和搜索速度较慢等问题,以时间最优为目标,提出一种基于种群迁移策略的鲸鱼粒子群混合优化算法(PSO-WOA-POS).首先,该算法通过在WOA的开发阶段引入粒子群优化(PSO)算法,利用群体的全局信息来提升搜索效率和精度;同时,将WOA中的包围与捕食策略融入PSO,通过动态路径调整和随机扰动,避免陷入局部最优,增强了解空间的多样性.此外,通过引入指数衰减函数在早期阶段增强探索性,扩大搜索范围;在后期阶段减少惯性,增强对最优解的聚焦,从而平衡全局搜索和局部搜索,提升优化效果.在Matlab中用6种测试函数进行模拟实验,验证了混合算法在基本性能和全局搜索能力方面的优势.通过在IRB 2600机器人上对3种算法进行实验,结果表明PSO-WOA-POS所需时间最短,且仿真与实际时间误差控制在5%以内,验证了算法的有效性和硬件适应性.

To address the limitations of the Whale Optimization Algorithm(WOA),such as its proneness to local optima and slow search speed,this paper proposes a hybrid optimization algorithm named Particle Swarm Optimization-Whale Optimization Algorithm-Population Update Strategy(PSO-WOA-POS)with the minimization of computation time as the core objective is proposed in the present paper.The hybrid algorithm embeds Particle Swarm Optimization(PSO)into the exploitation phase of WOA.It utilizes the global information sharing mechanism of particle swarms to improve search efficiency and optimization accuracy.Meanwhile,the inherent encircling and predation strategies of WOA are introduced into the iterative process of PSO.With dynamic path adjustment and random perturbation operations,the improved algorithm avoids premature convergence to local optima and maintains high diversity of the solution space.In addition,an exponential decay function is incorporated into the algorithm's update framework.This mechanism strengthens global exploration in early iterations to expand the search range and reduces the inertia weight in later stages for intensive exploitation around the optimal regions.It realizes a delicate balance between global exploration and local exploitation and further improves the overall optimization performance.Comprehensive simulation experiments based on six benchmark functions are carried out in the Matlab environment,which verify the outstanding performance and global search capacity of the proposed hybrid algorithm.Comparative hardware verification experiments are conducted on an IRB 2600 industrial robot with three alternative algorithms.The results show that PSO-WOA-POS achieves the minimum running time;the deviation between simulated and actual runtime is controlled within 5%.These results fully confirm the effectiveness of the algorithm and its strong adaptability to physical hardware platforms.

夏重洋;王鹏彧;李欣宇

安徽理工大学 人工智能学院,安徽 淮南 232001安徽理工大学 机电工程学院,安徽 淮南 232001安徽理工大学 人工智能学院,安徽 淮南 232001

信息技术与安全科学

指数衰减惯性权重机械臂轨迹规划粒子群优化算法鲸鱼优化算法基准函数

exponential decay inertia weighttrajectory planning of robot armParticle Swarm Optimization algorithmWhale Optimization Algorithmreference function

《四川轻化工大学学报(自然科学版)》 2026 (4)

74-85,12

安徽省科技专项重大项目(202203a05020049)

10.11863/j.suse.2026.04.07

评论