改进粒子群算法的自动充电机械臂时间最优轨迹研究OA北大核心CSTPCD
Exploring Time-optimal Trajectory of Automatic Charging Manipulator with Improved Particle Swarm Optimization Algorithm
针对桁架充电机械臂关节空间轨迹规划的时间优化问题,提出了一种非线性动态学习因子的粒子群算法.通过运动学分析获取工作空间,引入 3-5-3 多项式插值进行轨迹规划.结合运动过程中的速度与加速度约束,寻求运动过程中的最短时间.对比改进粒子群算法和基本粒子群算法的收敛速度,分析各关节优化前后运动时间的变化情况,并进行仿真实验验证.结果表明:改进粒子群算法的收敛性能较基本粒子群算法更快,整体运动时间缩短约33%,证实改进粒子群算法的可行性.
A particle swarm optimization(PSO)algorithm based on the nonlinear dynamic learning factor was proposed to solve the time optimization problem in the joint space trajectory planning of a truss charging manipulator.The workspace was obtained through kinematic analysis,and the 3-5-3 polynomial interpolation was introduced for the trajectory planning.The shortest motion time was sought through combining velocity constraints with acceleration constraints.The convergence speed of the improved PSO algorithm was compared with that of the basic PSO algorithm,and the variation of motion time of each joint before and after optimization was analyzed.The simulation results show that the convergence performance of the improved PSO algorithm is faster than that of the basic PSO algorithm and that the overall motion time is shortened by about 33%,confirming the feasibility of the improved PSO algorithm.
朱浩;赵清海;郑群锋;宁长久
青岛大学机电工程学院,山东青岛 266071青岛大学机电工程学院,山东青岛 266071||青岛大学电动汽车智能化动力集成技术国家地方联合工程研究中心,山东青岛 266071北京理工大学机械与车辆学院,北京 100081
机械工程
桁架充电机械臂时间优化非线性动态学习因子粒子群算法
truss charging manipulatortime optimizationnonlinear dynamic learning factorparticle swarm optimization algorithm
《机械科学与技术》 2024 (003)
423-429 / 7
国家自然科学基金项目(52175236)
评论