摘要
针对机械臂运行时间最短,消耗能量最少,脉动最小等多目标下的轨迹避障规划需求,提出了一种基于改进MOABC算法的机械臂多目标轨迹避障规划方法。首先,建立分段多项式轨迹的描述方程,将方程中不能确定的中间点相关参数作为待定参数,通过优化参数改变轨迹形状,使机械臂避开障碍物的同时优化轨迹的多个性能指标;其次,综合碰撞检测,运动学约束条件,建立时间-能量-脉动优化目标函数,应用改进的多目标人工蜂群算法(MOABC)优化参数,规划出满足条件的Pareto解集,从而得到可供选择的兼备精英性和多样性的理想轨迹;最后,通过MATLAB进行仿真并在实物上进行实验验证。结果表明该方法能有效地规划出满足机械臂性能需求的无碰撞运动轨迹。
Aiming at the requirements of multi-objective trajectory obstacle avoidance planning with the shortest running time,the least energy consumption and the least pulsation of the manipulator,a multi-objective trajectory obstacle avoidance planning method of the manipulator based on the improved MOABC algorithm is proposed.First,the description equation of piecewise polynomial trajectory is established,and the uncertain intermediate point related parameters in the equation are taken as undetermined parameters.By changing the trajectory shape through optimization parameters,the robot arm can avoid obstacles and optimize multiple performance indicators of the trajectory.Secondly,the collision detection and kinematics constraints are integrated,the time energy pulsation optimization objective function is established,the parameters are optimized by the improved multi-objective artificial bee colony algorithm(MOABC),and the Pareto solution set satisfying the conditions is planned,so as to obtain an ideal trajectory that can be selected and has both elitism and diversity.Finally,the simulation is carried out by MATLAB and the experimental verification is carried out on the real object.The results show that the method can effectively plan the collision free motion trajectory that meets the performance requirements of the manipulator.
作者
余自淳
邹艳丽
谭宇飞
许江杰
YU Zichun;ZOU Yanli;TAN Yufei;XU Jiangjie(School of Electronic and Information Engineering,Guangxi Normal University,Guilin 541004,China)
出处
《组合机床与自动化加工技术》
北大核心
2023年第6期63-67,共5页
Modular Machine Tool & Automatic Manufacturing Technique
基金
广西创新驱动发展专项资助项目(桂科AA21077015)。
关键词
机械臂
避障
轨迹规划
MOABC
多目标优化
manipulator
obstacle avoidance
trajectory planning
MOABC
multi-objective optimization
