改进遗传算法搜索中间点的机械臂轨迹规划
|
摘要
为了实现机械臂轨迹的多目标优化,提出了插入最优中间点的轨迹规划方法,旨在实现避障的同时保证轨迹最短且关节角变化量最小。介绍了仿人手臂机械臂工作原理,建立了机械臂运动学模型;提出了插入最优中间点的轨迹规划方法,明确了碰撞检测策略;对于最优中间点的选取,在传统遗传算法基础上,提出了均匀方式的初始种群生成方法,使用种群分布熵和适应度值方差构建了交叉概率,依据个体适应度值与种群平均值构建了变异概率,从而提出了基于改进遗传算法的最优中间点寻优方法。经仿真验证,插入中间点数量与障碍物数量相同时,才能够兼顾计算量和轨迹优化程度;与传统遗传算法规划的轨迹相比,改进遗传算法规划的轨迹长度较少了12. 1%,关节角变化量减少了24. 4%,适应度值增加了23. 6%。
In order to realize multi-objective optimization of robotic arm trajectory,trajectory planning method by inserting optimal intermediate point is proposed,aimed at avoiding obstacles,lessening trajectory length and minimizing joint angle variation. Working principle of humanoid robotic arm is introduced,and dynamic model of humanoid robotic arm is built. Robotic arm trajectory planning method based on inserting optimal is put forward,and collision detection method is given. For the chosen of optimal intermediate point,on basis of traditional genetic algorithm,uniform way is used to generate original population. Population distribution entropy and variance of fitness value are used to improve crossover probability,and variation probability is optimized by individual fitness value,so that intermediate point optimizing method based on improved genetic algorithm is raised. It is clarified by simulation that balancing copulation and trajectory optimizing level,quantity of intermediate point should be the same with quantity of obstacles. Compared with the trajectory planned by traditional genetic algorithm,the length of trajectory planned by improved algorithm decreases by 12. 1%,joint angle variation decreases by24. 4%,and fitness value decreases by 23. 6%.
In order to realize multi-objective optimization of robotic arm trajectory,trajectory planning method by inserting optimal intermediate point is proposed,aimed at avoiding obstacles,lessening trajectory length and minimizing joint angle variation. Working principle of humanoid robotic arm is introduced,and dynamic model of humanoid robotic arm is built. Robotic arm trajectory planning method based on inserting optimal is put forward,and collision detection method is given. For the chosen of optimal intermediate point,on basis of traditional genetic algorithm,uniform way is used to generate original population. Population distribution entropy and variance of fitness value are used to improve crossover probability,and variation probability is optimized by individual fitness value,so that intermediate point optimizing method based on improved genetic algorithm is raised. It is clarified by simulation that balancing copulation and trajectory optimizing level,quantity of intermediate point should be the same with quantity of obstacles. Compared with the trajectory planned by traditional genetic algorithm,the length of trajectory planned by improved algorithm decreases by 12. 1%,joint angle variation decreases by24. 4%,and fitness value decreases by 23. 6%.
引文
[1]胡平,刘璇,张建华.一种任务转化的动态避障算法[J].机械设计与制造,2018(7):55-58.
[2]Xin Pengfei,Rong Jili,Yang Yongtai,et al.Inverse kinematics analysis of a 7-DOF space manipulator for trajectory design[J].北京理工大学学报:英文版,2017,26(3):285-291.
[3]王学琨,李刚,周东凯,等.基于DE的时间最优6-DOF机械臂轨迹规划算法[J].计算机仿真,2015,32(8):332-337.
[4]Dizangian B,Ghasemi M R.A fast decoupled reliability-based design optimization of structures using B-spline interpolation curves[J].Journal of the Brazilian Society of Mechanical Sciences&Engineering,2016,38(6):1-13.
[5]于瑞,王成军,郭永存,等.基于杂交算法的机器人时间最优轨迹规划[J].机械传动,2018,42(7):55-60.
[6]应文杰,徐开,徐少平.羽毛球机器人机械臂运动轨迹多目标规划[J].计算机工程与应用,2017,53(3):258-265.
[7]张程,张卓.5-DOF仿人型机械臂关节空间轨迹规划研究[J].组合机床与自动化加工技术,2017(9):23-26.
[8]李小为,胡立坤,王琥.速度约束下PSO的六自由度机械臂时间最优轨迹规划[J].智能系统学报,2015,10(3):393-398.
[9]沈明秀,刘泓滨.基于改进遗传算法的车辆转向机构优化与误差仿真[J].中国工程机械学报,2017,15(6):513-516
[10]Bharathi C,Rekha D,Vijayakumar V.Genetic algorithm based demand side management for smart grid[J].Wireless Personal Communications,2017,93(2):481-502.
[2]Xin Pengfei,Rong Jili,Yang Yongtai,et al.Inverse kinematics analysis of a 7-DOF space manipulator for trajectory design[J].北京理工大学学报:英文版,2017,26(3):285-291.
[3]王学琨,李刚,周东凯,等.基于DE的时间最优6-DOF机械臂轨迹规划算法[J].计算机仿真,2015,32(8):332-337.
[4]Dizangian B,Ghasemi M R.A fast decoupled reliability-based design optimization of structures using B-spline interpolation curves[J].Journal of the Brazilian Society of Mechanical Sciences&Engineering,2016,38(6):1-13.
[5]于瑞,王成军,郭永存,等.基于杂交算法的机器人时间最优轨迹规划[J].机械传动,2018,42(7):55-60.
[6]应文杰,徐开,徐少平.羽毛球机器人机械臂运动轨迹多目标规划[J].计算机工程与应用,2017,53(3):258-265.
[7]张程,张卓.5-DOF仿人型机械臂关节空间轨迹规划研究[J].组合机床与自动化加工技术,2017(9):23-26.
[8]李小为,胡立坤,王琥.速度约束下PSO的六自由度机械臂时间最优轨迹规划[J].智能系统学报,2015,10(3):393-398.
[9]沈明秀,刘泓滨.基于改进遗传算法的车辆转向机构优化与误差仿真[J].中国工程机械学报,2017,15(6):513-516
[10]Bharathi C,Rekha D,Vijayakumar V.Genetic algorithm based demand side management for smart grid[J].Wireless Personal Communications,2017,93(2):481-502.