改进RBF神经网络在智能机器人轨迹规划中的研究
Improved RBF Neural Network in the Intelligent Robot Research on Trajectory Planning
刘翔 1王开科 2李菲3
作者信息
- 1. 四川城市职业学院,四川 成都 610110
- 2. 成都理工大学,四川 成都 610059
- 3. 重庆大学,重庆 400044
- 折叠
摘要
针对工业生产中对智能机器人轨迹规划的要求越来越高,在工业机器人运动模型的基础上,提出了一种将RBF神经网络和遗传算法相结合的工业机器人轨迹规划方法.通过遗传算法对RBF神经网络的网络结构、连接权值和阈值进行优化,精确跟踪机器人的轨迹.通过仿真将与未改进前的轨迹规划算法进行比较,验证该方法的优越性.结果表明,与改进前的规划算法相比,文中规划方法误差小,适应性强,能够满足工业机器人轨迹规划的预期要求.为工业机器人轨迹规划方法的发展提供了一定的参考.
Abstract
Aiming at the increasing demand of intelligent robot trajectory planning in industrial production,a trajectory plan-ning method of industrial robot combining RBF neural network and genetic algorithm was proposed based on the motion model of industrial robot.The network structure,connection weight and threshold of RBF neural network were optimized by genetic algo-rithm,and the trajectory of robot was accurately tracked.The simulation results show that the proposed method is superior to the unimproved trajectory planning algorithm.The results show that,compared with the previous improved planning algorithm,the proposed method has less error and stronger adaptability,and can meet the expected requirements of trajectory planning of indus-trial robots.It provides some reference for the development of trajectory planning method of industrial robot.
关键词
工业机器人/轨迹规划/RBF神经网络/遗传算法/关节轨迹Key words
Industrial Robot/Trajectory Planning/RBF Neural Network/Genetic Algorithm/Joint Trajectory引用本文复制引用
基金项目
四川省教育厅自然科学研究项目(18ZB0351)
出版年
2024
NETL
NSTL
维普
万方数据