基于PSO优化极限学习机的机器人控制研究  

Research on robot control based on PSO optimized limit learning machine

作　　者：杜玉香[1] 赵月爱[2] DU Yuxiang;ZHAO Yueai(School of Mechatronics and Automobile,Guangzhou Nanyang Polytechnic Gollege,Guangzhou 510925,China;Department of Computer Science,Taiyuan Normal University,Jinzhong 030619,China)

机构地区：[1]广州南洋理工职业学院机电与汽车学院,广东广州510925 [2]太原师范学院计算机系,山西晋中030619

出　　处：《辽宁科技大学学报》2020年第4期299-303,共5页Journal of University of Science and Technology Liaoning

基　　金：国家自然科学基金(61273294);山西省重点研发计划项目(201803D121088)。

摘　　要：为了提高机器人控制的准确率,采用PSO优化的极限学习机算法来实现机器人的行为控制,建立单隐藏层神经网络机器人控制模型,采用PSO算法对极限学习机的权重和阈值进行优化,根据最小范数二乘解定理,借助可逆矩阵求解权重和阈值最优解,最后获得稳定的极限学习机机器人控制模型。以趋向目标精确度和障碍避开准确度作为主要控制目的进行实例仿真,证明基于PSO优化极限学习机的机器人控制趋向目标准确度高,收敛速度快。In order to improve the accuracy of robot control,a limit learning machine algorithm optimized by PSO is used to control robot behaviors,and a single hidden layer of neural network robot control model is established.The PSO algorithm is used to optimize weight and threshold of the limit learning machine.Based on the least norm square solution theorem,optimal weights and thresholds are solved by means of reversible matrix,and finally a stable robot control model based on the limit learning machine is obtained.Taking the accuracies of moving towards targets and avoiding obstacles as main control objectives,the simulations show that the robot controlled by the PSO-optimized extreme learning machine exhibits high moving accuracy towards the target with fast convergence speeds.

关 键 词：极限学习机 机器人控制 粒子群算法 趋向目标 速度权重 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]