摘要
针对弧焊机器人动态特性中的非线性和不确定因素,对机器人的轨迹跟踪控制问题进行了研究。为提高跟踪精度和控制性能,提出一种基于高斯基模糊神经网络的轨迹跟踪控制方法。该方法以高斯基作为隶属函数,结合神经网络和模糊算法,设计了高斯基模糊神经网络控制器。采用非线性规划中的最速下降法对模糊神经网络进行自学习,能够在线调节隶属度函数的中心以及关节耦合权值,使得控制器具有更好的自学习与自适应能力。数值仿真结果表明该控制方法能高效地控制机器人的轨迹跟踪。
Because the nonlinear dynamic characteristics and uncertain factors exist in the arc welding robot,to improve the tracking accuracy and the controlling performance,this paper makes a study of its trajectory control and puts forward,the trajectory tracking control method based on Gauss Fuzzy neural network. In this method,Gauss function is taken as the membership function and the neural network is combine with the fuzzy control to design Gauss Fuzzy network controller. The steep descent method in nonlinear programming is applied to the self- learning of the fuzzy neural network,which can be used for the online adjustmen of the center of the membership function and the joint coupling weights and to make the controller have a better self- learning and adaptive ability.The simulation results show that this method can be used to well control the robot trajectory tracking.
出处
《机械制造与自动化》
2016年第6期159-163,共5页
Machine Building & Automation
关键词
模糊神经网络
高斯基函数
弧焊机器人
轨迹跟踪
fuzzy neural network
gauss function
arc welding robot
trajectory tracking