基于广义二型模糊神经网络的移动机器人轨迹跟踪控制  

TRAJECTORY TRACKING CONTROL OF MOBILE ROBOT BASED ON GENERAL TYPE-2 FUZZY NEURAL NETWORK

作　　者：周俊 赵涛[1] Zhou Jun;Zhao Tao(College of Electrical Engineering,Sichuan University,Chengdu 610065,Sichuan,China)

机构地区：[1]四川大学电气工程学院,四川成都610065

出　　处：《计算机应用与软件》2023年第4期68-74,79,共8页Computer Applications and Software

基　　金：四川省科技计划项目(2020YFG0115);成都市科技计划项目(2019-YF05-00958-SN)。

摘　　要：针对轮式移动机器人的轨迹跟踪问题,提出一种广义二型模糊神经网络控制方法。模糊控制可以弥补机器人动态特性中的非线性和不确定性因素,而广义二型模糊系统能更有效地处理外界干扰和参数扰动等不确定性,广义二型模糊神经网络系统结合了神经网络强大的非线性拟合能力和自学习能力,能够更有效地对规则库中可能存在的不确定性进行建模。它可以进一步提高控制精度,达到跟踪的目的。仿真结果表明,与PID控制器、模糊控制器和一型模糊神经网络控制器相比,该方法能更好地跟踪轮式移动机器人的运动轨迹且拥有更好的抗干扰能力。Aimed at the trajectory tracking problem of wheeled mobile robots,a general type-2 fuzzy neural network control method is proposed.Fuzzy control could make up for the non-linear and uncertain factors in the dynamic characteristics of robot,while the general type-2 fuzzy system could deal with the uncertainties such as external interference and parameter disturbance more effectively.The general type-2 fuzzy neural network system combined the strong non-linear fitting ability and self-learning ability of neural network,which could more effectively model the uncertainty that may exist in the rule base.It could further improve the control accuracy and achieve the purpose of tracking.The simulation results show that compared with PID controller,fuzzy controller and type-1 fuzzy neural network controller,this method can better track the trajectory of wheeled mobile robot and has better anti-interference ability.

关 键 词：广义二型模糊控制 神经网络 轨迹跟踪 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]