非完整移动机器人路径跟踪的快速非线性模型预测控制方法  

Fast nonlinear model predictive control method for nonholonomic mobile robot path following

作　　者：占红武[1] 厉晟 胥芳[1] Zhan Hongwu;Li Sheng;Xu Fang(Key Laboratory of E&M(Zhejiang University of Technology),Ministry of Education&Zhejiang Province,Hangzhou 310023)

机构地区：[1]浙江工业大学特种装备制造与先进加工技术教育部/浙江省重点实验室,杭州310023

出　　处：《高技术通讯》2021年第4期405-416,共12页Chinese High Technology Letters

基　　金：国家重点研发计划(2018 YFB1309404)资助项目。

摘　　要：本文用快速非线性模型预测控制方法进行了非完整移动机器人路径跟踪问题的研究。采用虚拟目标跟踪法建立非完整移动机器人路径跟踪问题的非线性系统,转化为以Bolza形式的最优控制问题模型。通过一种障碍函数法处理不等式约束将其整合至性能指标函数中,基于拉格朗日乘数法建立最优化必要条件的线性方程组,以连续/广义极小残余算法(C/GMRES)为计算核心求解线性方程组并引入多重打靶法增加数值精度,同时提出一种优化求解方法降低算法计算负载。仿真对比结果表明,所提控制方法在满足实时性与保证控制精度的条件下具有更高的计算效率。最后通过实验结果证明所提出的控制方法在非完整移动机器人路径跟踪过程中的有效性。A fast nonlinear model predictive control method is used to study the trajectory tracking problem of nonholonomic mobile robots.A nonlinear system model of the path tracking problem of a nonholonomic mobile robot is established using the virtual target tracking method,which is transformed into an optimal control problem model in the form of Bolza.It is integrated into the performance index function by processing an inequality constraint using a barrier function method.The Lagrangian multiplier method is used to establish a set of linear equations for optimization.The continuation/general minimum residual(C/GMRES)algorithm is used as the calculation core to solve the linear equations and the multiple shooting method is introduced to increase the numerical accuracy.At the same time,an optimized solution method is introduced to reduce the calculation of the control method’s load.The simulation comparison results show that the proposed control method has higher calculation efficiency under the conditions of real-time performance and guaranteed control accuracy.Finally,experimental results demonstrate the effectiveness of the proposed control method in nonholonomic mobile robot path tracking.

关 键 词：非线性模型预测控制(NMPC) 路径跟踪 虚拟目标跟踪 障碍函数 连续/广义极小残余算法(C/GMRES) 多重打靶法 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]