Gaussian process based model predictive tracking control with improved iLQR
简介：This article proposes a Gaussian process(GP)based model predictive control(MPC)method to solve the tracking control of wheeled mobile robot(WMR)with uncertain model parameters.Firstly,a Gaussian process velocity prediction model is proposed to compensate for the unknown dy-namic model,as the kinematic model cannot accurately characterize the motion characteristics of the robot.Then,by introducing the Lorentz function,the improved iterative linear quadratic regulator(iLQR)method is used to solve the nonlinear MPC(NMPC)controller with constraints.In addi-tion,in order to reduce computational burden,a closed gradient calculation method is introduced to improve algorithm efficiency.Finally,the feasibility and effectiveness of this method are verified through simulation and experiment.展开
学者：LIHengZhuGongcaiLiuAndongNiHongjie
关键词：model predictive controlGaussian processiterative linear quadratic regulatortrajectory tracking
在线出版日期：2026-04-17 （网站首发日期）
