基于准无限时域MPC的无人艇轨迹跟踪控制器

Trajectory Tracking Controller for Unmanned Surface Vehicle Based on Quasi-infinite Horizon Model Predictive Control

为解决水面无人艇(unmanned surface vehicle,USV)如何稳定地跟踪任意给定轨迹的问题,设计了一种基于准无限时域模型预测控制(model predictive control,MPC)的轨迹跟踪控制器.基于USV三自由度动力学状态空间模型,对模型近似线性化和离散化处理,以便于设计控制器.为保证控制器系统的稳定性,应用准无限时域模型预测控制理论,设计了一种终端惩罚函数,将其融入到线性模型预测控制方法中,提出了一种USV准无限时域轨迹跟踪的模型预测控制器.通过理论证明了所设计控制器的稳定性.并在仿真平台上进行直线轨迹、正弦曲线轨迹仿真,验证了所设计控制器的适用性、稳定性.

To solve the problem of how unmanned surface vehicle(USV)can stably track any given trajectory,a trajectory tracking controller based on quasi-infinite time domain model predictive control(MPC)was designed.Based on the three-degree-of-freedom dynamic state space model of USV,the model was approximately linearized and discretized to facilitate the design of the controller.In order to ensure the stability of the controller system,a terminal penalty function was designed by applying the quasi-infinite time domain model predictive control theory,which was integrated into the linear model predictive control method,and a model predictive controller of USV quasi-infinite time domain trajectory tracking was proposed.The stability of the designed controller was proved by theory.The simulation of linear trajectory and sinusoidal trajectory on the simulation platform verifies the applicability and stability of the designed controller.