结合速度和干扰观测的船舶路径跟踪模型预测控制
2021-04-30李宗宣卜仁祥范艺
李宗宣 卜仁祥 范艺



摘要:
针对船舶路径跟踪控制中的舵角需优化﹑舵幅和舵速受约束等问题,提出模型预测控制(model predictive control, MPC)算法。为处理系统高阶状态值不易测量以及环境干扰问题,设计高阶非线性观测器,同时对船速和包含模型不确定项和外界干扰的总未知项进行估计。以引入舵机响应系统的MMG模型作为预测模型,不仅能提高预测精度,而且更符合船舶运动控制的实际情况。仿真结果表明,所设计的控制器在风浪流时变干扰下仍能跟踪参考路径,舵角幅值小且其变化是光滑的,速度值和总未知项也均能被准确地逼近,验证了所提算法的有效性。
关键词:
船舶运动控制; 模型预测控制(MPC); 路径跟踪; 非线性观测器
中图分类号: U664.82
文献标志码: A
Model predictive control of ship path following combined
with velocity and disturbance observers
LI Zongxuan, BU Renxiang, FAN Yi
(Navigation College, Dalian Maritime University, Dalian 116026, Liaoning, China)
Abstract:
To solve the rudder angel optimization requirement, rudder amplitude constraint and rudder rate constraint in the ship path following control, a model predictive control (MPC) algorithm is proposed. To deal with the problem of high-order state values of the system being not easy to measure and environmental
disturbance, a high-order nonlinear observer is designed to estimate the ship velocity and the total unknown terms including model uncertainties and external disturbance. The MMG model with the rudder response system is used as the predictive model, which not only can improve the prediction accuracy, but also accords with the actual situation of ship motion control. The simulation results show that: the designed controller can follow the desired path under the time-varying disturbance of wind, wave and current; the rudder angle amplitude is small and its variation is smooth, and the velocity value and the total unknown terms can be accurately approximated, which verifies the effectiveness of the proposed algorithm.
Key words:
ship motion control; model predictive control (MPC); path following; nonlinear observer
收稿日期: 2020-03-19
修回日期: 2020-05-09
基金項目:
国家自然科学基金(51939001,61976033,51379026);辽宁省自然科学基金(20170540089)
作者简介:
李宗宣(1996—),男,陕西眉县人,硕士研究生,研究方向为船舶运动控制、模型预测控制,(E-mail)zongxuanli@126.com;
卜仁祥(1973—),男,辽宁岫岩人,副教授,二副,博士,研究方向为船舶运动控制、船舶非线性控制,
(E-mail)burenxiang@dlmu.edu.cn
0 引 言
欠驱动船舶路径跟踪是一个不考虑时间限制的控制领域问题[1]。为解决此问题,FOSSEN将视觉制导算法应用到船舶路径跟踪中[2],使船舶在实际可行的情况下实现参考路径跟踪。由于船速不易被直接测量,文献[3]利用高增益状态观测器对船舶横、纵向速度进行估计。文献[4]为估计船速设计出线性观测器,但线性观测器对较小偏差的估计精度较低,而在估计较大偏差时系统会产生振荡。……
