基于MPC的无人船运动控制及可视化仿真系统实现

摘要

Motion control visualization is of great significance to the research on motion control of un‐manned surface vessel (USV) under the environmental disturbance and demand of the automatic colli‐sion avoidance .First ,motion mathematical model of USV under the wind and wave disturbance is es‐tablished ,a series of constraints are set based on the model ,and Model Predictive Control (MPC) al‐gorithm is used to control the motion trajectory of USV .Using Visual C+ + (VC) as the develop‐ment platform ,the visual simulation platform of USV motion is established .In addition ,multi-con‐straint MPC controller based on the hybrid programming of Matlab and VC is developed ,which can realize real-time display of the vessel motion trajectories and dynamic changes in parameters related to vessel motion on the software interface .The experimental results show that the system can rela‐tively well simulate the true navigation track of full‐actuated and under-actuated vessels ,change the motion parameters of vessel according to different vessel features ,and provide test simulation plat‐form for the research on motion control of USV .%运动控制可视化对研究在环境干扰和自动避碰需求下的无人船运动控制具有重要意义。建立了风浪干扰下的船舶运动数学模型，基于模型设定一系列约束，运用模型预测控制（model pre‐dictive control ，MPC）算法对船舶运动轨迹进行控制。以Visual C＋＋为开发环境，搭建了无人船运动可视化仿真平台，实现了基于M atlab与VC混合编程的多约束M PC控制器，在软件界面上能实时显示船舶运动轨迹和船舶运动相关参数的动态变化。实验结果表明，系统能够较好地模拟船舶真实航行轨迹，并能根据不同船舶特性更改船舶运动参数，为无人船运动控制研究提供测试仿真平台。