基于MPC的USV自主航行仿真研究

Research on the simulation of USV autonomous navigation based on MPC

水面无人艇的主要任务包括按照设定航线循迹航行以及在面临复杂环境下能够安全经济地避开海上静态障碍物和/或动态障碍物并最终顺利到达目的地。本文基于模型预测控制的设计框架,在船舶运动模型中充分考虑船舶的水动力特性以及风流等外界的干扰,以航迹偏移量、航速变化量、碰撞危险度和规则符合度构造评价函数,以无人艇执行机构的动力为约束条件,设计实现了一种水面无人艇自主航行模型。通过仿真实验,验证了所设计的控制算法可实现水面无人艇在期望航迹上的自主航行。

The main tasks of unmanned surface vehicle include following a set route and being able to safely and economically avoid static and/or dynamic obstacles at sea in a complex environment and finally reach the destination smoothly.Based on the design framework of model predictive control, this paper fully considers the ship’s hydrodynamic characteristics and additional disturbances such as wind and current in the ship motion model, and constructs the evaluation function based on track offset, speed variation, collision risk and rule compliance. Taking the power of the unmanned surface vehicle’s executive mechanism as the constraint condition, the design and realization of an unmanned surface vehicle autonomous navigation model is achieved. The simulation experiment verifies that the involved control algorithm can realize the autonomous navigation of the unmanned surface vehicle on the desired trajectory.