基于智能控制的船舶路径规划及目标循迹

SHIP PATH GENERATION AND TARGET TRACKING BASED ON AUTOMATIC CONTROL

赵健男

11849154

硕士

力学

安松

2020-05-29

2020-07-20

哈尔滨工业大学

深圳

随着世界经济的发展，船舶运输业任务加剧。智能控制可以自主识别环境，为船舶规划路径并进行控制。该系统不仅可以节省人力资源，避免人为原因引起的航行安全隐患，更可以提高系统效率，满足当今航运业务发展要求。船舶智能控制系统复杂程度高，包含众多控制问题。本文主要涉及船舶控制系统的三个方面，分别是船舶在海洋中的状态估计、船舶路径规划以及循迹控制。为了完成船舶的控制以及仿真系统的模拟，本文建立了船舶在海洋环境中的受力与动力学模型。在描述了船舶模型后，又详细模拟了不规则风载荷、浪载荷、与流载荷作用在船体的大小及方向。船舶数学模型和风、浪、流等环境载荷模型的建立为船舶路径生成与循迹控制提供载体。建立了船舶的数学模型之后，本文研究了船舶航行时基于模型的状态观测器。观测器是船舶智能控制重要模块之一。观测器可以实时的估计船舶航行的位置、速度等信息并对测量信号进行滤波处理，去掉一阶波浪运动成分，得到可供控制器使用的状态信息与未知环境载荷估计。本文重点研究了基于智能控制的船舶制导系统。在路径规划方面，完成了对现有算法的改进，开发了离散化人工势场算法。在循迹方面，针对不同路径类型，分别研究了视线制导法和样条曲线插值法。此外，研究了循迹路径生成算法的时间优化问题，得到了给定任意初始条件并满足最终条件的时间优化算法。在船舶循迹控制方面采用非线性PID控制器，以及推力分配问题。最后，借助船舶运动模拟器的辅助，将船舶的数学模型、环境载荷模型、船舶状态观测器、制导算法与循迹算法和船舶运动控制系统进行整合，在Matlab环境下编写了船舶控制仿真系统，该系统可以完成船舶水动力建模、路径规划、目标循迹控制等工作，给出了详细的仿真结果，验证了理论的可行性。

With the development of the world economy, the tasks of the ship transportation industry have intensified. Automatic control can autonomously identify the environment, generate routes for ships and control them. The system can not only save human resources and avoid hidden safety hazards caused by human factors, but also improve system efficiency and meet the development requirements of today's shipping business.The ship's automatic control system is complex and contains many control problems. Three aspects of ship control related problems are studied in the thesis. They are ship’s state estimation, path generation and tracking control. The environmental forces and ship’s dynamics in the sea are studied. Irregular wind loads, wave loads, and current loads acting on the hull are simulated in detail. This mathematical model of ship in seaway is built to help further development of ship’s path generation and tracking control strategy.Model-based state observer used by ship’s dynamic positioning system are applied. Observer is one of the most important components for ship’s navigation. The observer estimates ship's position, speed, and unknown forces on the vessel in real time and filtered first-order wave motion component, which is not favored by low frequency motion control.Ship’s guidance system based on automatic control has also been studied. In terms of path generation, improvements to existing algorithms based on discrete artificial potential field algorithms are developed. For trajectory generation, the line-of-sight guidance law and polynomial curve interpolation method are investigated for different path types. In addition, the trajectory generation problem associated with limited ship’s speed acceleration and jerk when traveling with minimum time is investigated. For the path following controller, we use a non-linear PID controller. The controller contains wind feedforward forces, acceleration-reference feedforward forces, proportion forces due to position deviation, forces due to speed deviation, as well as bias force rejection terms. Total control demand are allocated to each individual thruster by solving a quadratic programming problem with linearized objective function and constraints.Finally, with the aid of ship motion simulator, the ship state observer, guidance algorithm and tracking control are integrated, and the ship control simulation system is written in the Matlab. Simulation results are further discussed.

船舶制导 路径规划 目标循迹 状态观测 控制仿真

guidance path planning target tracking observer control simulation

中文

联合培养

南方科技大学

赵健男. 基于智能控制的船舶路径规划及目标循迹[D]. 深圳. 哈尔滨工业大学,2020.