摘 要

水面无人艇是目前世界各个国家研究人员重点关注的高度智能化的海洋无人装备，在商业开发，科学考察以及军事作战等方面具有广阔的应用前景。航向控制系统是无人艇运动控制系统开发过程中的重要子系统之一，稳定的航向控制也是无人艇完成各项任务使命的最基本前提。由于无人艇的操纵性方程中的模型参数通常不易获得，而航向控制问题研究过程中又需要无人艇的操纵性运动方程作为测试验证的基础，因而本论文开展了基于船舶操纵性参数辨识建模的无人艇航向自动控制研究工作。

论文首先介绍了选题的背景和意义，对船舶操纵运动参数辨识和无人艇航向控制的研究现状做了简单介绍。建立了船舶操纵运动坐标系，通过对船舶运动方程的研究，得出了船舶在动坐标系下的水平面运动数学模型，通过船舶弱机动情况下的运动状态，对方程中的某些量进行化简，推导出了船舶操纵运动线性模型，消去方程中的或，可以得到船舶艏摇响应线性模型，由于模型是在弱机动状态下得出，所以为了适应高速环境，加入了非线性项，考虑到外界环境的影响，加入了压舵角，最终得到了船舶操纵运动二阶非线性响应模型。

然后针对如何获得精确的无人艇操纵运动模型这一问题，论文先根据常规的操纵响应模型设计了一种基于EKF方法的参数辨识模型，然后通过对20°Z形操纵实验进行仿真得到了辨识所需的数据，并且根据所设计的EKF方法得到了模型的各个参数，但是在对得到的参数进行分析之后，发现模型部分参数的辨识精度不高，针对这一问题，进一步设计了Levenberg-Marquardt方法对EKF辨识方法进行优化，以提升参数的辨识精度。优化之后，部分参数的辨识误差大大降低。对优化后的模型进行仿真，并且对改进辨识模型与原操纵模型做了平均绝对误差、均方根误差等的计算，通过对比仿真结果和误差计算的结果，验证了改进方法的有效性。

最后，利用目前常用的PID控制原理，设计了一种基于PID控制技术的无人艇航向自动控制方法，并进行了相应的计算机仿真实验和分析；同时考虑到传统PID控制方法，控制参数不能随着外界环境的变化自适应调整，因而论文进一步研究了一种基于模糊控制技术改进的PID控制方法，通过模糊控制技术实现了对比例-积分-微分项控制参数的在线优化，从而设计出了一种可以用于无人艇航向自动控制的自适应模糊PID控制方法，并通过计算机仿真验证了所设计控制方法的有效性和优越性。

关键词：无人艇 操纵运动模型 参数辨识 模糊PID 航向控制

abstract

At present, the surface unmanned vehicle is a highly intelligent marine unmanned equipment which is focused by researchers from all over the world. It has broad application prospects in commercial development, scientific investigation and military operations. The course control system is one of the important subsystems in the development process of the motion control system of the unmanned boat, and the stable course control is also the most basic premise for the unmanned boat to complete all tasks and missions. Because the model parameters in the maneuverability equation of the unmanned boat are usually not easy to obtain, and the maneuverability equation of the unmanned boat is needed as the basis of the test and verification in the course of the study of the course control problem, this paper carries out the research work of the automatic course control of the unmanned boat based on the identification and modeling of the ship maneuverability parameters.

First of all,the background and significance of the topic are introduced, and the research status of ship maneuvering motion parameter identification and unmanned boat course control is briefly introduced. Secondly, the coordinate system of Ship Manoeuvring Motion is established. Through the study of the equation of ship motion,the mathematical model of ship's horizontal motion in the dynamic coordinate system is obtained. Through the state of ship's motion in the case of weak maneuvering, some quantities in the equation are simplified, the linear model of Ship Manoeuvring Motion is derived, and the linear model of ship's yaw response can be obtained by eliminating the or in the equation In order to adapt to the high-speed environment, the nonlinear term is added, considering the influence of the external environment, the rudder angle is added, and finally the second-order nonlinear response model of ship maneuvering motion is obtained.

Then aiming at the problem of how to get the precise maneuvering motion model of the unmanned boat, the paper first designs a parameter identification model based on EKF method according to the conventional maneuvering response model,and then gets the data needed for identification through the simulation of the 20 ° Z-shaped maneuvering experiment, and gets the parameters of the model according to the EKF method designed,but the parameters obtained are improved After line analysis, it is found that the identification accuracy of some parameters of the model is not high. To solve this problem, Levenberg Marquardt method is further designed to optimize the EKF identification method to improve the identification accuracy of parameters. After optimization, the identification error of some parameters is greatly reduced. The optimized model is simulated, and the average absolute error and root mean square error of the improved identification model and the original control model are calculated. The effectiveness of the improved method is verified by comparing the simulation results with the error calculation results.

Finally,using the current PID control principle,an automatic course control method of unmanned boat based on PID control technology is designed, and the corresponding computer simulation experiment and analysis are carried out; at the same time, considering the traditional PID control method, the control parameters can not be self-adaptive adjusted with the change of the external environment, so the paper further studies a P-control technology based on the improvement of fuzzy control technology The ID control method, through the fuzzy control technology, realizes the on-line optimization of the proportional integral differential control parameters, thus designs an adaptive fuzzy PID control method which can be used in the course automatic control of the unmanned boat, and verifies the effectiveness and superiority of the designed control method through the computer simulation.

Key words: unmanned surface vehicle ; motion model ;parameter identification; fuzzy PID ;ship course control

目录

第1章 绪论 2

1.1 选题的背景和意义 2

1.2 船舶操纵性参数辨识建模国内外研究现状 3

1.3 无人艇航向自动控制国内外研究现状 4

1.4 本文的主要内容和结构安排 5

第2章 船舶操纵运动响应模型 6

2.1 船舶操纵运动坐标系 7

2.2 船舶操纵运动线性模型 8

2.3 船舶操纵运动响应模型 9

2.4 本章小结 11

第3章 基于扩展卡尔曼滤波的船舶操纵性参数辨识 11

3.1 参数辨识的基本原理 12

3.2 基于扩展卡尔曼滤波的参数辨识 12

3.2.1 EKF算法原理 13

3.2.2 仿真数据生成 13

3.2.3 数据离散化处理 14

3.2.4 构造辨识模型 15