基于多永磁同步电机协同推进的两栖无人艇跟踪控制研究

The unmanned surface vehicle (USV) and unmanned underwater vehicle (UUV) are two kinds of intelligent marine vehicles. However, previous studies mainly focus on a specific motion control system (i.e. USV or UUV). Nevertheless, complicated nonlinear dynamics of the thruster have also not been considered. Considering both USV and UUV dynamics, a novel amphibious unmanned marine vehicle (AUMV) system is proposed by combining the investigation of the unmanned vehicle dynamics with the research on the actuator nonlinearity, whereby the coordination propulsion is employed by multiple permanent magnet synchronous thrusters (PMST). By virtue of the novel AUMV system, innovative trajectory tracking control schemes for intelligent marine vehicles are developed in this proposal. The significant contributions are as follows: 1) The unified motion model of the AUMV is established; 2) Considering the unmodeled dynamics and unknown external disturbances, the robust adaptive motion control methodology for the AMUV is proposed; 3) The entire motion model of the AUMV is proposed by incorporating the PMST dynamics,; 4) Combining with the PMST control synthesis, fuzzy-neural-based adaptive robust tracking controller is implemented to track any smooth reference trajectory with high accuracy. Finally, simulation and experiment results together with comprehensive comparisons demonstrate the effectiveness and superiority of the proposed schemes. In this context, this project aims to set up the novel robust adaptive tracking control theory for multi-mode AUMV including complicated thruster dynamics, and thereby contributing to autonomous control of marine vehicles under unknown complex environmental disturbances.

水面无人艇和水下航行器是两大智能海洋载运工具，然而现有理论结果仍集中于单一运动控制系统研究，而且推进器的复杂非线性动态并未考虑。本项目兼具水面和水下航行动态，采用多永磁同步电机协同推进方式，通过结合无人艇运动模型探讨和复杂执行器动态特性研究，提出新颖的两栖无人艇系统，为智能海洋航行器的航迹跟踪控制等研究提供新的系统化研究思路和方法。研究内容包括：建立两栖无人艇的统一运动模型；考虑未建模动态和未知外界干扰，研究两栖无人艇的鲁棒自适应运动控制方法；考虑永磁同步推进器动态特性，构建两栖无人艇整体运动模型；结合永磁同步电机控制研究，设计模糊神经网络鲁棒自适应跟踪控制器，实现任意光滑参考航迹的精确跟踪控制；通过仿真分析和实船实验，验证所得理论方法的有效性和优越性。本项目旨在面向考虑复杂推进器动态特性的多工作模式两栖无人艇，提出新的鲁棒自适应跟踪控制理论，实现未知复杂航行环境下的海洋航行器自主控制。

本项目针对无人艇水面和水下工作模式下的模型不一致性，充分考虑不同模式下的模型不确定性及外界干扰，结合各模式下艇体水动力、推进器推力等因素以及相应的外界干扰模型，将各推进器的非线性机桨特性、输入饱和非线性及死区非线性融入到两栖无人艇六自由度动态模型中，建立了具有复杂推进器动态的两栖无人艇Lagrange动态模型。针对单一无位置传感器算法无法适用于全速度范围两栖无人艇永磁同步推进电机控制的问题，采用转子预定位和高频信号注入法提取静止状态下的永磁同步推进电机转子位置，采用I/F流频比法控制其低速工况，提出一种可消除直流偏置的改进型反电动势法和基于旋转坐标系下新型滑模观测器的中高速范围永磁推进电机无位置传感器控制方法，结合低速/中高速切换融合算法，实现了全速度范围两栖无人艇永磁推进电机无位置传感器控制功能，经仿真和电机台架试验，验证了所提算法的有效性。针对两栖无人艇航迹跟踪控制问题，本项目以具有复杂推进器动态的两栖无人艇动态模型为研究对象，在输出反馈控制框架下，基于动态面控制和二阶滑模控制策略，结合简洁模糊神经在线逼近器的估算性能，设计了模糊神经鲁棒自适应航迹跟踪控制器，实现了对任意给定光滑航迹的精确跟踪控制。