多自主式水下航行器协同控制的人工物理法研究

发布时间：2018-10-23 20:01

【摘要】：对水下多自主式水下航行器(Autonomous Underwater Vehicle,AUV)的编队协同控制和队形重构技术进行研究是水下AUV系统协同完成作业任务的重要研究内容。人工物理法通过设定虚拟的物理力完成机器人速度和方向信息的计算,并根据结果进行实时控制。由于分布式具有对水下传感信息和通信的依赖度较低的控制特点,因此可以很好的应用于可扩展的水下多AUV协同控制中。本文采用人工物理法完成多AUV的队形协同控制,研究了7个AUV采用人工物理法保持六边形队形,进行编队控制穿越障碍区间,AUV通过传感器探测外界环境,获得相关环境信息并确定障碍物的位置;AUV编队在穿越障碍区间时转换为一字队形,并采用边界检测法,沿边行走绕过障碍物;在穿越障碍区间后,AUV编队再次转换为六边形;依据判决条件完成多智能体控制的模态转换,进行队形重构。仿真结果证明本章所提方法的有效性,并具有较好的适时性与柔性。
[Abstract]:The research on formation coordination control and formation reconstruction technology of underwater multi-autonomous underwater vehicle (Autonomous Underwater Vehicle,AUV) is an important research content for underwater AUV system to cooperate in accomplishing the task. The artificial physical method can calculate the velocity and direction information of the robot by setting the virtual physical force, and control it in real time according to the result. Because of its low dependence on underwater sensing information and communication, distributed control can be well applied to the extended underwater multi-AUV collaborative control. In this paper, the formation coordination control of multiple AUV is accomplished by artificial physics method. Seven AUV using artificial physics method to keep hexagonal formation are studied, and the formation control is carried out to cross the obstacle interval. AUV detects the external environment through sensors. The AUV formation is converted into a word formation when crossing the obstacle interval, and the boundary detection method is used to walk around the obstacle along the edge. After crossing the obstacle interval, the AUV formation is converted to hexagonal again. According to the decision condition, the mode conversion of multi-agent control is completed, and the formation reconstruction is carried out. The simulation results show that the proposed method is effective and has good timeliness and flexibility.
【作者单位】：中国海洋大学信息科学与工程学院;山东省科学院海洋仪器仪表研究所;北京师范大学政府管理学院;
【基金】：山东省科技攻关项目(2013GHY11519) 青岛市战略性新兴产业培育计划项目(13-4-1-45-hy) 青岛创新与创业领导人才项目(13-cx-2)资助~~
【分类号】：TP242

【参考文献】