IVP模型在AUV路径规划与实时避障中的应用研究  

Application Research of IVP Model in AUV Path Planning and Real-time Obstacle Avoidance

作　　者：李聪 胡超芳 沈同圣 赵德鑫 LI Cong;HU Chaofang;SHEN Tongsheng;ZHAO Dexin(School of Electrical and Information Engineering,Tianjin University,Tianjin 300072,China;National Innovation Institute of Defense Technology,Beijing 100071,China)

机构地区：[1]天津大学电气自动化与信息工程学院,天津300072 [2]国防科技创新研究院,北京100071

出　　处：《机械科学与技术》2022年第10期1477-1482,共6页Mechanical Science and Technology for Aerospace Engineering

基　　金：国家自然科学基金项目(41906169)。

摘　　要：针对传统基于行为的控制体系结构在解决AUV局部路径修复问题时,AUV整体行为容易出现不可接受的缺陷等不足,设计一种新的行为融合-区间优化(IVP)方法。首先将路径规划任务划分为路径点跟踪行为、实时避障行为等多个具体的行为,结合环境约束,将AUV控制决策选择视为一个多目标优化问题,然后利用IVP模型进行行为的协调,最后基于MOOS-IVP体系以静态目标周围的环绕式路径点跟踪进行了仿真分析,结果表明IVP模型在解决决策空间随决策变量增加呈指数增长的同时,能够实时避障并保证了结果的最优化。Aiming at the shortcomings of traditional behavior-based control architecture in solving the problem of local path repair of autonomous underwater vehicle(AUV), such as unacceptable defects of AUV′s overall behavior, a new behavior fusion-Interval Programming(IVP) method was designed in this paper. Firstly, the path planning task is divided into multiple specific behaviors such as path point tracking behavior, real-time obstacle avoidance behavior;and combined with environmental constraints, the AUV control decision selection is considered as a multi-objective optimization problem. Then the IVP model is used to coordinate these behaviors above. Finally, the simulation analysis is carried out based on MOOS(Mission oriented operating suite)-IVP system with the surrounding path point tracking around the static target. The results show that the IVP model in proposed method can avoid obstacles in real time and ensure the optimization of results while solving the decision space grows exponentially with the increase of decision variables.

关 键 词：AUV 路径点跟踪 区间优化 实时避障 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]