人工势场和虚拟结构相结合的多水下机器人编队控制  

A Multi-AUV Formation Algorithm Combining Artificial Potential Field and Virtual Structure

作　　者：潘无为 姜大鹏[1] 庞永杰[1] 李岳明[1] 张强[1] 

机构地区：[1]哈尔滨工程大学水下机器人技术重点实验室,黑龙江哈尔滨150001

出　　处：《兵工学报》2017年第2期326-334,共9页Acta Armamentarii

基　　金：国家自然科学基金项目(51209051;51509057;51309066);中央高校基本科研项目(HEUCFD1514);海洋工程国家重点实验室开放课题项目(1415)

摘　　要：传统的多水下机器人(AUV)编队算法,例如领航跟随法、虚拟结构法,对编队形成过程中AUV间的避碰问题和编队行进过程中的避障问题,没有进行有效解决。人工势场法可利用势函数进行避碰、避障,但队形组织能力略显不足。针对上述问题,提出了一种人工势场和虚拟结构相结合的多AUV编队控制算法。将系统分为3个部分:编队参考点、虚拟结构质点和AUV.以编队参考点为中心形成期望的虚拟结构以组织队形;虚拟结构质点以期望的虚拟结构为运动目标,并在运动的过程中,通过人工势场斥函数,实现避碰和避障;AUV对虚拟结构质点进行目标跟踪,从而渐进形成AUV的队形。通过编队路径跟踪、编队队形变换和编队避障等一系列仿真实验,对算法的可靠性和灵活性进行充分验证。实验结果表明:在随机的初始位置条件下,多AUV系统可以快速无碰撞地形成队形;在编队行进过程中进行灵活的队形变换,并对障碍物有效避碰。Collision avoidance in the forming process of formation and obstacle avoidance on the formation proceeding are not solved using traditional multi-AUV formation algorithms, such as leader-follower and virtual structure. Artificial potential field algorithm can be used to solve the above problems, while its a- bility of formation organization is not satisfied. A multi-AUV formation control algorithm combing artificial potential field and virtual structure is proposed to overcome the problems. The system is divided into three parts： formation reference point, virtual structure particle and AUV. The desired virtual structure, which is the moving target of virtual structure particles, is organized to surround the formation reference point. Repulsive artificial potential is used to avoid collision and obstacle while the particles move. The AUVs track the particles to form a certain formation asymptotically. The formation path following, formation changing and obstacle avoidance are simulated to verify the availability and flexibility of the proposed al- gorithm. The results indicate that the multi-AUV system can form a formation without collision from ran- dom initial positions, and the fleet can change formation flexibly and avoid obstacles effectively while pro- ceeding.

关 键 词：兵器科学与技术 多水下机器人 编队控制 人工势场 虚拟结构 

分 类 号：TP273.5[自动化与计算机技术—检测技术与自动化装置]