北京航空航天大学学报 ›› 2016, Vol. 42 ›› Issue (9): 1952-1957.doi: 10.13700/j.bh.1001-5965.2015.0568

• 论文 • 上一篇    下一篇

磁异常干扰下基于约束策略的仿生导航方法

刘坤, 刘明雍, 杨盼盼, 李红, 彭星光   

  1. 西北工业大学 航海学院, 西安 710072
  • 收稿日期:2015-09-02 出版日期:2016-09-20 发布日期:2016-01-04
  • 通讯作者: 刘明雍,Tel.:029-88493006,E-mail:liumingyong@nwpu.edu.cn E-mail:liumingyong@nwpu.edu.cn
  • 作者简介:刘坤,男,博士研究生。主要研究方向:地磁仿生导航、多目标优化、进化搜索算法。Tel.:13689250068,E-mail:liukunkmz@126.com;刘明雍,男,博士,教授,博士生导师。主要研究方向:地球物理场导航、自主水下航行器与自主移动机器人、控制理论与应用。Tel.:029-88493006,E-mail:liumingyong@nwpu.edu.cn
  • 基金资助:
    国家自然科学基金(51379176,51179156,61473233)

Bio-inspired navigation method based on constraint strategy under geomagnetic abnormal disturbance

LIU Kun, LIU Mingyong, YANG Panpan, LI Hong, PENG Xingguang   

  1. School of Marine Science and Technology, Northwestern Polytechnical University, Xi'an 710072, China
  • Received:2015-09-02 Online:2016-09-20 Published:2016-01-04
  • Supported by:
    National Natural Science Foundation of China (51379176, 51179156, 61473233)

摘要: 地磁场异常将会在导航空间中形成极值区域,扰乱磁趋性运动行为,容易造成基于搜索的地磁仿生导航方法陷入局域极小,使得载体迷失航向难以逃离异常区域,最终导致导航失败。针对这一问题,提出了一种基于行为约束策略的导航搜索方法,通过约束搜索行为强制载体扩大探索范围,进而摆脱异常区域干扰,促使载体进入正常地磁场环境。利用磁趋性统计特征与多目标函数的收敛状态构建行为约束的触发及终止条件,将磁趋性历史数据中较好的样本作为约束行为,实施导航搜索行为约束。伴随载体的移动,实现对进化种群的时序更新,最终引导载体到达目标位置,完成导航任务。仿真结果表明,该方法能够有效克服磁场异常环境对地磁仿生导航的干扰,完成导航任务,并能够有效提高远程自主导航的成功率。

关键词: 地磁异常场, 仿生导航, 多目标优化, 进化策略, 磁趋性

Abstract: The abnormity of geomagnetic field will form an extreme value region in the navigation space and disturb the magnetotaxis behavior. Such abnormity will cause the searching based bio-inspired geomagnetic navigation to fall into local minimum, and make the vehicle lose its navigation direction and fail to get out of the abnormal region, which leads to the failure of navigation behavior. In this paper, a navigation method based on behavior constraint strategy is proposed by expanding the exploration scope of vehicle under the constrained searching behavior, and the vehicle is then forced to get rid of the disturbance of abnormal region and get into the normal geomagnetic field. The statistic characteristics of magnetotaxis and convergence state of multi-objective functions are utilized to construct the trigger and termination conditions of behavioral constraints, and better historical data of magnetotaxis are taken as the constraint behavior. The evolution population is updated with the movement of vehicle and finally leads the vehicle to get to the target position. Simulation results show that this method can effectively overcome the abnormal disturbance to the bio-inspired geomagnetic navigation and enhance the success rate of autonomous long range navigation.

Key words: geomagnetic abnormal field, bio-inspired navigation, multi-objective optimization, evolutionary strategy, magnetotaxis

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发