火力与指挥控制2024,Vol.49Issue(3) :73-79,87.DOI:10.3969/j.issn.1002-0640.2024.03.010

一种动态不确定海洋环境下的USV态势自主感知及避障算法

An Algorithm for USV Situation Autonomous Awareness and Obstacle Avoidance in Dynamic and Uncertain Marine Environment

王潋 陈霄 完颜娟
火力与指挥控制2024,Vol.49Issue(3) :73-79,87.DOI:10.3969/j.issn.1002-0640.2024.03.010

一种动态不确定海洋环境下的USV态势自主感知及避障算法

An Algorithm for USV Situation Autonomous Awareness and Obstacle Avoidance in Dynamic and Uncertain Marine Environment

王潋 1陈霄 2完颜娟1
扫码查看

作者信息

  • 1. 解放军 91054 部队,北京 102442
  • 2. 军事科学院,北京 100091
  • 折叠

摘要

针对无人艇在动态不确定海洋环境下的态势自主感知和实时避障问题,提出了一种遵守COLREGs海事规则,且考虑传感器测量误差和障碍物随机运动不确定信息的无人艇航行态势自主感知及避障算法.阐述了无人艇在海上航行时须遵守的 3 类海事规则,追越、交叉相遇和正面相遇;考虑到传感器存在的测量误差、观测漂移以及障碍物运动的随机不确定性,对相关误差进行建模,并基于融合COLREGs规则的速度障碍法,设计提出了一种融合COLREGs海事规则的USV态势自主感知及避障算法;通过正面相遇、追越、左右交叉相遇 4 种典型场景下的仿真实验,验证了所提出算法的有效性.

Abstract

Aiming at the problem of autonomous situation awareness and real-time obstacle avoidance of the USV in dynamic and uncertain marine environment,an autonomous navigation situation awareness and obstacle avoidance algorithm for the USV is proposed,which complies with the COLREGs maritime rules and considers the measurement errors of sensors and the uncertain information of random movement of obstacles.Firstly,three kinds of maritime rules,overtaking,cross encounter and head-on en-counter,must be observed by the USV when navigating at sea are expounded;After that,the measurement errors,observation drift and random uncertainty of obstacle movement existing in the sensor are considered,the relevant error is modeled,and the USV situation autonomous awareness and obstacle avoidance algorithm integrating COLREGs maritime rules is designed and proposed based on the speed obstacle method integrating COLREGs rules;Finally,the effectiveness of the proposed algorithm is verified by four typical scenarios:frontal encounter,overtaking,left and right cross encounter.

关键词

无人艇/海事规则/冲突场景/自主避障/不确定性建模/极大似然法

Key words

USV/maritime rules/conflict scenarios/autonomous obstacle avoidance/uncertainty modeling/maximum likelihood method

引用本文复制引用

出版年

2024
火力与指挥控制
火力与指挥控制研究会,火力与指挥控制专业情报网

火力与指挥控制

CSTPCDCSCD北大核心
影响因子:0.312
ISSN:1002-0640
参考文献量14
段落导航相关论文