Dynamic Obstacles Trajectory Prediction and Collision Avoidance of USV

详细信息    查看官网全文

• 英文论文题名：Dynamic Obstacles Trajectory Prediction and Collision Avoidance of USV
• 论文作者：Yuanhui WANG ; Jiaying SHEN ; Xiangbo LIU
• 英文论文作者：Yuanhui WANG ; Jiaying SHEN ; Xiangbo LIU ; College of Automation ; Harbin Engineering University
• 年：2017
• 作者机构：College of Automation,Harbin Engineering University;
• 英文论文关键词：USV ; Trajectory Prediction ; Collision Avoidance ; Elman Network ; Artificial Potential Field
• 会议召开时间：2017-07-26
• 会议录名称：第36届中国控制会议论文集（B）
• 英文会议录名称：Proceedings of the 36th Chinese Control Conference（B）
• 语种：英文
• 分类号：U664.82
• 学会代码：KZLL
• 会议名称：第36届中国控制会议
• 会议地点：中国辽宁大连
• 主办单位：中国自动化学会控制理论专业委员会
• 学会名称：中国自动化学会控制理论专业委员会
• 页数：5
• 文件大小：378k
• 原文格式：D
• 会议级别：全国

摘要

This paper focuses on solving the Unmanned Surface Vehicle autonomous dynamic collision avoidance problem. Firstly, a kinematic and dynamics model of three degree of freedom for USV is established, researches USV's treat method about the data of obstacles, and structures whole environment information to make USV avoid obstacles. Secondly, according to obstacles' position data, Elman predictive model is deduced to predict its motion state. After optimizing in long time domain, obstacles' information can be obtained which close the actual motion state. Thirdly, two factors, distance between USV and target point and estimated collision time, are added to improve traditional artificial potential field. The improved artificial potential field can solve problems that USV cannot find path near obstacles and arrive the target point because of entering the local minimums point. Finally, this research is proved by simulation experiments to improve the level of intelligent and safety, and reduce the personnel expenses effectively.
This paper focuses on solving the Unmanned Surface Vehicle autonomous dynamic collision avoidance problem. Firstly, a kinematic and dynamics model of three degree of freedom for USV is established, researches USV's treat method about the data of obstacles, and structures whole environment information to make USV avoid obstacles. Secondly, according to obstacles' position data, Elman predictive model is deduced to predict its motion state. After optimizing in long time domain, obstacles' information can be obtained which close the actual motion state. Thirdly, two factors, distance between USV and target point and estimated collision time, are added to improve traditional artificial potential field. The improved artificial potential field can solve problems that USV cannot find path near obstacles and arrive the target point because of entering the local minimums point. Finally, this research is proved by simulation experiments to improve the level of intelligent and safety, and reduce the personnel expenses effectively.

引文

[1]Leilei Li.Application Research on Artificial Neural Network in the Construction Engineering Estimate Cost[D].North China Electric Power University,2012
    [2]Mark Richey.ACCC Hosts Dima Khatib[J].The Washington Report on Middle East Affairs,2016,356
    [3]Krogh B H.A Generalized Potential Field Approach to Obstacle Avoidance Control[C].Robotics Research:The Next Five Years and Beyond.Bethlehem,1984,950-955.
    [4]Rui Meng.Soccer Robot Path Planning Based on Improved Artificial Potential Field[D].Beijing Technology and Business University,2010.
    [5]Cetin O,Zagli I.Continuous Airborne Communication Relay Approach Using Unmanned Aerial Vehicles[J].Journal of Intelligent&Robotic Systems,2012,65(1):549-562.
    [6]Wei Li.Ship Track Keeping Control Based on Artificial Neural Networks and Navigation Safety Application Research[D].Dalian Maritime University,2007.