Efficient Path Planning Method of USV for Intelligent Target Search

Jing Zhang,Fan Zhang,Yunsong Li,Zengyuan Liu

doi:10.1007/s41651-019-0035-0

Jing Zhang, Fan Zhang + Show 2 more

Open Access

https://doi.org/10.1007/s41651-019-0035-0

Copy DOI

Abstract

In order to accomplish a target search task safely and efficiently and make full use of prior information and real-time information, a path planning method of unmanned surface vehicle (USV) for intelligent target search is proposed. The overall strategy is divided into three parts: global path planning based on prior information, local path planning based on real-time information, and improved A* obstacle avoidance algorithm. Before the start of the task, the global path planning is carried out based on prior information such as the initial position of USV, the predicted position of the target and range of search area. After the start of the task, if USV finds suspicious targets, in order to further approach these suspicious targets, it will enter different local path planning modes according to the characteristics of these targets. During task execution, if obstacles are encountered, an improved A* obstacle avoidance algorithm is adopted. The simulation results show that the proposed method can improve the efficiency of target recognition and reduce the turning cost of USV when encountering obstacles. So, for USV intelligent target search, the proposed path planning method can save resources and improve search efficiency.

Highlights

In the military and civilian fields, with the rapid development of technology, unmanned surface vehicle (USV) is increasingly used to perform various tasks such as target search, post-disaster personnel search and rescue, surface warning, and security patrol (Kim and Lee, 2019)
Aiming at the shortcomings of existing path planning methods, this paper proposes a path planning algorithm of USV for intelligent target search, which combines global path planning with local path planning and improves the efficiency of search targets
This paper is organized as follows: Firstly, the global path planning based on prior information is introduced, which can search the target area by expanding spiral trajectory; secondly, the local path planning based on real-time information is introduced, and the quick confirmation of suspicious targets is achieved by switching automatically among three modes; thirdly, the traditional A* path planning algorithm is introduced, and the A* algorithm is improved by reducing turning times

Summary

Introduction

In the military and civilian fields, with the rapid development of technology, unmanned surface vehicle (USV) is increasingly used to perform various tasks such as target search, post-disaster personnel search and rescue, surface warning, and security patrol (Kim and Lee, 2019). Aiming at the shortcomings of existing path planning methods, this paper proposes a path planning algorithm of USV for intelligent target search, which combines global path planning with local path planning and improves the efficiency of search targets. It increases the map coverage of USV, and considering the cost of USV turning, adds the turning cost to the evaluation function, improves the A* obstacle avoidance algorithm and saves resources. This paper improves the planning method of global path planning, completes the search of the target area by expanding spiral trajectory and makes the whole motion process more stable and safer. During the whole target search task, the automatic switching of these three modes can ensure that the target is detected quickly

13 Page 4 of 9

13 Page 6 of 9

13 Page 8 of 9

Findings

Conclusion