基于目标干扰意图识别的无人艇避碰算法

茅云生; 彭伟; 向祖权; 宋利飞; 刘梦雪

doi:10.16411/j.cnki.issn1006-7736.2021.02.004

大连海事大学学报 >

2021 , Vol. 47 >Issue 2: 26 - 34

DOI: https://doi.org/10.16411/j.cnki.issn1006-7736.2021.02.004

船舶与海洋工程

基于目标干扰意图识别的无人艇避碰算法

茅云生 ,
彭伟 ,
向祖权 ,
宋利飞 ,
刘梦雪

展开

（1.高性能船舶技术教育部重点实验室（武汉理工大学），武汉 430063；2.武汉理工大学交通学院，武汉　430063）

宋利飞（1989 — ），男，博士，副教授，E-mail:songlifei@whut.edu.cn.

收稿日期: 2020-12-15

修回日期: 2021-03-05

网络出版日期: 2021-03-05

基金资助

国家自然科学基金资助项目（51809203).

收起

Collision avoidance algorithm of unmanned surface vehicle based on target interference intent recognition

MAO yun-sheng ,
PENG wei ,
XIANG zu-quan ,
SONG li-fei ,
LIU meng-xue

Expand

(1.Key Laboratory of High Performance Ship Technology (Wuhan University of Technology), Ministry of Education,Wuhan 430063,China;2.Transportation School, Wuhan University of Technology,Wuhan 430063,China;; 3. Information Center, Renmin Hospital of Wuhan University,Wuhan 430060, China）

Received date: 2020-12-15

Revised date: 2021-03-05

Online published: 2021-03-05

Fold

摘要

在无人艇避碰规划过程中,为准确识别障碍干扰意图，提出一种结合主成分分析（Principal Component Analysis，PCA）和支持向量机（Support Vector Machine，SVM）的分类算法.首先，利用PCA对船舶模型运动的高维运动特征数据集进行降维，获取特征集的主成分；再通过SVM对经过处理得到的低维主成分进行分类处理，以识别障碍对无人艇的干扰意图类型为主动干扰还是非主动干扰；最后,根据障碍的不同干扰类型，采用不同算法进行避碰、脱逃，通过这种差异化的处理，提升无人艇运动的安全性.仿真结果证明了该算法的有效性.

关键词： 无人艇; 避碰; 主动干扰; 意图识别; 分类算法

本文引用格式

茅云生 , 彭伟 , 向祖权 , 宋利飞 , 刘梦雪 . 基于目标干扰意图识别的无人艇避碰算法[J]. 大连海事大学学报, 2021 , 47(2) : 26 -34 . DOI: 10.16411/j.cnki.issn1006-7736.2021.02.004

Abstract

In order to accurately identify the obstacles interference intention in the process of collision avoidance planning for unmanned surface vehicle (USV), the classification algorithm based on principal component analysis (PCA) and support vector machine (SVM) was proposed. Firstly, PCA was used to reduce the dimension of the highdimensional motion feature data set of ship model motion to obtain the principal components of the feature set. Then, SVM was used to classify the low dimensional principal components, so as to identify whether the interference intention type of obstacles to the UAV is active interference or non active interference. Finally, according to the different interference types of obstacles, different algorithms were used to avoid collision and escape. Through this differential processing, the safety of the USV was improved. The simulation results verify the effectiveness of the algorithm.

Key words： unmanned surface vehicle (USV); collision avoidance; active interference; intent recognition; classification algorithm

参考文献

[1]Ru-jian Yan, Shuo Pang, et al.Development and missions of unmanned surface vehicle[J].Journal of Marine Science and Application, 2010, 9(4):451-457 [2] Victor J Sussman.USVs Are Key in the Counter–A2/AD Fight[J].United States Naval Institute. Proceedings, 2020, 12(1):24-34 [3]Lee S M, Kwon K Y, Joh J.A fuzzy logic for autonomous navigation of marine vehicles satisfying COLREG guidelines[J].International Journal of Control Automation & Systems, 2004, 2(2):171-181 [4]Song L.A Study on Path Planning Algorithms Based upon Particle Swarm Optimization[J].Journal of Information & Computational Science, 2015, 12(2):673-680 [5] V.Karasev, A. Ayvaci, B. Heisele, et al. Intent-aware long-term prediction of pedestrian motion[C]. IEEE International Conference on Robotics and Automation, Stockholm, 2016:2543-2549. [6] A.Rasouli, I. Kotseruba and J. K. Tsotsos. Are they going to cross? A benchmark dataset and baseline for pedestrian crosswalk behavior[C]. Proceedings of the IEEE International Conference on Computer Vision Workshops, Venice, 2017:206-213. [7]季学武, 费聪, 何祥坤, 刘玉龙, 刘亚辉.基于网络的驾驶意图识别及车辆轨迹预测[J].中国公·学报, 2019, 32(06):34-42 [8]JI X W, FEI C, HE X K, LIU Y L, LIU Y H.Driving intention recognition and vehicle trajectory prediction based on LSTM network[J].China Journal of Highway and Transport, 2019, 32(06):34-42 [9]ZYNER A, WORRAL S, NEBOT E.A Recurrent Neural Network Solution for Predicting Driver Intention at Unsignalized Intersections[J].IEEE Robotics and Automation Letters, 2018, 3(3):1759-1764 [10]于大腾.面对接近式威胁的在轨飞行器安全性分析方法[J].宇航学报, 2020, 41(8):1084-1093 [11]YU D T.Safety analysis method of on-orbit aircraft facing approach threat[J].Journal of Astronautics, 2020, 41(8):1084-1093 [12]Wang P Y, Guo Y N, Wie B.Orbital rendezvous performance comparison of differential geometric and ZEMZEV feedback guidance algorithms[J].Astrodynamics, 2019, 3(1):79-92 [13]杨柏丞, 赵志垒.基于改进模拟退火算法的多船会遇避碰决策[J].大连海事大学学报, 2018, 44(02):22-26 [14]YANG B C, ZHAO Z L.Multi-ship encounter collision avoidance decision based on improved simulated annealing algorithm[J].Journal of Dalian Maritime University, 2018, 44(02):22-26 [15] Hemanth Mithun Praveen, Divya Shah, Krishna Dutt Pandey, et al.PCA based health indicator for remaining useful life prediction of wind turbine gearbox[J].VIBROENGINEERING PROCEDIA, 2019, 4(2):31-36 [16]关峰, 于少强, 靳廉洁, 等.基于支持向量机的巴拿马型干散货船队运营区域选择[J].大连海事大学学报:自然科学版, 2017, 43(2):75-80 [17]GUAN F, YU S Q, JIN L J, et al.Operation area selection of Panamax dry bulk fleet based on support vector machine[J].Journal of Dalian Maritime University: Natural Science Edition, 2017, 43(2):75-80 [18]张铎.我国对海上避碰规则的保留与适用[J].世界海运, 2017, 40(11):1-12 [19]ZHANG D.The reservation and application of China's rules for preventing collisions at sea[J].World shipping, 2017, 40(11):1-10

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献