本文研究船舶模拟驾驶系统障碍物自动识别方法，满足船舶模拟驾驶系统在未知环境下的避障需求。船舶模拟驾驶系统的电子导航雷达单元，利用激光雷达传感器采集船舶环境信息的激光点云数据；控制中心依据所采集激光点云数据，通过自适应距离阈值聚类法聚类激光点云数据，提取障碍物特征向量。将提取障碍物特征向量作为支持向量机的输入，利用粒子群优化算法确定支持向量机的最优核参数，利用设置最优核参数的支持向量机，输出船舶模拟驾驶系统障碍物自动识别结果。实验结果表明，船舶模拟驾驶系统采用该方法，自动识别模拟驾驶时的静态障碍物以及动态障碍物，满足船舶安全航行需求。
This paper studies the automatic obstacle identification method of the ship simulation driving system to meet the obstacle avoidance requirements of the ship simulation driving system in unknown environment. The electronic navigation radar unit of the ship navigation simulation system collects the laser point cloud data of ship environmental information by using lidar sensors. According to the collected laser-point cloud data, the control center uses adaptive distance threshold clustering method to cluster the laser-point cloud data and extract the obstacle feature vectors. The extracted obstacle feature vectors were set as the input of the support vector machine, and the optimal kernel parameters of the support vector machine were determined by the particle swarm optimization algorithm. The automatic obstacle recognition results of the ship simulation system were output by the support vector machine with the optimal kernel parameters. The experimental results show that this method can automatically identify static and dynamic obstacles during simulated driving, and meet the requirements of safe navigation of ships.
2022,44(22): 144-147 收稿日期：2022-08-17
DOI：10.3404/j.issn.1672-7649.2022.22.028
分类号：TP391
基金项目：江苏省现代教育技术研究课题（2021-R-91015）
作者简介：刘晓峰(1978-),男,硕士,讲师,研究方向为航海技术
参考文献：
[1] 时光志, 周毅, 李萌, 等. 基于虚拟现实技术的LNG船舶仿真系统[J]. 船海工程, 2021, 50(3): 25–28+33
SHI Guangzhi, ZHOU Yi, LI Meng, et al. LNG ship navigation simulation system based on virtual reality technology[J]. Ship & Ocean Engineering, 2021, 50(3): 25–28+33
[2] 陈立家, 卢学, 李世刚. 基于实际交通流的航海模拟器智能目标船系统[J]. 大连海事大学学报, 2020, 46(4): 8–16
[3] 杨琪森, 王慎执, 桑金楠, 等. 复杂开放水域下智能船舶路径规划与避障方法[J]. 计算机集成制造系统, 2022, 28(7): 2030–2040
[4] 严荣慧, 谢海成, 花敏恒, 等. 基于运动与表象特征的广域船舶目标识别方法[J]. 中国舰船研究, 2022, 17(1): 227−234.
YAN Ronghui, XIE Haicheng, HUA Minheng, et al. Wide-area ship target recognition method based on motion and appearance features[J]. Chinese Journal of Ship Research, 2022, 17(1): 227−234.
[5] 张照亿, 李颖, 董双, 等. 基于船舶领域模型的船舶碰撞危险识别方法[J]. 中国航海, 2021, 44(2): 1–7+14
ZHANG Zhaoyi, LI Ying, DONG Shuang, et al. Ship domain model in identification of collision risk[J]. Navigation of China, 2021, 44(2): 1–7+14
[6] 罗贤程, 尹建川, 李宏宾. 基于动态分阶势场法的船舶自动避碰系统[J]. 上海海事大学学报, 2020, 41(1): 1–7
LuO Xiancheng, YIN Jianchuan, LI Hongbin. An automatic collision avoidance system of ships based on dynamic hierarchical potential field method[J]. Journal of Shanghai Maritime University, 2020, 41(1): 1–7
