设计基于机器视觉技术的船舶智能导航系统，有效规划船舶路径，避开障碍物，实现船舶智能导航。该系统利用CCD摄像头拍摄船舶航行环境信息，利用机器视觉图像采集模块采集并处理环境信息，提取导航需要的信息；核心控制器利用基于小生境粒子群优化的多路径规划算法，设计导航路径，并传输至控制电机驱动模块与舵机控制模块；舵机控制模块接收导航信息后，精准调整舵机角度，控制船舶航行方向。通过电机驱动模块为船舶电机提供可靠的驱动电路，令船舶按照指定速度航行；利用电源模块为整个系统各模块提供可靠电源电压；无线通信模块依据控制器局域网络总线，为整个系统提供信息传输服务。实验证明：在不同障碍物数量时，该系统均可有效规划多条船舶航行路径，实现船舶智能导航；在不同发送速率与波特率时，该系统均可成功传输信息，且信息传输完整性较优。应用该系统后，可有效降低船舶舵角，提升船舶航行稳定性。
A ship intelligent navigation system based on machine vision technology is designed to effectively plan the ship's path, avoid obstacles and realize intelligent navigation of the ship. The system uses CCD camera to capture ship navigation environment information. The machine vision image acquisition module is used to collect and process environmental information and extract navigation information. The core controller designs the navigation path by using the multi-path planning algorithm based on niche particle swarm optimization and transmits it to the control motor drive module and steering gear control module. After receiving the navigation information, the steering gear control module accurately adjusts the steering gear Angle to control the navigation direction of the ship. Through the motor drive module for the ship motor to provide reliable driving circuit, so that the ship according to the specified speed sailing; Use the power module to provide reliable power supply voltage for each module of the whole system; The wireless communication module provides information transmission service for the whole system according to the controller LAN bus. Experimental results show that the system can effectively plan multiple navigation paths and realize intelligent navigation of ships under different number of obstacles. At different transmission rates and baud rates, the system can successfully transmit information, and the integrity of information transmission is better. After the application of the system, the rudder angle of the ship can be effectively reduced and the navigation stability of the ship can be improved.
2022,44(10): 111-114 收稿日期：2021-11-26
DOI：10.3404/j.issn.1672-7649.2022.10.022
分类号：TP391
基金项目：江苏省本科生创新项目(202110289044Z)
作者简介：谢董正(2000-)，男，研究方向为海洋工程与技术
参考文献：
[1] 陈志超, 薛圻蒙, 罗凯, 等. 基于改进LOS导航算法的智能船舶自动靠泊控制研究[J]. 中国航海, 2021, 44(2): 126–133
[2] 章怀宇, 陈熙源, 王健. 面向深海无人潜航器的导航系统设计与实现[J]. 船舶工程, 2019, 41(5): 121–127
[3] 张莉. 基于Qt的船用CCD星敏感器导航系统仿真研究[J]. 船舶工程, 2021, 43(5): 114–120+128
[4] 许威, 胡勇, 陈兴怖. 基于机器视觉的船体双曲度板测量方法[J]. 船舶工程, 2019, 41(5): 101–106
[5] 肖长诗, 陈芊芊, 文元桥, 等. 基于LoRa的封闭水域船舶自动识别系统设计[J]. 船海工程, 2019, 48(1): 158–162+167
[6] 祝亢, 黄珍, 王绪明. 基于深度强化学习的智能船舶航迹跟踪控制[J]. 中国舰船研究, 2021, 16(1): 105–113
[7] 李明峰, 王胜正, 谢宗轩. 恶劣气象海况下船舶航线的多变量多目标优化建模[J]. 中国航海, 2020, 43(2): 14–19+30
