摘要
针对多无人艇协同避障软件系统的需求,首先,提出多无人艇协同避障软件系统总体框架,并将总体框架划分为多无人艇层、岸基层和云系统层3个层级;然后,详细设计岸基层的岸基监控软件的数据处理与显示、自主航行路径规划模块;接着,建立Hadoop数据采集2层Flume架构,并基于HBase实现Hadoop数据存储,基于LSTM实现Hadoop避障预测;最后,验证该软件系统可满足多无人艇协同避障实际需求。
According to the requirements of multi-USV cooperative obstacle avoidance software system,firstly,the overall framework of multi-USV cooperative obstacle avoidance software system is proposed,and the overall framework is divided into three levels:multi-USV layer,shore-based layer and cloud system layer;secondly,the data processing and display module,autonomous navigation path planning module of shore-based monitoring software of shore-based layer are designed in detail;then the two-layer Flume architecture for Hadoop data acquisition is established,and Hadoop data storage is realized based on HBase,and Hadoop obstacle avoidance prediction is realized based on LSTM;the developed software system can meet the actual needs of multi-USV cooperative obstacle avoidance.
作者
子文江
庄伟涛
彭荣发
饶银辉
陈远明
洪晓斌
Zi Wenjiang;Zhuang Weitao;Peng Rongfa;Rao Yinhui;Chen Yuanming;Hong Xiaobin(School of Mechanical and Automotive Engineering,South China University of Technology,Guangzhou 510641,China;Guangzhou Shipyard International Company Limited,Guangzhou 511462,China)
出处
《自动化与信息工程》
2021年第1期12-17,共6页
Automation & Information Engineering
基金
广东省科技计划项目(2019B151502057,2018A030313209)
广东省自然资源厅科技项目(GDoE[2019]A13)
广州市科技计划项目(201902010024)。
关键词
无人艇
协同避障
HADOOP
软件设计
unmanned surface vehicle
cooperative obstacle avoidance
Hadoop
software design