利用FastSLAM框架的多自治水下航行器同时定位与跟踪算法
Simultaneous localization and tracking algorithm utilizing FastSLAM framework for autonomous underwater vehicles
摘要点击 252  全文点击 221  投稿时间:2018-09-28  修订日期:2019-04-18
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2019.80747
  2020,37(1):89-97
中文关键词  协同定位  自治水下航行器  同时定位与制图  同时定位与跟踪  声呐  传感器  误差  仿真
英文关键词  cooperative localization  autonomous underwater vehicles  simultaneous localization and mapping  simultaneous localization and tracking  sonar  sensors  errors  simulation
基金项目  国家自然科学基金项目(51607133);陕西省教育厅专项科学研究计划项目(17JK0332);陕西省科技厅科技发展计划项目(2011K06-01)
作者单位E-mail
卢健 西安工程大学 406170365@qq.com 
陈旭 西安工程大学 273662018@qq.com 
刘通 西安工程大学  
马成贤 西安工程大学  
何金鑫 西安工程大学  
中文摘要
      协同定位是共融机器人研究领域的重要问题. 协同定位方案的制定受限于机器人间信息交互的能力. 针对长时间通讯中断时多自治水下航行器(AUV)协同定位精度明显下降的问题, 借鉴同时定位与制图(SLAM)方法, 提出了基于FastSLAM框架的同时定位与跟踪(SLAT)算法. 将主AUV视为非合作目标, 在从AUV上建立起一个关于主AUV的运动估计器, 利用从AUV上声呐传感器实时获取的相对量测信息, 在对主AUV运动状态估计的同时, 完成对从AUV自定位精度的提升. 仿真实验结果表明, 在长时间通讯中断发生的条件约束下, 相比于传统的航位推算方法, 所提出的SLATF1.0和2.0算法能够有效减小定位误差, 2.0算法对于探测精度变化等因素的影响具有更好适应性.
英文摘要
      The cooperative localization is an important research question in the field of Tri-Co Robots study. The scheme of the cooperative localization algorithm depends on the ability of information interaction between the robots. To solve the problem that the cooperative localization accuracy is obviously reduced when the communication is interrupted for a long time between the autonomous underwater vehicles(AUV), the simultaneous localization and tracking(SLAT) algorithms based on the FastSLAM framework are developed in this research, borrowing the principle of the simultaneous localization and mapping(SLAM) algorithms. The master AUV is regarded as a non-cooperative target and a motion estimator used to track the master AUV is built in the slaver AUV. When the motion state of the master AUV is estimated, the improvement of the self localization accuracy of the slaver AUV is achieved, using the relative measurement information obtained from the sonar sensor on the slaver AUV in real time. The simulation experimental results show that the proposed SLATF1.0 and 2.0 algorithms can effectively reduce the localization errors compared to the conventional dead reckoning method under the condition of long-term communication interruption, and the 2.0 algorithm has better adaptability to the influence of the detection accuracy variety.