摘要
针对基于容积卡尔曼滤波(CKF)的无人水下航行器(UUV)同步定位与地图构建(SLAM)存在对模型参数变化的鲁棒性差、收敛速度慢、对突变状态跟踪能力低等问题,通过在CKF中引入渐消因子和弱化因子,提出了一种基于强跟踪CKF(STCKF)的SLAM算法(STCKF-SLAM)。首先建立UUV的运动模型、特征模型、测距声呐模型,然后基于霍夫变换从多测距声呐测量数据中提取堤岸线特征,最终采用STCKF实现了UUV的同步定位与地图构建。基于UUV海试数据的仿真实验结果表明:相比CKF-SLAM算法,STCKF-SLAM算法保持了对突变状态的强跟踪能力,且均方根误差降低了13%,提高了SLAM系统的精确性,可应用于UUV长航时水下隐蔽作业。
Aiming at the issues that the simultaneous localization and mapping （SLAM） algorithm for unmanned un- derwater vehicle （UUV） has the problem of low robustness to model parameter variation, slow convergence and unde- sirable tracking ability to abrupt state-changes ,a strong tracking cubature Kalman filter （STCKF） based SLAM algo- rithm （STCKF-SLAM） is proposed through introducing the fading factor and weakening factor in the CKF. Firstly, the motion model, feature model and ranging sonar model of UUV are set up. Then, the embankment line features are ex- tracted based on Hough transform from the multi-ranging sonar measurement data;and finally, the SLAM of UUV is realized with the STCKF. Simulation experiment results based on UUV sea trial data show that, compared with the CKF-SLAM algorithm, the STCKF-SLAM algorithm maintains the strong tracking ability to abrupt state-changes, the root-mean square error is reduced by 13% ;The STCKF-SLAM algorithm enhances the accuracy of SLAM system,and could be used in long-time underwater hiding work of UUV.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2013年第11期2542-2550,共9页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(E091002/50979017)
教育部高等学校博士学科点专项科研基金(20092304110008)
中央高校基本科研业务费专项资金(HEUCFZ1026)
哈尔滨市科技创新人才(优秀学科带头人)研究专项资金(2012RFXXG083)
教育部新世纪优秀人才支持计划(NCET-10-0053)资助项目
关键词
无人水下航行器
同步定位与地图构建
多测距声呐
霍夫变换
强跟踪容积卡尔曼滤波
unmanned underwater vehicle（UUV）
simultaneous localization and mapping（SLAM）
multi-ranging so-nar
Hough transform（HT）
strong tracking cubature Kalman filter（STCKF）
