摘要
同时定位与地图构建是机器人在未知环境中自主导航的一个重要研究内容。为了提高全协方差SLAM算法的计算和存储效率,通过对系统状态向量进行重构和选择适当的观测向量,系统模型和系统协方差矩阵可以表示成特殊形式的矩阵。基于这两个矩阵的特性,提出了一个改进的SLAM方法,对机器人的位置和方向进行间接地估计,同时降低了SLAM算法的时间复杂度和空间复杂度。实验表明改进算法是一致的和收敛的。
Simultaneous localization and map building（SLAM） is an important research topic in the field of intelligent mobile robot.It is necessary for a mobile robot to navigate in an unknown environment.In order to improve the computational efficiency of the full covariance SLAM solution,the system state vector is reconstructed and the observation vector is selected properly.The system motion model,system observation model,system augmentation model and system covariance matrix can be respectively represented with a kind of special matrix form.An improved SLAM method is proposed based on the property of special matrix.The position and orientation of mobile robot is estimated indirectly.The computational requirement and memory requirement are decreased by half.The experimental results indicate that the proposed method is consistent and convergent.
出处
《火力与指挥控制》
CSCD
北大核心
2010年第12期134-137,140,共5页
Fire Control & Command Control
基金
江苏省"青蓝工程"基金
常州市青年人才基金(CQ2007005)
江苏教育自然基金资助项目(08KJD520002)
