摘要
未知复杂环境中不规则的障碍物使传感数据具有不确定性,单依靠激光测距仪进行移动机器人的定位与自主导航可靠性不足;针对此问题,提出了一种基于激光测距仪和双目视觉传感器信息融合的精确定位方法,利用加权最小二乘拟合方法和尺度不变特征变换(SIFT)算法分别从对激光信息与视觉信息中提取直线和点特征,进行特征级的信息融合。通过对实验结果和数据分析,多传感器信息融合可以有效提高移动机器人SLAM(即时定位与地图构建)的精度和鲁棒性。
Relying solely on laser range finder make the localization and autonomous navigation of mobile robot highly uncertainty due to the irregular obstacles in unknown complex environment, This paper studied the accurate location of the method based on laser range finder and binocular vision sensor information fusion, extracting straight lines and some points features by using the weighted least squares fitting method and SIFT （Scale Invariant Feature Transform）. The data fusion is performed at the level of features. The experiment results and da ta analysis show that multi -- sensor fusion is an efficient way to improve the precision and robustness of mobile robot SLAM （Simultaneous Localization and Mapping）.
出处
《计算机测量与控制》
北大核心
2013年第1期180-183,共4页
Computer Measurement &Control
基金
国家自然科学基金项目(61075087)
湖北省科技计划自然科学基金重点项目(2010CDA005)
关键词
移动机器人
即时定位与地图构建
信息融合
尺度不变特征变换
mobile robot~ simultaneous localization and mapping~ data fusion
scale invariant feature transform
