【摘要】 针对移动机器人同时定位与地图创建(SLAM)系统在复杂特殊环境中应用时,单一传感器存在漏检且难以解决数据关联的问题,在多维匹配模型的基础上,提出一种多传感器数据融合模型,将数据关联问题转化为数学规划中的约束组合优化问题进行求解。实验结果表明,所提出的方法与联合相容分支定界算法(JCBB)的关联结果相当,且降低了计算代价。更多还原

【Abstract】 When robots locate themselves in complex and hazardous environments,the simultaneous localization and mapping(SLAM) system with a single sensor might overlook some features which will lead to difficult data association.A multi-sensor based data fusion model for robotic SLAM is proposed to cope with environmental complexity by solving data association problems through combinatorial optimization on multiple data sets.Experimental results on real-world data show that our method produces data association as accurate as the joint compatibility brance and bound(JCBB) method but with reduced computing cost.更多还原