SLASEM中的基于广义距离的数据关联

Data Association Using General Distance in SLASEM

介绍了一种同时定位与采样环境地图创建(SLASEM)中的数据关联方法.针对地图中的采样环境粒子与实际环境之间没有一一对应性、传统的马氏距离不能描述SLASEM中两个物体之间相似性的特点,提出了两个点集合之间的一种广义距离函数,并利用该距离函数进行数据关联.同时,提出了基于环境拓扑结构的方法解决多重数据关联问题,并且利用前一时刻的数据关联的结果辅助当前时刻的数据关联.最后,用两个室内环境的实验验证了所提算法的有效性.

A method of data association in SLASEM（simultaneous localization and sampled environment mapping） is presented.According to the features that environment samples in the map have no one-to-one correspondence with the real environment and the traditional Mahalanobis distance cannot describe the similarity of two objects in SLASEM,a general distance function between two point sets is proposed for data association.The multi-association problem is solved by utilizing the topological structure of the environment.The result of data association in previous instant is also used to aid the data association in current instant.Results of two indoor experiments validate the effectiveness of the proposed algorithm.