随着运行轨迹的增加，submap的数量逐渐增多，搜索空间也越来越大，直

With the increase of the running trajectory, the number of submaps gradually increases, and the search space also becomes larger and larger. Searching directly in the search space will cause the performance of the algorithm to deteriorate seriously and cannot meet real-time performance. Therefore, the principle of branch and bound [38] is used to improve the search space data structure, so as to efficiently find t.

With the increase of the running track, the number of submap gradually increases, the search space is also larger and larger, search directly in the search space will lead to the algorithm performance degradation is serious, can not meet the real-time. Therefore, the search spatial data structure is improved by using the principle of branch boundary.

With the increase of the trajectory, the number of submaps increases gradually, and the search space is also growing. Searching directly in the search space will lead to serious performance degradation of the algorithm, which can not meet the real-time performance. Therefore, we use the principle of branch and bound [38] to improve the search spatial data structure, so as to obtain t efficiently.<br>