论文部分内容阅读
针对多机器人协同SLAM(同步定位与地图构建)的地图融合中,由于通信距离受限或网络拓扑变化造成信息缺失、从而影响全局地图构建的问题,提出一种基于信息增益一致性原理的动态地图融合算法.该算法是完全分布式的,且不依赖于任何特殊的机器人通信网络结构.该算法利用机器人所测局部地图的历史数据和当前数据之间的新增信息,使每个机器人都能同步地获取一致的、最新的全局地图.在有限的网络连接条件下,所提出的地图融合算法能够通过渐近收敛的方式获得准确的全局地图.在每一次迭代中,每个机器人得到的全局地图都是无偏的.在实验中通过实际环境的RGB-D(彩色-深度)数据验证了算法的有效性.
To solve the problem of global map building due to the limitation of communication distance or the change of network topology caused by the multi-robot collaborative SLAM (Simultaneous Localization and Map Construction) map fusion, a dynamic map based on the consistency of information gain is proposed The algorithm is completely distributed and does not depend on any special robot communication network structure.The algorithm makes use of the historical information of the local map measured by the robot and the new information between the current data so that each robot can Simultaneous acquisition of a consistent and up-to-date global map The proposed map fusion algorithm can obtain an accurate global map by means of asymptotic convergence under the condition of limited network connection.In each iteration, the global The maps are unbiased, and the validity of the algorithm is verified through real-world RGB-D (color-depth) data.