基于高斯混合模型最大期望聚类的同时定位与地图构建数据关联

Simultaneous localization and mapping data association based on maximum expectation clustering for Gaussian mixture model

中文关键词  同时定位与地图构建  数据关联  联合兼容分支定界  高斯混合模型  最大期望聚类  移动机器人

英文关键词  simultaneous localization and mapping  data association  joint compatibility branch and bound  gaussian mixture model  Expectation-Maximization clustering  mobile robot

中文摘要

      数据关联是移动机器人同时定位与地图构建中状态估计的前提和基础, 针对当前联合兼容分支定界算法存在计算复杂高, 耗时长的问题, 提出了基于高斯混合模型(GMM)最大期望聚类分组的SLAM数据关联算法. 首先,为减少同一时刻参与关联的观测值数目，在局部区域内采用GMM最大期望聚类算法对当前时刻的观测值进行分组; 其次, 在各观测小组中采用联合兼容分支定界算法进行数据关联; 最后, 综合各观测小组的观测值同局部地图特征得到的关联解, 得到最优的关联结果. 仿真实验结果表明, 基于高斯混合模型最大期望聚类分组的SLAM数据关联算法在保证数据关联准确度的前提下, 计算复杂度得到了降低, 缩短了运行时间。

英文摘要

      Data association is the premise and basis of state estimation of mobile robot simultaneous localization and mapping(SLAM). In order to solve the problem of complex and time-consuming computation of joint compatiblity branch and bound algorithm, a SLAM data association algorithm based on gaussian mixture model(GMM) maximum expectation(EM) clustering is proposed. Firstly, in order to reduce the number of observations participating in the association at the same time, group the current measurement using maximum expectation clustering algorithm for gaussian mixture model in the local region. Secondly, conduct data association using joint compatibility branch and bound algorithm for each group. Finally, obtain the optimal correlation result by combining the correlation result between each observation group and the local map features.The simulation results show that the SLAM data association algorithm based on gaussian mixture model maximum expectation clustering guarantees the accuracy of data association, the computational complexity of this method is reduced and the running time is shortened.