Building Map and Locating System for Indoor Robot Based on Player
In this paper, we have proposed a method for building an indoor service robot map and a positioning system based on the open source platform of Player. Firstly, the DP-SLAM algorithm is transplanted to the Player and builds dynamic offline maps, in order to reduce the errors and constraints caused by manual map building. Secondly, we introduce the KLD-Sampling Adaptive Monte Carlo locating (KLD-AMCL) algorithm to adjust the number of particles required adaptively by calculating the MLE and the real posterior KL distance, to get higher accuracy of localization. Finally, an indoor service robot positioning system is built by combining the Player platform, dynamic map building and KLD-AMCL algorithm. Empirical results show that this system has better environmental adaptability and higher positioning accuracy.
Player Indoor Robot Locating Map Building DP-SLAM KLD-AMCL
Bin Wang Wei Lu Bin Kong
University of Science and Technology of China, Hefei 230027, China Institute of Intelligent Machines, Chinese Academy of Science, Hefei 230031, China
国内会议
合肥
英文
1-6
2011-10-15（万方平台首次上网日期，不代表论文的发表时间）