Self-adaptive monte carlo for single-robot and multi-robot localization

Abstract

In order to achieve the autonomy of mobile robots, effective localization is a necessary prerequisite. In this paper, we propose an improved Monte Carlo localization using self-adaptive samples, abbreviated as SAMCL. This algorithm is able to solve the multi-robot localization problem as well as the single-robot localization problem. By employing a pre-caching technique to reduce the on-line computational burden, SAMCL is more efficient than regular MCL. We define the concept of Similar Energy Region (SER), which is a set of grid cells having similar energy with the robot in the robot space. By distributing global samples in SER instead of distributing randomly in the map, SAMCL obtains a better performance in localization. Thanks to self-adaptive samples that can automatically separate themselves into a global sample set and a local sample set according to need, SAMCL can solve position tracking, global localization and the kidnapped robot problem together. SAMCL can be extended to handle multi-robot localization through a Position Mapping (PM) algorithm. This algorithm enables one robot to calculate its possible positions according to positions of other robots and mutual relations between each other. The validity and the efficiency of our algorithm are demonstrated by experiments carried out with different intentions. Extensive experiment results are also given in this paper.