Date of Award


Publication Type

Master Thesis

Degree Name



Computer Science


multi-robot, localization, Monte Carlo, belief, entropy, density estimation


Wu, Dan




Cooperative multi-robot localization techniques use sensor measurements to estimate poses (locations, orientations) of robots relative to a given map of the environment. Existing approaches update a robot's pose instantly whenever it detects another robot. However, such instant update may not be always necessary and effective, since both robots' pose estimates could be highly uncertain at the time of the detection. In this thesis, we develop a new information exchange mechanism to collaborative multi-robot localization. We also propose a new scheme to calculate how much information is contained in a robot's belief by using entropy. Instead of updating beliefs whenever detection occurs, our approach first compares the beliefs of the robots which are involved in the detection, and then decide whether the information exchange is necessary. Therefore, it avoids unnecessary information exchange whenever one robot perceives another robot. On the other hand, this approach does allow information exchange between detecting robots and such information exchange always contributes positively to the localization process, hence, improving the effectiveness and efficiency of multi-robot localization. The technique has been implemented and tested using two mobile robots as well as simulations. The results indicate significant improvements in localization speed and accuracy when compared to the single mobile robot localization.