Multi-agent Collaborative Localization
Multi-agent Collaborative Localization(MACL) is used in multi-agent localization system to optimize robot localization accuracy. The main idea is sharing information between robots. Specifically, when robot 0 detects robot 1, robot 0 estimates robot 1’s position and send its belief to robot 1. Robot 1 receives the message from robot 0 and combines it with its own belief. By fusing two position estimates, robot 1 gets better estimate of its position.
More introduction is: http://www-users.cselabs.umn.edu/~moxxx066/MACL_2015.pdf