Active Localization of Multiple Mobile Robots By Moving to Best Frontiers

Shivudu Bhuvanagiri    K Madhava Krishna    Rakesh Goyal   

Robotics Research Center, IIIT Hyderabad, India   

We envisage a multi robotic scenario where several robots are in ambiguity about their states and require help of other robots to overcome their ambiguity. Ambiguity here is used in the sense of more than one hypothesis of a robot’s state. In such a scenario the method presented here moves the robots to locations where probability of eliminating several ambiguous states among multiple robots is a maximum. For this purpose the best frontiers are identified and robots dispatched to those frontiers. The best frontiers are those that have the highest probability of realizing a unique hypothesis if the robots were to arrive there. The method presented has been tested in both simulation and real-time on robots and its efficacy verified. Extensive comparative analysis portrays the advantage of the current method over others that do not perform active localization in a multi-robotic sense.