A frontier based multi-robot approach for coverage of unknown environments

With the advent of latest technical advancements in the field of robotics, a stage has arrived where autonomous robots are expected to help humans in tasks that are either dangerous or too monotonous such as mining, search and rescue, floor cleaning. All these problems are derivatives of the coverage problem wherein the motto is to complete coverage of the environment in a time effective manner. Most of the coverage methods developed till date have access to the map prior to exploration and only few of them made use of multiple robots. In view of the drawbacks of the existing approaches, we develop a novel frontier based multi robot approach for coverage of unknown environments. In this work, multiple robots are employed to simultaneously explore and map the environment. Global map is computed by merging the individual maps of the robots. Frontiers which are the boundaries between explored and unexplored areas are identified and the robots are navigated toward frontiers using the proposed exploration strategy. Robot operating System (ROS) is used for implementation and Stage is used for simulating robots and their environments. Comparisons are made with existing approaches taking into consideration of time to explore, percentage of area explored. Results demonstrate the efficiency and effectiveness of our approach.