Autonomous navigation strategy for robot swarms using local communication

Our motivation focuses on answering a simple question: What is the minimum robotic struc­ture necessary to solve a navigation problem? Our research deals with environments that are unknown, dynamic, and denied to sensors. In particular, the paper addresses problems concerning how to coordinate the navigation of multi­ple autonomous mobile robots without requiring system identification, geometric map building, localization or state estimation. The proposed navigation algorithm uses the gradient of the environment to set the navigation control. This gradient is continuously modified by all the ro­bots in the form of local communication. The design scheme, both for the algorithm and for its implementation on robots, searches for a mi­nimal approximation, in which it minimizes the requirements of the robot (processing power, communication and kind of sensors). Besides, our research finds autonomous navigation for each robot, and also scales the system to any number of agents. The navigation algorithm is formulated for a grouping task, where the robots form autonomous groups without any external interaction or prior information of the environ­ment or information from other robots. Finally, task performance is verified through simulation for the laboratory prototypes of the group