A Multi-model Distributed Controller for Swarm Navigation: Obstacle Clearance Strategy

This paper considers a multi-model Distributed Controller for swarm Navigation using the obstacle Clearance strategy (DCNC). If there is an obstacle blocking the swarm’s path, some agents are collectively assigned to push the obstacle away from its original position creating free space for the rest of the swarm navigating towards the destination. To clear the obstacle, we developed a distributed obstacle-pushing strategy consisting of a self-identified push section on an obstacle’s boundary and a collective pushing force control. Our pushing strategy ensures the uniform distribution of robots on the self-identified push section, resulting in the maximum pushing force over the pushing direction.