Augmented Costmap-Based Path Planning and Control for Multi-Mobile Robot Rigid Formation

This work proposes a novel method to generate an environment costmap to plan a feasible path for multiple autonomous mobile robots in a rigid formation using a leader-follower approach and a regulation algorithm to control the mobile robots during navigation. The method is based on a costmap merging technique in which the involved robots’ two-dimensional costmaps are merged into a three-dimensional one that embeds the environment description and the rigid structure of the multi-robot formation by means of sampling the leader’s rotations on the additional dimension of the generated costmap. The resulting three-dimensional grid is then employed to plan an optimal path for the robots’ formation, using a graph-based path-finding algorithm such as A^∗ or Theta^∗. This enables the whole planning task to be pursued only with respect to the leader while generating feasible trajectories also for followers. Finally, the method includes a control method to regulate the following mobile platforms in the nominal configuration during the navigation. The devised model is tested in different simulated scenarios and on a real setup.