Parallel Motion Execution and Path Rerouting for a Team of Mobile Robots

Abstract

This work achieves parallel movements and collision free paths for a team of identical robots evolving in a known environment while satisfying a global Boolean-based formula over a set of regions of interest. The movement capabilities of the robots within the grid-based environment are modeled as a Petri net system. The solution starts from initially known robot paths given by the approach in (Mahulea et al., 2020b). The main advantages of this work are two-folded: (i) it reduces the number of waiting states of the robots throughout their paths, (ii) it considers path rerouting action solved by a MILP problem, without generating new observations along their paths. The rerouting is triggered when the number of robots that are waiting is greater than a threshold, or when some stopped robots inhibit the movement of others. Thus, the proposed method yields parallel movement of robots and path rerouting when needed. The algorithm is integrated in the open-source toolbox RMTool (González et al., 2015).