The core message of this article is that when a multi-agent path finding plan becomes invalid due to unexpected delays, the plan can be efficiently repaired by introducing additional delays to some agents, rather than replanning from scratch.
ITA-ECBS, a new bounded-suboptimal algorithm, efficiently solves the Combined Target Assignment and Path Finding (TAPF) problem by incorporating an additional lower bound matrix and employing the shortest path algorithm to accelerate the focal search.