Despite the increased interest in multi-agent planning, one question has remained largely unaddressed: “under what conditions are multiple agents actually needed to solve a planning problem?”
It is well understood that, through cooperation, multiple agents can achieve tasks that are unachievable by a single agent. However, there are no formal characterizations of situations where cooperation is required to achieve a goal, thus warranting the application of multiple agents. In this work, we provide such a formal characterization for multi-agent planning problems with sequential action execution. We first show that determining whether there is required cooperation (RC) is in general intractable even in this limited setting. As a result, we start our analysis with a subset of more restrictive problems where agents are homogeneous. For such problems, we identify two conditions that can cause RC. We establish that when none of these conditions hold, the problem is single-agent solvable; otherwise, we provide upper bounds on the minimum number of agents required. For the remaining problems with heterogeneous agents, we further divide them into two subsets. For one of the subsets, we propose the concept of transformer agent to reduce the number of agents to be considered which is used to improve planning performance. We implemented a planner using our theoretical results and compared it with one of the best IPC CoDMAP planners in the centralized track. Results show that our planner provides significantly improved performance on IPC CoDMAP domains.
Distributed Multi-agent Pathfinding
The task here is to address the multi-agent pathfinding problem with distributed systems that are subject to limited sensing and communication range.
Cooperative pathfinding is often addressed in one of two ways in the literature. In fully coupled approaches, robots are considered together and the plans for all robots are constructed simultaneously. In decoupled approaches, the plans are constructed only for a subset of robots at a time. While decoupled approaches can be much faster than fully coupled approaches, they are often suboptimal and incomplete. Although there exist a few decoupled approaches that achieve completeness, global information (which makes global coordination possible) is assumed. Global information may not be accessible in distributed robotic systems. In this paper, we provide a window-based approach to cooperative pathfinding with limited sensing and communication range in distributed systems (called DisCoF). In DisCoF, robots are assumed to be fully decoupled initially, and may gradually increase the level of coupling in an online and distributed fashion. In some cases, e.g., when global information is needed to solve the problem instance, DisCoF would eventually couple all robots together. DisCoF represents an inherently online approach since robots may only be aware of a subset of robots in the environment at any given point of time. Hence, they do not have enough information to determine non-conflicting plans with all the other robots. Completeness analysis of DisCoF is provided.