Meta-Agent Conflict-Based Search For Optimal Multi-Agent Path Finding

Abstract

The task in the multi-agent path finding problem (MAPF) isto find paths for multiple agents, each with a different startand goal position, such that agents do not collide. It is possibleto solve this problem optimally with algorithms that arebased on the A* algorithm. Recently, we proposed an alternativealgorithm called Conflict-Based Search (CBS) (Sharonet al. 2012), which was shown to outperform the A*-basedalgorithms in some cases. CBS is a two-level algorithm. Atthe high level, a search is performed on a tree based on conflictsbetween agents. At the low level, a search is performedonly for a single agent at a time. While in some cases CBSis very efficient, in other cases it is worse than A*-based algorithms.This paper focuses on the latter case by generalizingCBS to Meta-Agent CBS (MA-CBS). The main idea isto couple groups of agents into meta-agents if the number ofinternal conflicts between them exceeds a given bound. MACBSacts as a framework that can run on top of any completeMAPF solver. We analyze our new approach and provideexperimental results demonstrating that it outperforms basicCBS and other A*-based optimal solvers in many cases.

Suboptimal Variants of the Conflict-Based Search Algorithm for the Multi-Agent Pathfinding Problem

“…In other domains, MSG1 was better, demonstrating that there is no universal winner. This is a known phenomenon for MAPF on optimal solvers too (Sharon et al 2013;2012a;2012b). Fully identifying which algorithm works best under what circumstances is a challenge for future work.…”

“…(1) The increasing cost tree search (ICTS) algorithm solves MAPF optimally by converting it into a set of fast-to-solve decision problems (Sharon et al 2013). (2) The conflict based search (CBS) (Sharon et al 2012a) and its extension meta-agent CBS (MA-CBS) (Sharon et al 2012b). CBS is the main focus of this paper and is detailed below.…”

On Merging Agents in Multi-Agent Pathfinding Algorithms

“…Meta-Agent Conflict-Based Search (MA-CBS) (Sharon et al 2012b) is an optimal algorithm that combines the split and merge operators. When a conflict is found between two agents, MA-CBS employs a policy that decides whether to use the merge or the split operators.…”

“…Therefore, the paths of agents inside the same meta-agent never conflict. The merge operator was directly used in Independence Detection (ID) (Standley 2010), Meta-Agent Conflict-Based Search (MA-CBS) (Sharon et al 2012b), and Improved Conflict-Based Search (Boyarski et al 2015), all optimal MAPF algorithms. But, as we will show, a different type of merge operator is also used in Priority-Based Search (Ma et al 2019), a suboptimal MAPF algorithm.…”

Recursive Constraint Manifold Subsearch for Multirobot Path Planning with Cooperative Tasks

“…Solving the CPP problem requires solving two qualitatively different problems: coordinating the simultaneous motion of many teams of robots (Wagner and Choset 2011) (Felner et al 2012) (Sharon et al 2012) (Standley 2010) and finding paths for the robots within a team that satisfy the constraints of the task being executed. While there has been a large amount of work in formation control (Belta and Kumar 2001) (Leonard and Fiorelli 2001) (Lewis and Tan 1997) (Spears and Gordon 1999) (Yang, Freeman, and Lynch 2008) and cooperative manipulation (Alonso-Mora et al 2015) (Desai and Kumar 1999) (Koga and Latombe 1994) (Rus, Donald, and Jennings 1995), little work has been done in the context of path planning with the dynamic formation and dissolution of teams.…”