Lifelong Multi-Agent Path Finding in Large-Scale Warehouses

Abstract

Multi-Agent Path Finding (MAPF) is the problem of moving a team of agents to their goal locations without collisions. In this paper, we study the lifelong variant of MAPF, where agents are constantly engaged with new goal locations, such as in large-scale automated warehouses. We propose a new framework Rolling-Horizon Collision Resolution (RHCR) for solving lifelong MAPF by decomposing the problem into a sequence of Windowed MAPF instances, where a Windowed MAPF solver resolves collisions among the paths of the agents only within a bounded time horizon and ignores collisions beyond it. RHCR is particularly well suited to generating pliable plans that adapt to continually arriving new goal locations. We empirically evaluate RHCR with a variety of MAPF solvers and show that it can produce high-quality solutions for up to 1,000 agents (= 38.9% of the empty cells on the map) for simulated warehouse instances, significantly outperforming existing work.

Optimal and Bounded-Suboptimal Multi-Goal Task Assignment and Path Finding

“…Multi-Label Space-Time A* (MLA*) finds a time-optimal path for an agent a i (a path with the smallest finish time T i ) that visits all goal locations of its assigned task g j in sequence and obeys a set of constraints. MLA* was first introduced for two goal locations [31] and then extended to more than two goal locations [37]. MLA* extends Space-Time A* [38] by adding a label indicating the different segments between the goal locations, where label k indicates that the next goal location to visit is g j [k].…”

Multi-Goal Multi-Agent Pickup and Delivery

“…Multi-Label A* (MLA*) [3] was invented for planning paths for pairs of goal locations, namely the pickup location and the delivery location of a task. Li et al [7] generalize MLA* for planning paths for longer sequences of goal locations.…”

“…In LNS-PBS and LNS-wPBS, each agent maintains (1) a dummy endpoint, i.e., an endpoint that it can move to and stay indefinitely at without collisions (initially, this dummy endpoint is its start location), (2) a task sequence, that consists of the uncompleted tasks that it has to execute, (3) a corresponding goal sequence, that consists of all goal locations of the tasks in its task sequence plus its dummy endpoint at the end, and (4) a path, that moves the agent from its current location through all locations in its goal sequence without collisions. Algorithm 1 without the blue parts (i.e., Lines [6][7][8]) shows how LNS-PBS works. Many of its steps (not shown in the pseudo-code but introduced later), including the use of dummy endpoints, the strategy of which unexecuted tasks can be assigned to agents, and the modification of PBS, are designed to ensure its completeness.…”

“…However, these algorithms solve only one-shot problems where each agent has only one task, and their scalability is limited. Li et al [7] propose the efficient lifelong MAPF algorithm Rolling-Horizon Collision Resolution (RHCR) for planning collision-free paths for a sequence of goal locations. It uses a rolling-horizon framework that repeatedly calls a windowed MAPF algorithm to resolve collisions for only a few timesteps ahead.…”

“…tasks) (seq. goals) (well-formed) CENTRAL [1] TA-Hybrid [4] HBH+MLA* [3] RMCA [5] (SMT-)HCBS [6] N/A RHCR [7] LNS-PBS LNS-wPBS window for the windowed MAPF algorithm. Empirically, LNS-PBS and LNS-wPBS often yield smaller service times than state-of-the-art MAPD algorithms, and LNS-wPBS scales to thousands of agents and thousands of tasks in a large warehouse.…”