# GRACE: A Unified 2D Multi-Robot Path Planning Simulator & Benchmark for Grid, Roadmap, And Continuous Environments

arXiv: https://arxiv.org/abs/2603.10858 · HTML: https://mlsanity.com/paper/2603.10858/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 56 references that are in the map.
- First posted on arXiv 11 Mar 2026.

## Builds on (most cited first, 26 of 26)

- [Multi-Agent Pathfinding: Definitions, Variants, and Benchmarks](https://mlsanity.com/paper/1906.08291/) (2019)
- [PRIMAL: Pathfinding via Reinforcement and Imitation Multi-Agent Learning](https://mlsanity.com/paper/1809.03531/) (2018)
- [EECBS: A Bounded-Suboptimal Search for Multi-Agent Path Finding](https://mlsanity.com/paper/2010.01367/) (2020)
- [SCRIMP: Scalable Communication for Reinforcement- and Imitation-Learning-Based Multi-Agent Pathfinding](https://mlsanity.com/paper/2303.00605/) (2023)
- [Flatland-RL : Multi-Agent Reinforcement Learning on Trains](https://mlsanity.com/paper/2012.05893/) (2020)
- [Pairwise Symmetry Reasoning for Multi-Agent Path Finding Search](https://mlsanity.com/paper/2103.07116/) (2021)
- [POGEMA: A Benchmark Platform for Cooperative Multi-Agent Pathfinding](https://mlsanity.com/paper/2407.14931/) (2024)
- [Asymptotically Optimal Sampling-based Kinodynamic Planning](https://mlsanity.com/paper/1407.2896/) (2014)
- [Representation-Optimal Multi-Robot Motion Planning using Conflict-Based Search](https://mlsanity.com/paper/1909.13352/) (2019)
- [dRRT*: Scalable and Informed Asymptotically-Optimal Multi-Robot Motion Planning](https://mlsanity.com/paper/1903.00994/) (2019)
- [Improving Learnt Local MAPF Policies with Heuristic Search](https://mlsanity.com/paper/2403.20300/) (2024)
- [Experimenting with robotic intra-logistics domains](https://mlsanity.com/paper/1804.10247/) (2018)
- [Conflict-based Search for Multi-Robot Motion Planning with Kinodynamic Constraints](https://mlsanity.com/paper/2207.00576/) (2022)
- [LNS2+RL: Combining Multi-Agent Reinforcement Learning with Large Neighborhood Search in Multi-Agent Path Finding](https://mlsanity.com/paper/2405.17794/) (2024)
- [Quick Multi-Robot Motion Planning by Combining Sampling and Search](https://mlsanity.com/paper/2203.00315/) (2022)
- [db-CBS: Discontinuity-Bounded Conflict-Based Search for Multi-Robot Kinodynamic Motion Planning](https://mlsanity.com/paper/2309.16445/) (2023)
- [Advancing Learnable Multi-Agent Pathfinding Solvers with Active Fine-Tuning](https://mlsanity.com/paper/2506.23793/) (2025)
- [Anytime Multi-Agent Path Finding using Operation Parallelism in Large Neighborhood Search](https://mlsanity.com/paper/2402.01961/) (2024)
- [HiMAP: Learning Heuristics-Informed Policies for Large-Scale Multi-Agent Pathfinding](https://mlsanity.com/paper/2402.15546/) (2024)
- [Lightweight and Effective Preference Construction in PIBT for Large-Scale Multi-Agent Pathfinding](https://mlsanity.com/paper/2505.12623/) (2025)
- [Near-Optimal Multi-Robot Motion Planning with Finite Sampling](https://mlsanity.com/paper/2011.08944/) (2020)
- [Path Planning with Kinematic Constraints for Robot Groups](https://mlsanity.com/paper/1704.07538/) (2017)
- [Conflict-Based Search for Explainable Multi-Agent Path Finding](https://mlsanity.com/paper/2202.09930/) (2022)
- [Clique Analysis and Bypassing in Continuous-Time Conflict-Based Search](https://mlsanity.com/paper/2312.16106/) (2023)
- [CBS with Continuous-Time Revisit](https://mlsanity.com/paper/2501.07744/) (2025)
- [SkyRover: A Modular Simulator for Cross-Domain Pathfinding](https://mlsanity.com/paper/2502.08969/) (2025)

## Abstract (arXiv)

Advancing Multi-Agent Pathfinding (MAPF) and Multi-Robot Motion Planning (MRMP) requires platforms that enable transparent, reproducible comparisons across modeling choices. Existing tools either scale under simplifying assumptions (grids, homogeneous agents) or offer higher fidelity with less comparable instrumentation. We present GRACE, a unified 2D simulator+benchmark that instantiates the same task at multiple abstraction levels (grid, roadmap, continuous) via explicit, reproducible operators and a common evaluation protocol. Our empirical results on public maps and representative planners enable commensurate comparisons on a shared instance set. Furthermore, we quantify the expected representation-fidelity trade-offs (MRMP solves instances at higher fidelity but lower speed, while grid/roadmap planners scale farther). By consolidating representation, execution, and evaluation, GRACE thereby aims to make cross-representation studies more comparable and provides a means to advance multi-robot planning research and its translation to practice.