Multi-Agent Pathfinding: Definitions, Variants, and Benchmarks

Stern, Roni; Sturtevant, Nathan R.; Felner, Ariel; Koenig, Sven; Ma, Hang; Walker, Thayne T.; Li, Jiaoyang; Atzmon, Dor; Cohen, Liron; Kumar, T. K. Satish; Boyarski, Eli; Barták, Roman

doi:10.48550/arxiv.1906.08291

Cited by 15 publications

(26 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All the algorithms were implemented in Python and tested on a computer with an Intel Core i7 CPU and 16 GB RAM. We selected maps (grids) from [19] and generated an undirected graph by making each grid four-connected. The run time limit for each test instance is five minutes.…”

Section: Numerical Resultsmentioning

confidence: 99%

“…Note that when B = 0, MA-CBS is the same as LS-rM* since all agents are always merged as one meta-agent, and when B = ∞, MA-CBS is the same as CCBS since all agents are never merged. We select three grids from different categories (room, maze, game map) from [19] and report the success rates of finding a solution within the time limit, as well as average run times for a different number of agents N ∈ {2, 4, 8, 12, 16, 20}. The best performance for each N is highlighted in bold text in the table.…”

Section: B Meta-agent Conflict-based Searchmentioning

confidence: 99%

“…Multi-agent path finding (MAPF), as its name suggests, determines an ensemble of collision-free paths for multiple agents between their respective start and goal locations. Most MAPF methods [19] describe the workspace with a graph, where vertices represent possible locations of agents and edges are actions that move agents between locations. Conventional MAPF planners [6], [16], [17], [8], including our own [21] typically consider the case where each agent executes an action concurrently in that all agents start and stop together.…”

Section: Introductionmentioning

confidence: 99%

“…We also fuse LS-rM* with Meta-agent Conflict-based Search (MA-CBS) [15], which is an algorithm that combines both Conflict-based Search (CBS) [16] and A*-based planners to achieve better performance in various maps. We test the algorithms in maps from [19] and our results (Sec. VII) show the following:…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Loosely Synchronized Search for Multi-agent Path Finding with Asynchronous Actions

Ren,

Rathinam,

Choset

2021

Preprint

View full text Add to dashboard Cite

Multi-agent path finding (MAPF) determines an ensemble of collision-free paths for multiple agents between their respective start and goal locations. Among the available MAPF planners for workspaces modeled as a graph, A*based approaches have been widely investigated and have demonstrated their efficiency in numerous scenarios. However, almost all of these A*-based approaches assume that each agent executes an action concurrently in that all agents start and stop together. This article presents a natural generalization of MAPF with asynchronous actions (MAPF-AA) where agents do not necessarily start and stop concurrently. The main contribution of the work is a proposed approach called Loosely Synchronized Search (LSS) that extends A*-based MAPF planners to handle asynchronous actions. We show LSS is complete and finds an optimal solution if one exists. We also combine LSS with other existing MAPF methods that aims to tradeoff optimality for computational efficiency. Extensive numerical results are presented to corroborate the performance of the proposed approaches. Finally, we also verify the applicability of our method in the Robotarium, a remotely accessible swarm robotics research platform.

show abstract

Section: Numerical Resultsmentioning

confidence: 99%

Section: B Meta-agent Conflict-based Searchmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Loosely Synchronized Search for Multi-agent Path Finding with Asynchronous Actions

Ren,

Rathinam,

Choset

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Multi-Agent Path-Finding (MAPF) [30]- [32] methods are methods tailored for multi-agent control problems. These methods tend to focus on collision avoidance rather than optimality.…”

Section: Multi-agent Path-findingmentioning

confidence: 99%

A Neural Network Approach for High-Dimensional Optimal Control Applied to Multi-Agent Path Finding

Onken,

Nurbekyan,

et al. 2021

Preprint

View full text Add to dashboard Cite

We propose a neural network approach for solving high-dimensional optimal control problems arising in real-time applications. Our approach yields controls in a feedback form and can therefore handle uncertainties such as perturbations to the system's state. We accomplish this by fusing the Pontryagin Maximum Principle (PMP) and Hamilton-Jacobi-Bellman (HJB) approaches and parameterizing the value function with a neural network. We train our neural network model using the objective function of the control problem and penalty terms that enforce the HJB equations. Therefore, our training algorithm does not involve data generated by another algorithm. By training on a distribution of initial states, we ensure the controls' optimality on a large portion of the state-space. Our grid-free approach scales efficiently to dimensions where grids become impractical or infeasible. We demonstrate the effectiveness of our approach on several multi-agent collisionavoidance problems in up to 150 dimensions. Furthermore, we empirically observe that the number of parameters in our approach scales linearly with the dimension of the control problem, thereby mitigating the curse of dimensionality.

show abstract

High-efficiency inspecting method for mobile robots based on task planning for heat transfer tubes in a steam generator

Zhang

et al. 2023

Front. Mech. Eng.

View full text Add to dashboard Cite

Many heat transfer tubes are distributed on the tube plates of a steam generator that requires periodic inspection by robots. Existing inspection robots are usually involved in issues: Robots with manipulators need complicated installation due to their fixed base; tube mobile robots suffer from low running efficiency because of their structural restricts. Since there are thousands of tubes to be checked, task planning is essential to guarantee the precise, orderly, and efficient inspection process. Most in-service robots check the task tubes using row-by-row and column-by-column planning. This leads to unnecessary inspections, resulting in a long shutdown and affecting the regular operation of a nuclear power plant. Therefore, this paper introduces the structure and control system of a dexterous robot and proposes a task planning method. This method proceeds into three steps: task allocation, base position search, and sequence planning. To allocate the task regions, this method calculates the tool work matrix and proposes a criterion to evaluate a sub-region. And then all tasks contained in the sub-region are considered globally to search the base positions. Lastly, we apply an improved ant colony algorithm for base sequence planning and determine the inspection orders according to the planned path. We validated the optimized algorithm by conducting task planning experiments using our robot on a tube sheet. The results show that the proposed method can accomplish full task coverage with few repetitive or redundant inspections and it increases the efficiency by 33.31% compared to the traditional planning algorithms.

show abstract

Multi-Agent Pathfinding: Definitions, Variants, and Benchmarks

Cited by 15 publications

References 0 publications

Loosely Synchronized Search for Multi-agent Path Finding with Asynchronous Actions

Loosely Synchronized Search for Multi-agent Path Finding with Asynchronous Actions

A Neural Network Approach for High-Dimensional Optimal Control Applied to Multi-Agent Path Finding

High-efficiency inspecting method for mobile robots based on task planning for heat transfer tubes in a steam generator

Contact Info

Product

Resources

About