Abstract | ||
---|---|---|
In a multi-agent pathfinding (MAPF) problem, agents need to navigate from their start to their goal locations without colliding into each other. There are various MAPF algorithms, including Windowed Hierarchical Cooperative A*, Flow Annotated Replanning, and Bounded Multi-Agent A*. It is often the case that there is no a single algorithm that dominates all MAPF instances. Therefore, in this paper, we investigate the use of deep learning to automatically select the best MAPF algorithm from a portfolio of algorithms for a given MAPF problem instance. Empirical results show that our automatic algorithm selection approach, which uses an off-the-shelf convolutional neural network, is able to outperform any individual MAPF algorithm in our portfolio. |
Year | Venue | DocType |
---|---|---|
2019 | CoRR | Journal |
Volume | Citations | PageRank |
abs/1906.03992 | 0 | 0.34 |
References | Authors | |
0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Devon Sigurdson | 1 | 1 | 2.39 |
Vadim Bulitko | 2 | 670 | 67.16 |
Sven Koenig | 3 | 3125 | 361.22 |
carlos hernandez | 4 | 56 | 3.85 |
William Yeoh | 5 | 66 | 8.46 |