Name
Title | ||
|---|---|---|
A hierarchical training method of generating collective foraging behavior for a robotic swarm |
Abstract | ||
|---|---|---|
Swarm robotics is a field in which multiple robots coordinate their collective behavior autonomously to accomplish a given task without any form of centralized control. Training a robotic swarm to complete a multi-objective task under sparse rewards is a challenging task in reinforcement learning (RL). This research has applied a hierarchical training method for the RL training process to address the multi-objective task with sparse rewards. We conduct experiments where a robotic swarm has to accomplish a complex collective foraging problem using computer simulations. The results show that the proposed approach leads to perform more effectively than a conventional RL approach. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1007/s10015-021-00714-x | ARTIFICIAL LIFE AND ROBOTICS |
Keywords | DocType | Volume |
Robotic swarm, Reinforcement learning, Task partitioning | Journal | 27 |
Issue | ISSN | Citations |
1 | 1433-5298 | 0 |
PageRank | References | Authors |
0.34 | 0 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Boyin Jin | 1 | 0 | 0.34 |
| Yupeng Liang | 2 | 0 | 0.34 |
| Ziyao Han | 3 | 0 | 0.34 |
| Motoaki Hiraga | 4 | 2 | 2.41 |
| Kazuhiro Ohkura | 5 | 0 | 0.34 |