2021
DOI: 10.3390/app12010272
|View full text |Cite
|
Sign up to set email alerts
|

Cooperative Multi-Robot Task Allocation with Reinforcement Learning

Abstract: This paper deals with the concept of multi-robot task allocation, referring to the assignment of multiple robots to tasks such that an objective function is maximized. The performance of existing meta-heuristic methods worsens as the number of robots or tasks increases. To tackle this problem, a novel Markov decision process formulation for multi-robot task allocation is presented for reinforcement learning. The proposed formulation sequentially allocates robots to tasks to minimize the total time taken to com… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
9
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
7

Relationship

0
7

Authors

Journals

citations
Cited by 18 publications
(9 citation statements)
references
References 21 publications
0
9
0
Order By: Relevance
“…Our approach relies on the continuous training of a smart auctioneer which is not attached to a particular drone and can be simply secured with backups on one or more other drones. Compared to our work, [22], [23] address different problems than MR(p)TA and [24] assumes single-task robots with delayed allocation and does not consider robotic clusters or task parallelization. Finally our solution is meant to support high dimensional action spaces.…”
Section: Related Workmentioning
confidence: 97%
See 1 more Smart Citation
“…Our approach relies on the continuous training of a smart auctioneer which is not attached to a particular drone and can be simply secured with backups on one or more other drones. Compared to our work, [22], [23] address different problems than MR(p)TA and [24] assumes single-task robots with delayed allocation and does not consider robotic clusters or task parallelization. Finally our solution is meant to support high dimensional action spaces.…”
Section: Related Workmentioning
confidence: 97%
“…However, its application to multi-agent systems remains limited and complex due to the non-stationary environment introduced by simultaneous learning agents [21]. Recently, several RL-based approaches have been proposed to solve various decentralized MRS problems such as exploration [22], collaborative decision-making [23] and MRTA [24]. These solutions rely on a partial or total centralization of learning.…”
Section: Related Workmentioning
confidence: 99%
“…Park et al [102] propose a novel MDP formulation for multi-robot task allocation using reinforcement learning in problems involving multi-robot tasks. The problem considered involves sequentially allocating robots to spatial tasks with particular workload requirements and their deep reinforcement learning method makes use of a cross-attention mechanism to compute the importance of tasks for robots.…”
Section: Other Approachesmentioning
confidence: 99%
“…Authors in [20] developed a new allocation method based on a reinforcement learning algorithm for allocating some tasks into the free robots. The paper showed that the new method found results better than the heuristic methods.…”
Section: Previous Workmentioning
confidence: 99%