'Multi-agent systems are widely applicable to real-world problems and applications ranging from warehouse automation to environmental monitoring, autonomous driving, and even game playing. While multi-agent systems can execute time-sensitive, complex, and large-scale problems that are intractable for single agents, it is challenging to efficiently coordinate such systems for cooperative or non-cooperative tasks. On the other hand, decision making and reinforcement learning in single-agent scenarios have seen tremendous achievements in recent years; yet, translation of many single-agent techniques to the multi-agent domain may not be straightforward. The main challenges lie intrinsically in the nature of the multi-agent systems, including: complex interaction dynamics, constrained inter-agent communication, complex optimality, heterogeneity in the system and the potential presence of adversaries.
The objective of this Research Topic is to report on the recent advances in multi-agent planning and decision-making. While decision-making and planning for a single agent has been extensively studied, the multi-agent version of this problem has not received the same attention and is less understood. In addition to sensing and planning, multi-agent systems must also deal with the need to communicate and coordinate. Also, several multi-agent architectures assume very large numbers of agents each having limited computational resources. Coordination and cooperation can be either altruistic or individualistic and deciding between the two (and when) is not straightforward. By bringing together in this Research Topic experts from the field, it is hoped to make the community more aware of the existing challenges of the multi-agent decision-making problem, and at the same time also disseminate to the community promising, recent, and novel research trends in this area.
This Research Topic is linked to the IROS 2022 workshop titled ‘Decision Making in Multi-Agent Systems’, but it also welcomes relevant submissions from authors who did not attend. Contributions that were initially presented at the conference should be extended to include at least 30% original content. Papers reporting both on theoretical developments and non-trivial experimental demonstrations would be welcome. Of particular interest are articles that address topics related (but not limited) to:
• Multi-agent communication architectures
• Efficient solver for multi-agent path finding
• Risk-aware planning
• Integrated perception and planning for robot teams and/or swarms
• Application of reinforcement learning and deep learning techniques to multi-agent problems
• Multi-agent reinforcement learning
• Adversarial behaviors and team strategies
• Computational efficiency for large robot teams
• Mixed human-robot teams
'Multi-agent systems are widely applicable to real-world problems and applications ranging from warehouse automation to environmental monitoring, autonomous driving, and even game playing. While multi-agent systems can execute time-sensitive, complex, and large-scale problems that are intractable for single agents, it is challenging to efficiently coordinate such systems for cooperative or non-cooperative tasks. On the other hand, decision making and reinforcement learning in single-agent scenarios have seen tremendous achievements in recent years; yet, translation of many single-agent techniques to the multi-agent domain may not be straightforward. The main challenges lie intrinsically in the nature of the multi-agent systems, including: complex interaction dynamics, constrained inter-agent communication, complex optimality, heterogeneity in the system and the potential presence of adversaries.
The objective of this Research Topic is to report on the recent advances in multi-agent planning and decision-making. While decision-making and planning for a single agent has been extensively studied, the multi-agent version of this problem has not received the same attention and is less understood. In addition to sensing and planning, multi-agent systems must also deal with the need to communicate and coordinate. Also, several multi-agent architectures assume very large numbers of agents each having limited computational resources. Coordination and cooperation can be either altruistic or individualistic and deciding between the two (and when) is not straightforward. By bringing together in this Research Topic experts from the field, it is hoped to make the community more aware of the existing challenges of the multi-agent decision-making problem, and at the same time also disseminate to the community promising, recent, and novel research trends in this area.
This Research Topic is linked to the IROS 2022 workshop titled ‘Decision Making in Multi-Agent Systems’, but it also welcomes relevant submissions from authors who did not attend. Contributions that were initially presented at the conference should be extended to include at least 30% original content. Papers reporting both on theoretical developments and non-trivial experimental demonstrations would be welcome. Of particular interest are articles that address topics related (but not limited) to:
• Multi-agent communication architectures
• Efficient solver for multi-agent path finding
• Risk-aware planning
• Integrated perception and planning for robot teams and/or swarms
• Application of reinforcement learning and deep learning techniques to multi-agent problems
• Multi-agent reinforcement learning
• Adversarial behaviors and team strategies
• Computational efficiency for large robot teams
• Mixed human-robot teams
