In this paper, distributed leader–follower control algorithms are presented for linear multi-agent systems based on output regulation theory and internal model principle. By treating a leader to be followed as an exosystem, the proposed framework can be used to generalize existing multi-agent coordination solutions to allow the identical agents to track an active leader with different dynamics ...

Hardware agents as a part of cooperative multi-agent systems act in dynamically changing environments and accomplish tasks jointly. Since the pure hybrid plan representation provides no explicit means to represent particularly communication, task coordination and distribution in the multi-agent system, the proposed plan representation was adjusted to describe these aspects and combines the hybr...

Complex tasks are often solved by teams because no one individual has the collective expertise, information, or resources required for the e ective completion or performance of a task. This paper describes a prototype, implemented in the RETSINA multi-agent infrastructure, in which agents interact with each other via capability-based and team-oriented coordination. We propose a model of team-or...

This paper presents a conceptual model of an agent (called Collaborator Agent) intended to design collaborative software architectures based on multi-agent systems. The authors’ model combines astutely two research areas: Multi-Agent Systems (MAS) and Computer Supported Cooperative Work (CSCW). The particularity of their approach is the division of the collaborative process into three spaces ac...

Exploring agent conversation in the context of fine-grained agent coordination research has raised several intellectual questions. The major issues pertain to interactions between different agent conversations, the representations chosen for different classes of conversations, the explicit modeling of interactions between the conversations, and how to address these interactions. This paper is n...

In this paper, we propose to guide reinforcement learning (RL) with expert coordination knowledge for multi-agent problems managed by a central controller. The aim is to learn to use expert coordination knowledge to restrict the joint action space and to direct exploration towards more promising states, thereby improving the overall learning rate. We model such coordination knowledge as constra...

We explore deep reinforcement learning methods for multi-agent domains. We begin by analyzing the difficulty of traditional algorithms in the multi-agent case: Q-learning is challenged by an inherent non-stationarity of the environment, while policy gradient suffers from a variance that increases as the number of agents grows. We then present an adaptation of actor-critic methods that considers...

This paper provides a novel approach to multi-agent coordination in general sum Markov games. Contrary to what is common in multi-agent learning, our approach does not focus on reaching a particular equilibrium between agent policies. Instead, it learns a basis set of special joint agent policies, over which it can randomize to build different solutions. The main idea is to tackle a Markov game...

Auctions are proposed as a distributed negotiation mean, particularly useful in multiagent systems where both cooperative and self-interested agents compete for resources and services. The aim of this paper is to show how auction mechanisms on the Internet can be easily implemented by using programmable tuple spaces. Tuple spaces are shared repositories of information that follow the Linda mode...

Achieving joint objectives by teams of cooperative planning agents requires significant coordination and communication efforts. For a single-agent system facing a plan failure in a dynamic environment, arguably, attempts to repair the failed plan in general do not straightforwardly bring any benefit in terms of time complexity. However, in multi-agent settings the communication complexity might...