Correct-by-Design Teamwork Plans for Multi-Agent Systems

TL;DR

This paper introduces Teamwork Synthesis, a novel approach to automatically generate multi-agent teamwork plans by dynamically establishing minimal interactions, simplifying the design of distributed systems.

Contribution

It reformulates the distributed synthesis problem by removing fixed interaction architectures and automatically determining minimal interactions to achieve goals.

Findings

Successfully synthesizes teamwork plans with minimal interactions.

Automatically adapts interaction structures to meet specifications.

Enhances the feasibility of designing multi-agent systems.

Abstract

We propose Teamwork Synthesis, a version of the distributed synthesis problem with application to teamwork multi-agent systems. We reformulate the distributed synthesis question by dropping the fixed interaction architecture among agents as input to the problem. Instead, our synthesis engine tries to realise the goal given the initial specifications; otherwise it automatically introduces minimal interactions among agents to ensure distribution. Thus, teamwork synthesis mitigates a key difficulty in deciding algorithmically how agents should interact so that each obtains the required information to fulfil its goal. We show how to apply teamwork synthesis to provide a distributed solution.