IMPACT (Intelligent Multi-Agent System for Planning Cooperative Tasks)

To tackle these challenges, the project leverages Industry 4.0 principles through the design of Cyber Physical Systems (CPS) and Multi-Agent Systems (MAS). CPS networks consist of intelligent entities that communicate and interact using AI and ICT. MAS, in particular, offers a scalable and fault-tolerant solution for efficient task scheduling, involving autonomous agents that cooperate to achieve common goals.

The project proposes splitting the functionality of agents into two parts: a local component with basic connectivity and computational capabilities, and a remote component with high-performance computing resources, such as cloud servers. This approach enhances performance, especially in scenarios requiring extensive message exchange, like rescheduling or optimization in environments with shared production capabilities.

IMPACT aims to implement MAS-based, distributed, cooperative production systems using cloud technology, enhancing performance in demanding reconfiguration situations and easing interconnection across geographically different sites. The project will demonstrate this approach with physical demonstrators at Fraunhofer IPK and Fraunhofer Italia, addressing performance issues in distributed MAS while protecting the private networks of different production sites.