Ein Multi-Agentensystem für die modellbasierte Echtzeit-Fehlerdiagnose

Real-time AI systems, among them multiagent systems, are gaining importance in complex technical applications. This paper presents VEX, an ambitious multiagent system for model-based real-time fault diagnosis in modular production systems. The architecture comprises several process agents, a simulation agent and a user agent. We discuss the most interesting characteristics of this real-time AI architecture. As implementation examples we present the multi-DSP core of the simulation agent, the client-server communication protocol for the process agents and the user agent’s graphical interface.