Intelligent Distributed Multi-Agent Hybrid Control Systems

Autonomous intelligent multi-agent systems are systems that sense and manipulate their environment by gathering multi-model sensor data, compressing and representing it in symbolic form at various levels of granularity, and using the representations to reason and learn about how to optimally interact with the environment. This project proposes an innovative multi-agent distributed intelligent control methodology by integrating planning and decision making, soft computation, autonomous control and discrete event control techniques. The proposed intelligent system is organized by the distributed control functions in a hierarchical architecture that makes those functions relatively autonomous by introducing enough coordination and supervision to ensure the harmony of the distributed controllers necessary for high performance. An important new feature of our approach is that we work on both fully autonomous intelligent systems and intelligence augmentation of human centered systems. In Phase I, the detailed methodology, computation approaches and architectural concepts to support design and implementation of generic distributed multi-agent control systems are developed, and the applicability to the intelligent digital battlefield is illustrated. In Phase II, a fully integrated design and prototyping environment to support multi-agent control applications will be developed. BENEFITS: The research results of this SBIR program have the potential to lead to the following commercialized product: Intelligent Control for Multi-Agent Hybrid Dynamic Systems. The applications of this technology arise in all areas of the intelligent digital battlefield, manufacturing systems, intelligent vehicle highway systems, industrial processes, and command, control, communications and intelligence (C(3)I) systems.