Web-based multi-agent system architecture in a dynamic environment
International Journal of Knowledge-based and Intelligent Engineering Systems
Hi-index | 0.00 |
Research sought to develop agents that can stand in for human air traffic controllers in large-scale simulation studies of advanced air traffic management concepts. The agents use a high-level activity model to structure the air traffic control task, and apply heuristics to plan and formulate clearances. Results indicate the agents handle traffic flow spacing problems well, but experience some difficulties managing complex 'merge' problems. Follow-on research on how such agents can make plausible errors to support safety analyses identified interesting aspects of agent performance in the face of errors. The research was supported by the NASA Aviation System Capacity Program and the FAA/NASA Aviation Safety Program.