Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. SPONSOR/MONITOR'S ACRONYM(S) 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)Defense AFRL-IF-RS-TR-2006-220 DISTRIBUTION AVAILABILITY STATEMENT APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED. PA #06-46313. SUPPLEMENTARY NOTES ABSTRACTAs systems become complex with many interconnected subsystems, decentralized control becomes essential. When certain parameters of the system are unknown, and/or when subsystems are not aware of the signals from other subsystems that affect their behavior, we need decentralized adaptive control. The report deals with questions that arise while analyzing the stability and performance of decentralized adaptive control systems. The project produced three specific results: 1. Interconnected dynamical systems can be stable even when there is no communication between subsystems, provided all subsystems have common knowledge of the goals of the other subsystems. 2. Even though stability can be achieved without communication, the latter is necessary to satisfy performance requirements. To keep communication costs to a minimum, partial communication has to be used. This gives rise to stability problems which were resolved. 3. The problem as to when a subsystem in an interconnected-system communicates with another is an important one and needs to be investigated further. Simulation results have clearly shown that significant improvement in the performance of the overall system can be achieved by subsystems communicating only over critical intervals of time. SUBJECT TERMSDecentralized decision making, control theory, adaptive control, biological comparison, communication networks SECURITY

No abstract

No abstract