Multi-Agent Health Management System
Agency / Branch:
DOD / ARMY
We propose to develop an intelligent multi-agent approach for distributed, dynamic, and active health management of fleet-wide ground vehicle subsystems/systems by incorporating advanced technologies in multi-disciplinary areas, such as software agent, signal processing, and data mining. In this framework, server-side agents include a light-weight library of diagnostics/prognostics algorithms/models developed based on the fleet-wide ground vehicle information, maintenance log, and test data, etc. When a vehicle connects to the server, a diagnostic/prognostics model will be generated automatically by the server and downloaded to the vehicle. The diagnostic and prognostic decisions made by each vehicle will be sent back to the server to confirm and update the model library if needed. If necessary, an "updated" diagnostics/prognostics model may be re-generated and the vehicle-side diagnostics/prognostics agents will be updated. In this way, the vehicle-side diagnostics/prognostics agents have the capabilities of self-adaptation and self-learning utilizing the vehicle and fleet level information. The whole framework will be developed under a unified software agent framework.
Small Business Information at Submission:
Director, Signal Processing
Director, Contracts and Proposals
INTELLIGENT AUTOMATION, INC.
15400 Calhoun Drive Suite 400 Rockville, MD 20855
Number of Employees: