AI-Powered Maritime Anomaly Detection System Development

The contract seeks the development of AI and machine learning models designed to identify anomalous behavior in maritime vessels by analyzing multiple data sources including radar, AIS, and FMV feeds. These models must be capable of detecting patterns indicative of illicit or unusual activity, such as unauthorized vessel movements, spoofed identifiers, or deviations from typical maritime routines. The solution must be engineered for real-time performance and high accuracy, with robust training datasets to account for diverse environmental and operational conditions across global waters. The resulting AI system must integrate seamlessly with existing command and control platforms, particularly TAK, to enable actionable intelligence delivery to operational users. Performance will be evaluated based on detection reliability, False positive rates, latency, and interoperability with federal systems. The work is scoped as a subcontract under the C5I Division 3 Portsmouth, with performance centered in San Diego and a response deadline of August 14, 2026. The NAICS code 511210 indicates this effort falls under software publishing, emphasizing the delivery of proprietary algorithmic capabilities rather than hardware or general consulting services.