A new approach for testing autonomous and fully adaptive radars

A fundamental goal of cognitive radar is to sense, learn and adapt (SLA) to a complex environment. A further goal of cognitive fully adaptive radar (CoFAR) is to jointly optimize the adaptive transmit and receive functions. This type of emerging radar technology is anticipated to play a key part in the future of autonomous sensors and systems. CoFAR systems present a number of unique testing challenges relative to traditional systems. In particular the highly adaptive nature of these systems will generally require more rigorous testing to ensure good performance over a wide range of operating environments. For example, a CoFAR system will need to be presented with a wide range of operating scenarios and conditions to ensure a desired level of performance will be met during actual operations of the radar. This will typically preclude traditional approaches for testing using limited numbers of flight tests in test ranges that represent only a small number of scenarios and conditions expected to be encountered in real operations. Testing over a wider range of conditions will typically be cost prohibitive. This paper presents a new method for testing CoFAR systems in a “virtual test range” using high-fidelity simulated data. We provide a hardware in-the-loop demonstration of the approach using an actual maritime radar system.