Estimation and Prediction of Unmanned Aerial Vehicle Trajectories

There is serious concern about the introduction of Unmanned Aerial Vehicles (UAV) in the National Air Space (NAS) because of their potential to increase the risk of collision between aircraft. At present, many UAV platforms lack a Sense and Avoid (SAA) capability to mitigate collision risk, and this has prevented both the government and private contractors from using these platforms in critically needed reconnaissance, surveillance, and security enforcement missions. To demonstrate a SAA capability that is applicable to a wide range of UAV platforms, advanced trajectory estimation and prediction algorithms are developed and used to exploit a small collision avoidance radar currently under development for UAV operation. Collision prediction algorithms will assess potential risk in probabilistic terms using adaptive techniques that permit accurate predictions across long time horizons. Techniques to ensure these predictions are robust to modeling uncertainty increase the utility the developed SAA capability for realistic scenarios.