3D target tracking by UAVs subject to measurement uncertainties

Abstract

This paper proposes an Extended Kalman Filter (EKF) based approach to track a moving target UAV in 3D. The state of the target UAV is estimated from the range, azimuth, and elevation angle measurements that are assumed to be available from a ground based sensor or from on-board seeker antenna. The chaser vehicle states are also estimated using onboard sensors, namely GPS + IMU. These estimated states (chaser and the target) are utilized in a nonlinear guidance law which provides signals for the velocity, flight path, and course/heading angle to the chaser so as to track the target vehicle. It is to be mentioned that the guidance controller onboard the chaser was designed based on a backstepping like technique. All process and measurement noise are modelled as zero-mean Gaussian processes with known covariances. The performance of the combined estimation and guidance algorithm is shown via extensive simulation.