An investigation of an alternative coordinate space parameterisation for radar target tracking

Target tracking in radar uses mature methods that have been around for many years. The Extended Kalman Filter has long been the typical choice of tracking filter in radar systems due to its low computational effort and its simple implementation. With many recent advances in hardware and processing capability on airborne platforms, it is now becoming possible to integrate higher-complexity algorithms into real-time systems. This work presents an alternative method of tracking a single target using a hybrid filtering approach with an alternative state space parametrization. This approach will be compared to an EKF, UKF and a particle filter using simulated data and root mean square error.