Adaptive Turn Rate Estimation for Radar Maneuvering Target Tracking Based on Doppler Measurements

Abstract

It is generally accepted that the constant turn (CT) model is a standard model option for maneuvering target tracking under curvilinear trajectories, where accurate, real-time determination of the turn rate parameter is essential. However, in practical scenarios, estimation of the turn rate is still a considerable challenge. In this article, we first propose an adaptive turn rate estimation strategy using Doppler measurements and then incorporate it into an interaction multiple model (IMM) structure to obtain a complete tracking algorithm, namely, the IMM-PATR algorithm. Specifically, the proposed adaptive turn rate estimation strategy comprises three key modules. First, to solve the multivalued problem, a prediction approach including turn rate calculation and selection is proposed by using Doppler measurements. Second, with the known turn rate, a modified CT motion model is derived to provide a more precise representation of the target's motion characteristics. Third, a secondary update using a kinematic constraint of a constant speed is designed to improve the accuracy of the velocity estimation, which is subsequently incorporated into the update process of the turn rate. In fact, these three modules function as an integrated whole, interconnected and mutually influential. Comparative simulation analyses with existing CT models, adaptive turn rate estimation methods, and different measurement accuracies are implemented to demonstrate the efficacy of the IMM-PATR algorithm.