Adaptive Tracking Algorithm for Manoeuvring Targets Optimized Based on the CS-Jerk Model

To address the issue of limited tracking accuracy in the current statistical-Jerk model (CS-Jerk) during manoeuvring target tracking, which arises from the fixed values of two parameters (i.e., the jerk limit and manoeuvre frequency), an optimized adaptive CS-Jerk model is proposed. This proposal is based on an analysis of the applicable scenarios and inherent defects of the standard CS-Jerk model. Combined with the Kalman filtering algorithm, the optimized model achieves higher-precision tracking of manoeuvring g targets. Specifically, the optimized model describes the jerk limit through the relationship between the current filtered posterior estimate of the target's jerk and the jerk increment. Additionally, a weight function with a smooth variation trend is constructed to allocate weights between the prior predicted value and posterior estimated value of the jerk at the current moment, thereby characterizing the mean acceleration and manoeuvre frequency at the current moment. Finally, simulation results demonstrate that the optimized model can accurately track jerk manoeuvres of varying intensities, with relatively stable error variations and superior tracking accuracy, thus verifying the feasibility and effectiveness of the proposed algorithm.