Radar High-Speed Target Range-Doppler-Azimuth Coherent Extension Detection for Autonomous Vehicles

Robust motion target detection is a critical component in the pursuit of autonomous driving. However, this technology presents significant challenges in accurately detecting weak signals amidst strong background clutter and recovering the integrated energy loss caused by range migration effects of high-speed targets during single or multiple dwells. In this study, we propose a method for detecting high-speed targets under low signal-to-noise ratio (SNR) using time division multiplexing (TDM) multiple-input multiple-output (MIMO) radar sensors. The proposed method consists of three main steps. Firstly, we employ modified Doppler range processing with phase compensation to tackle the concerns regarding range/Doppler migration, velocity ambiguity, and the coupling of velocity and azimuth information for each dwell. Secondly, we introduce a coherent averaging method to generate multiple dwells range-Doppler maps, thereby leveraging the full coherent integration gains. Lastly, we utilize a parallel-processing optimization model to conduct azimuth super resolution estimation. The proposed method enables coherent extension of range, Doppler, and azimuth measurements across multiple dwells, ensuring robust integrated energy accumulation without any prior information. Through a series of simulations and measurements, we demonstrate the effectiveness of the proposed method in detecting the range, Doppler, and azimuth of high-speed targets while maintaining strong noise robustness. © 2024 IEEE.