Radar Sensor-Based Longitudinal Motion Estimation by Using a Generalized High-Gain Observer
H. Bessafa, Zehor Belkhatir, Cédric Delattre, Redouane Khemmar, Ali Zemouche, Rajesh Rajamani · 2024
This study explores vehicle longitudinal dynamic estimation using a noisy radar sensor. By incorporating additional velocity information, we propose an improved generalized high-gain observer that ensures exponential Input to State Stability (ISS) of estimation errors with explicit bound. The observer of this work deals with the extra measurement differently than our recent paper, that does not account for noisy measurement. The observer outperforms standard high gain in convergence speed, accuracy, and noise rejection. The proposed algorithm is tested and validated using a tracking scenario designed using the CARLA simulation environment. It is shown through the results that the proposed observer outperforms the standard high-gain observer in terms of convergence speed, accuracy and noise rejection.