Effect of EKF Localization on Path Tracking in Autonomous Vehicle Navigation Systems

This paper presents an integrated approach to evalu-ating the interplay between localization and path-tracking performances in autonomous vehicle navigation systems. Considering stochastic uncertainties in motion and measurement models, we design an extended Kalman filter (EKF) for localization and employ the pure pursuit method for steering control. We analyze how localization accuracy impacts path-tracking performance through extensive simulations, varying the number of landmarks and noise covariance settings. Our results highlight the critical role of accurate localization in improving path tracking, offering valuable insights for designing robust autonomous navigation systems.