An Error-State Extended Kalman Filter Based State Estimation and Localization Algorithm for Autonomous Systems

This research paper presents an in-depth analysis of how Error-State Extended Kalman (ES-EKF) Filter can be utilized to optimize the state estimation and localization of self-driving cars. The goal is to improve the accuracy and stability of state estimation in autonomous vehicle. The ES-EKF is a variant of the Kalman Filter that takes into account the modelling errors of the systems. The self-driving car systems is modelled as state-space system, where state is estimated using sensor measurements. By comparing the performance of the ES-EKF with different types of sensors,such as GPS, IMU, and LIDAR it is concluded that ES-EKF provides more accurate and stable state estimates with a combination of multiple sensors. The performance of the ES-EKF is evaluated by using CARLA simulator.