State and Localization Estimation for Self-Driving Cars with Unscented Kalman Filter
Sophia de Souza Nobre Benevides, Nícolas dos Santos Rosa, Valdir Grassi, Marco Henrique Terra · 2025
This paper introduces the formulation of an accurate localization method for positioning and navigation in autonomous vehicles. Positioning estimation is achieved through the development of an Unscented Kalman Filter (UKF), which integrates data from the following sensors in a ROS environment: an RTK (Real-Time Kinematic) Global Navigation Satellite System (GNSS) receiver and an Inertial Measurement Unit (IMU). The proposed method was implemented and tested using real-world data collected from an autonomous vehicle, ensuring the validation of the system under realistic driving conditions.