Localization of autonomous vehicle with low cost sensors
Résumé
This paper presents the design and real-time validation of an Inertial Measurement Unit (IMU) and Global Positioning System (GPS) data fusion algorithm for real-time localization in an autonomous vehicle system. The data fusion method is based on a low-pass filter and the Error State Extended Kalman Filter (ES-EKF). In this paper, the system's hardware and software design are detailed. Real-time validation of the proposed method is presented using low-cost sensors. The algorithm is deployed and tested employing ground truth data on an embedded microcontroller, the STM32 Nucleo, and achieved a 92% accuracy level on the road and proved reliable in actual industrial applications.