Riassunto analitico
This thesis outlines the integration of two approaches in simultaneous localization and mapping (SLAM), namely Lidar-Inertial Odometry SLAM (LIO SLAM) and Extended Kalman Filter SLAM (EKF SLAM), to enhance the accuracy and reliability of autonomous driving, focusing Formula Student Driverless competitions. LIO SLAM, by fusing data from LIDAR and IMU(s), provides a precise estimation of the vehicle's pose, to be further corrected through an EKF SLAM algorithm while simultaneously creating a map of the environment. Integrating LIO SLAM into EKF SLAM enables exploiting the accurate motion estimates provided by the former, to improve the accuracy of autonomous navigation specifically for Formula Student vehicles. The integrated approach seeks to enhance localization and mapping results beyond what can be achieved by using either method independently, providing an accurate solution to the SLAM problem in the Formula Student domain.
|
