Sensor fusion for vehicle odometry and path tracking: a ROS simulation

Autonomous vehicles are an emerging area with many problems without a definitive solution. One of these problems is the vehicle’s pose estimation which until these days is still an open problem. On this research we propose a sensor fusion strategy that estimates a vehicle’s position and orientation, this algorithm is used for the higher level tasks of trajectory generation and path following. The Visual Odometry (VO) algorithms that were tested during the research are ORB SLAM2. We have used Kalman filters as sensor fusion algorithm. The main odometry algorithm was designed in order to function with a GPS, VO, IMU, steering sensor and a speed sensor. The proposed strategy was tested in simulation using ROS and Gazebo as a prior stage of an experimental setup. This proposal impacts positively in economic and computational costs reduction with respect to traditional Visual-lidar approaches.