http://wiki.ros.org/robot_localization
Filtered using: 1- EKF 2- UKF
Input: Odometry, IMU, and GPS (.bag file)
Output: 1- Filtered path trajectory 2- Filtered latitude, longitude, and altitude
It runs 3 nodes: 1- An *kf instance that fuses Odometry and IMU, and outputs state estimate approximations 2- A second *kf instance that fuses the same data with GPS 3- An instance navsat_transform_node, it takes GPS data and produces pose data
Also compared memory consumption of EKF and UKF nodes via htop
Please refer docs folder for results