This is a demo fusing IMU data and Odometry data (wheel odom or Lidar odom) or GPS data to obtain better odometry.
Currently, I implement Extended Kalman Filter (EKF), batch optimization and isam2 to fuse IMU and Odometry data.
This demo has the following dependencies:
-
- I test it on ROS melodic and kinetic version
-
- use tag 4.28a now
Clone the repository and catkin_make:
cd ~/catkin_ws/src
git clone https://github.com/Guo-ziwei/fusion
cd ../
catkin_make
source ~/catkin_ws/devel/setup.bash
run EKF:
rosrun ekf_fusion ekf_fusion_node bag_filename.bag
run batch:
rosrun gtsam_imu imu_preintegrated bag_filename.bag
it will take a long time to get the result and I will fix it in the future
run isam2:
rosrun gtsam_imu imu_preintegrated_sm bag_filename.bag
Test on MIt dataset
Now the demo is running off-line (reading data from bag) and I will turn it to online.