Extended Kalman Filter for estimating 15-States (Pose, Twist & Acceleration) using Omni-Directional model for prediction and measurements from IMU and Wheel Odometry.
- ROS KInetic >=
- Eigen
- The position covariances keeps increasing in left column result as there is no correction data for position and it is being estimated by integrating velocities and acceleration.
- In the Second Column the Absolute Position is Fused, Hence position covariances converge as the position data from the odometry is being used for correction.
- Fix segmentation Fault when data selection vector has all 0 elements.
- Add way to set parameters through launch files.
This project is still in progress, and not recommended to be used apart from research purposes.