visual_inertial_SLAM
The goal of this project is to implement visual-inertial simultaneous localization and mapping (SLAM) using an extended Kalman filter (EKF).
Installation
conda create -n vslam python=3.8
conda activate vslam
pip install -r requirements.txt
Usage
- Use 'python main.py' to run the code with the default config.
- You can adjust the params in "./config/config.yaml" to get different mapping results. The params are clearly defined with annotation.
Results
- Full SLAM(measurement noise assumption=2.0)
- Full SLAM(measurement noise assumption=0.1)