To convert rplidar raw data (offline) to BHM compatible csv, run rplidar_to_bhm_convert_offline.py
. Data will be saved in datasets (and datasets/figs/).
To run BHM, run main_bhm_pytorch.py
. Parameters of BHM can be set in the yaml
files in the config folder.
To convert BHM compatible csv to an updatable Bayesian Hilbert Map, run
rplidar_to_bhm_convert_online.py
To create Bayesian Hilbert Maps from the Lidar directly, run rplidar_to_bhm_live_fromlidar2.py
- TensorFlow Lite
- PyTorch
- Matplotlib
- Pandas
- Numpy
- link websites for installation
- TensorFlow:
https://qengineering.eu/install-tensorflow-1.15.2-on-raspberry-pi-4.html
- PyTorch:
https://gist.github.com/akaanirban/621e63237e63bb169126b537d7a1d979
In the project directory, you can run:
To run the Edge TPU computer vision code go to the edge-tpu-tiny-yolo_ directory and run: