SOVA is a SLAM on Voxel Alignment and an open-source Python library, designed for fast and adaptive comparison of different approaches to solving the voxel-based planar SLAM problem.

Our main goal is to provide extendable, simple and efficient interfaces for testing various voxel SLAM hypotheses, which include different subdivision/segmenter/backend criteria.

---

Installation

To use this library you need to:

1. Download and install Python 3.10 from the official website.
2. Install pip package:

   pip install sova
   

   If you want to use MROBBackend robust optimisations, you have to install mrob library manually:
   1. Download wheel from source
   2. Install mrob from wheels

   python -m pip install mrob --no-index --find-links wheel/ --force-reinstall
   

Now you have everything you need to run your voxel-based pipeline.

Examples

Examples of using the voxel-based pipeline are presented in the examples directory with the all necessary instructions of how to run them.

Contributing

To contribute to the project you must:

1. Get to know the project structure:

   sova
   ├── backend
   ├── filter
   ├── pipeline
   ├── segmenter
   ├── subdivider
   ├── typing
   └── utils
       ├── dataset_reader
   

2. Implement new subdivision/segmenter/backend approach which satisfy the relevant interface.
3. Create PullRequest to the repository.
4. Go through the review and wait for your code to appear in the main branch.