Quadtree mapping is a library that can be used in the context of mobile robotics to obtain a 2D map of the environment the robot is in. It is not a SLAM (Simultaneour Localization and Mapping) approach. As input uses:
- the 2D pose of the robot (x, y, orientation)
- A 2D lidar scan (for example an Hokuyo laser was used) Internally it stores the data in a quadtree data structure. In the initial commit of main.cpp, you can see out to create as output a cpp file that compiled could create a 2D image of the map.
The original code is pretty primitive and the comments are in Italian.
My idea is to upload the project in gitlab, refactor the code, add meaningful comments in English, add unit tests, split it in a library and some viewer. As viewer it would be simple for me to create something in QT using QCustomPlot. To test input data, the data from RawSeeds project might be used.
Original authors of the projects are:
This project was an AIRLab project.