To run the program:
- Clone project and enter its main directory
- Either
- have your environment set to support 'ipython' and 'matplotlib'
- setup a new venv
- run
./setup
to deploy a venv - each time you want to run the app, first launch
. ./activate
(notice additional spaced out leading dot)
- run
- Run
./run
, wait a while and enjoy!
Data relevant to the C++ task is printed in console (alongside some additional logging info) Everything else should show up on a pyplot visualization
- To compute a location of the point we:
- imagine 'range circles' for each anchor reading
- take all intersections of each pair of readings relevant to the point
- plus, when such circles don't intersect, we take a point that is closest to both circles at once)
- filter out those intersection points that are too far away from a median of these points
- compute a mean of remaining intersection points