- these scripts assist you in processing the Interaction Dataset
- for details about and access to the dataset, visit https://interaction-dataset.com/
- both
python2andpython3are supported
csv: for reading the csv track files- to work with the map:
- either
lanelet2for most convenient map usage- see https://github.com/fzi-forschungszentrum-informatik/Lanelet2 for details
- or
pyprojxml
- either
math,numpy,os,argparse,timeandmatplotlibfor visualizing the scenarios- use
test_imports.pyto test whether you have all necessary packages installed
- copy/download the data into the right place
- copy/download the track files into the folder
recorded_tracks, keep one folder per scenario, as in your download - copy/download the maps into the folder
maps - your folder structure should look like in folder-structure.md
- copy/download the track files into the folder
- to visualize the data
- run
./main_visualize_data.py <scenario_name> <trackfile_number (default=0)>from folderpythonto visualize scenarios
- run
- if you only want to load and work with the track files
- run
./main_load_track_file.py <tracks_filename>from folderpythonto load tracks
- run
- to test the visualization
- run
./main_visualize_data.py .TestScenarioForScripts
- run
- to test loading the data
- run
./main_load_track_file.py ../recorded_trackfiles/.TestScenarioForScripts/vehicle_tracks_000.csv
- run