Drivers' Activity Tracking With JINS MEME Smart Glasses
The dataset contains the signal recording acquired on vehicle (car) drivers (ten experienced drivers and ten learner drivers) on the same 28.7 km route in the Silesian Voivodeship (in Polish województwo śląskie) in southern Poland. Experienced drivers performed the tasks in their own cars whereas the learner drivers performed the tasks under a supervison of a driving instructor in a specially marked cars (with L sign). The tasks include driving through a freeway (motorway), city traffic in Bytom, Radzionków, Piekary Śląskie, Tarnowskie Góry, driving in residential areas, driving on a roundabout, parking and driving through an intersection.
If you want to use this dataset, please cite the associating article:
Doniec, R.J.; Sieciński, S.; Duraj, K.M.; Piaseczna, N.J.; Mocny-Pachońska, K.; Tkacz, E.J. Recognition of Drivers’ Activity Based on 1D Convolutional Neural Network. Electronics 2020, 9(12), 2002. doi:10.3390/electronics9122002
The documentation of code is contained in the README.md file in the code archive.
Extract the archive file. The data available in CSV files are divided into four categories (traffic, roundabout, intersection, parking) and two groups of drivers (experienced and learners).
Each CSV file name follows the pattern:
category_variant_number of driver
Translations of categories and variants:
skrzyzowanie - intersection
rondo - roundabout
ruch - traffic
parkowanie - parking
lewo - left turn
prawo - right turn
prosto - straight ahead
miasto - city
miasto_niezabudowane - city outside built-up area
autostrada - freeway (motorway)
równoległe - parallel
prostopadłe - perpendicular
skos - slant
osiedle - neighbo(u)rhood
The CSV files contain date and time of the sample (timestamp), acceleration (X, Y, Z axis), gyration (X, Y, Z axis) and EOG signal (Left, Right, Horizontal, Vertical) samples. One line (after the header) represents one sample.