Estimation of the driving style based on the usersʹ activity and environment influence

Mikhail Sysoev (Author), Andrej Kos (Author), Jože Guna (Author), Matevž Pogačnik (Author)

Abstract

New models and methods have been designed to predict the influence of the user’s environment and activity information to the driving style in standard automotive environments. For these purposes, an experiment was conducted providing two types of analysis: (i) the evaluation of a self-assessment of the driving style; (ii) the prediction of aggressive driving style based on drivers’ activity and environment parameters. Sixty seven h of driving data from 10 drivers were collected for analysis in this study. The new parameters used in the experiment are the car door opening and closing manner, which were applied to improve the prediction accuracy. An Android application called Sensoric was developed to collect low-level smartphone data about the users’ activity. The driving style was predicted from the user’s environment and activity data collected before driving. The prediction was tested against the actual driving style, calculated from objective driving data. The prediction has shown encouraging results, with precision values ranging from 0.727 up to 0.909 for aggressive driving recognition rate. The obtained results lend support to the hypothesis that user’s environment and activity data could be used for the prediction of the aggressive driving style in advance, before the driving starts.

Keywords

agresivna vožnja;uporabniški kontekst;uporabniška aktivnost;predikcija stila vožnje;aggressive driving;user environment data;activity data;driving style prediction;

Data

Language:	English
Year of publishing:	2017
Typology:	1.01 - Original Scientific Article
Organization:	UL FE - Faculty of Electrical Engineering
UDC:	004.9:629.072
COBISS:	11867476
ISSN:	1424-8220
Views:	157
Downloads:	62
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	Slovenian
Secondary keywords:	agresivna vožnja;uporabniški kontekst;uporabniška aktivnost;predikcija stila vožnje;
Type (COBISS):	Article
Pages:	str. 1-15
Volume:	ǂno. ǂ10
Issue:	2404
Chronology:	Oct. 2017
DOI:	10.3390/s17102404
ID:	13505854