The process of urbanisation affects various aspects of human society, of which transportation is no exception. Different forms of mobility are offered to address the transportation challenges, such as congestion, parking spaces and pollution. Mobility as a Service (MaaS), a novel transportation paradigm, aims to improve the usability of the transportation network by combining different forms of travel. Currently, the most widespread option of travel are cars. Therefore, a possible approach to improve the sustainability aspect of MaaS solutions is to get a better understanding of car driver behaviour, such as driving style categorisation. This paper aims to present an advanced analytical study of contextually enriched automotive data set gathered from cars. Specifically, an unsupervised learning algorithm (i.e., k-means clustering) was used to identify patterns in the trips of different car drivers. The clustering process identified three driving styles among analysed car drivers. Identified driving styles were further analysed from the automotive, traffic, road and time perspectives, to better understand the sustainability aspect of transportation.
Driving style Categorisation based on Unsupervised Learning: a Step towards Sustainable Transportation
2021-09-08
2103844 byte
Conference paper
Electronic Resource
English
MATERIAL CATEGORISATION AND TRANSPORTATION SYSTEMS AND METHODS
European Patent Office | 2023
|Behaviourally relevant road categorisation: A step towards self-explaining rural roads
Online Contents | 2008
|MATERIAL CATEGORISATION AND TRANSPORTATION SYSTEMS AND METHODS
European Patent Office | 2023
|MATERIAL CATEGORISATION AND TRANSPORTATION SYSTEMS AND METHODS
European Patent Office | 2024
|Behaviourally relevant road categorisation: A step towards self-explaining rural roads
Online Contents | 2008
|