Driving Behavior Analysis Based On Spectral Cluster Algorithm

Paper #:
  • 2017-01-1980

Published:
  • 2017-09-23
Abstract:
The feature of driving behavior is vital to the design of advance driving assistant system (ADAS). Accurate driving behavior can help the ADAS to make proper assistant strategy and make the ADAS in harmony with the driver’s driving style. Traditional driving behavior analysis focuses on the detail driving process modelling which is always complex and not general to the various driving environment. So in this paper we propose a data mining method which can evaluate the driving behavior from the various driving data. First, we introduce the spectral cluster algorithm which is a machine learning method aiming at data cluster. Besides we also give the improved parallel calculating form of the spectral cluster in order to deal with the large amount of driving data. Second, we design the experiment under the fixed traffic condition and describe a data collecting platform to obtain the experimental driving data. At last, the proposed spectral cluster algorithm is used to classify the experimental driving data of the testing drivers into three groups and each group shows similar driving behavior or driving style. We give the detail cluster result and compare each group’s driving behavior feature to certify the efficiency of driving behavior analysis mechanism. The final result shows that the proposed method in this paper can achieve a well driving behavior classifier without the detail modelling of the driving behavior. In the end of this paper we discuss the future work we need to deepen this research direction.
Access
Now
SAE MOBILUS Subscriber? You may already have access.
Buy
Attention: This item is not yet published. Pre-Order to be notified, via email, when it becomes available.
Select
Price
List
Download
$27.00
Mail
$27.00
Members save up to 40% off list price.
Share
HTML for Linking to Page
Page URL

Related Items

Training / Education
2011-04-09
Training / Education
2013-04-09
Article
2016-08-15
Training / Education
2013-04-09
Article
2016-08-15