Genetic Programming for Channel Selection from Multi-stream Sensor Data with Application on Learning Risky Driving Behaviours

Dau, H, Song, A, Xie, F, Salim, F and Ciesielski, V 2014, 'Genetic Programming for Channel Selection from Multi-stream Sensor Data with Application on Learning Risky Driving Behaviours', in Grant Dick; Will N. Browne; Peter Whigham; Mengjie Zhang; Lam Thu Bui; Hisao Ishibuchi; Yaochu Jin; Xiaodong Li; Yuhui Shi; Pramod Singh; Kay Chen Tan; Ke Tang (ed.) Proceddings of the10th International Conference 2014 (LNCS 8886), Dunedin, New Zealand, 15-18 December 2014, pp. 542-553.


Document type: Conference Paper
Collection: Conference Papers

Title Genetic Programming for Channel Selection from Multi-stream Sensor Data with Application on Learning Risky Driving Behaviours
Author(s) Dau, H
Song, A
Xie, F
Salim, F
Ciesielski, V
Year 2014
Conference name SEAL 2014: Simulated Evolution and Learning
Conference location Dunedin, New Zealand
Conference dates 15-18 December 2014
Proceedings title Proceddings of the10th International Conference 2014 (LNCS 8886)
Editor(s) Grant Dick; Will N. Browne; Peter Whigham; Mengjie Zhang; Lam Thu Bui; Hisao Ishibuchi; Yaochu Jin; Xiaodong Li; Yuhui Shi; Pramod Singh; Kay Chen Tan; Ke Tang
Publisher Springer
Place of publication Switzerland
Start page 542
End page 553
Total pages 12
Abstract Unsafe driving behaviours can put the driver himself and other people participating in the traffic at risk. Smart-phones with builtin inertial sensors offer a convenient way to passively monitor the driving patterns, from which potentially risky events can be detected. However, it is not trivial to decide which sensor data channel is relevant for the task without domain knowledge, given the growing number of sensors readily available in the phone. Using too many channels can be computationally expensive. Conversely, using too few channels may not provide sufficient information to infer meaningful patterns. We demonstrate Genetic Programming (GP) technique's capability in choosing relevant data channels directly from raw sensor data. We examine three risky driving events, namely harsh acceleration, sudden braking and swerving in the experiment. GP performance on detecting these unsafe driving behaviours is consistently high on different channel combinations that it decides to use.
Subjects Neural, Evolutionary and Fuzzy Computation
Pattern Recognition and Data Mining
Keyword(s) feature selection
channel selection
Genetic Programming
risky driving behaviours
Copyright notice © Springer International Publishing Switzerland 2014
ISBN 9783319135632
Versions
Version Filter Type
Access Statistics: 121 Abstract Views  -  Detailed Statistics
Created: Wed, 28 Jan 2015, 10:59:00 EST by Catalyst Administrator
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us