Understanding Individualization Driving States via Latent Dirichlet Allocation Model

Automatic driving technology has become one of the hottest research topics of the Intelligent Transportation System (ITS) and Artificial Intelligence (AI) in the recent years. The development of automatic driving technology can be promoted through understanding driving states of each driver (individualization driving). Although some methods for driving states understanding are proposed by previous studies, the latent driving states and structured driving behaviors has not yet been automatically discovered. The purpose of this study is to develop an unsupervised method for deeply understanding the individualization driving. First, an encode method is proposed to extracted driving behavior from vehicle motion data. Then, Latent Dirichlet Allocation (LDA) model is innovatively developed to understand latent driving states and quantified structure of the driving behavior patterns (topics) from individualization driving (documents) using driving behaviors (words). In order to validate the performance and effectiveness of the proposed method, twenty-two drivers (15 males and 7 females) were recruited to carry out road experiments in Wuhan, China for experiments data collection. In addition, two typical unsupervised methods, including k-means and the random method are established and their performances are compared in our experiments. Experimental results verify the superiority of proposed method compared with other methods.

[1]  Gregory D. Abowd,et al.  Driver Classification Based on Driving Behaviors , 2016, IUI.

[2]  Ziyou Gao,et al.  Dynamic passenger demand oriented metro train scheduling with energy-efficiency and waiting time minimization: Mixed-integer linear programming approaches , 2017 .

[3]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[4]  Chaozhong Wu,et al.  Classification of vessel motion pattern in inland waterways based on Automatic Identification System , 2018, Ocean Engineering.

[5]  Hongwen He,et al.  Continuous reinforcement learning of energy management with deep Q network for a power split hybrid electric bus , 2018, Applied Energy.

[6]  Bin Ran,et al.  Vehicle Behavior Learning via Sparse Reconstruction with $\ell_{2}-\ell_{p}$ Minimization and Trajectory Similarity , 2017, IEEE Transactions on Intelligent Transportation Systems.

[7]  Rok Sosic,et al.  Driver identification using automobile sensor data from a single turn , 2016, 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC).

[8]  Wenjia Li,et al.  Driver identification and authentication with active behavior modeling , 2016, 2016 12th International Conference on Network and Service Management (CNSM).

[9]  Anthony G. Cohn,et al.  Latent Dirichlet Allocation for Unsupervised Activity Analysis on an Autonomous Mobile Robot , 2017, AAAI.

[10]  Jian-Da Wu,et al.  Driver identification based on voice signal using continuous wavelet transform and artificial neural network techniques , 2009, Expert Syst. Appl..

[11]  Faiyaz Doctor,et al.  A real-time driver identification system based on artificial neural networks and cepstral analysis , 2014, 2014 International Joint Conference on Neural Networks (IJCNN).

[12]  Julien Jacques,et al.  Funclust: A curves clustering method using functional random variables density approximation , 2013, Neurocomputing.

[13]  Thomas Hofmann,et al.  Unsupervised Learning by Probabilistic Latent Semantic Analysis , 2004, Machine Learning.

[14]  Vili Podgorelec,et al.  Text classification method based on self-training and LDA topic models , 2017, Expert Syst. Appl..

[15]  Jamal Hussain Shah,et al.  Facial expressions classification and false label reduction using LDA and threefold SVM , 2017, Pattern Recognit. Lett..

[16]  Reza Malekian,et al.  Hierarchical Fuzzy Logic-Based Variable Structure Control for Vehicles Platooning , 2019, IEEE Transactions on Intelligent Transportation Systems.

[17]  Ying Fu,et al.  Automated classification of software change messages by semi-supervised Latent Dirichlet Allocation , 2015, Inf. Softw. Technol..

[18]  Honghai Li,et al.  Driving style estimation by fusing multiple driving behaviors: a case study of freeway in China , 2018, Cluster Computing.

[19]  Hakan Erdogan,et al.  Video-based driver identification using local appearance face recognition , 2007 .

[20]  Chien-Hsing Chou,et al.  Short Papers , 2001 .

[21]  Bin Ran,et al.  Short-Term Traffic Prediction Based on Dynamic Tensor Completion , 2016, IEEE Transactions on Intelligent Transportation Systems.

[22]  Yiman Du,et al.  Leveraging longitudinal driving behaviour data with data mining techniques for driving style analysis , 2015 .

[23]  Nuno Vasconcelos,et al.  Latent Dirichlet Allocation Models for Image Classification , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[24]  Alois Ferscha,et al.  Supporting Implicit Human-to-Vehicle Interaction: Driver Identification from Sitting Postures , 2008 .

[25]  Fei Peng,et al.  Robust Coverless Image Steganography Based on DCT and LDA Topic Classification , 2018, IEEE Transactions on Multimedia.