A Traffic Flow Simulation Framework for Learning Driver Heterogeneity from Naturalistic Driving Data using Autoencoders
This paper proposes a novel data-centric framework for microscopic traffic flow simulation with intra and inter driver heterogeneity. We utilized a naturalistic driving corpus of 46 different drivers to learn and model the behavior divergence of Japanese drivers. First, ego-driver behavior signals a...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Society of Automotive Engineers of Japan, Inc.
2019-01-01
|
| Series: | International Journal of Automotive Engineering |
| Online Access: | https://www.jstage.jst.go.jp/article/jsaeijae/10/1/10_20194087/_article/-char/ja |
| _version_ | 1797356793765036032 |
|---|---|
| author | Ekim Yurtsever Chiyomi Miyajima Kazuya Takeda |
| author_facet | Ekim Yurtsever Chiyomi Miyajima Kazuya Takeda |
| author_sort | Ekim Yurtsever |
| collection | DOAJ |
| description | This paper proposes a novel data-centric framework for microscopic traffic flow simulation with intra and inter driver heterogeneity. We utilized a naturalistic driving corpus of 46 different drivers to learn and model the behavior divergence of Japanese drivers. First, ego-driver behavior signals are used to extract unique features of each driver with an auto-encoder. Then, using these features, drivers are divided into groups using unsupervised clustering algorithms. For each driver group, a feedforward neural network is trained for predicting the desired speed given the road topology. The trained network is then used in a microscopic traffic flow model for simulations. We used a macroscopic traffic survey conducted in Japan to evaluate the proposed framework. Our findings indicate that the proposed framework can simulate a realistic traffic flow with high driver heterogeneity. |
| first_indexed | 2024-03-08T14:32:43Z |
| format | Article |
| id | doaj.art-2bd2726fa54342a1a5134b01040322fa |
| institution | Directory Open Access Journal |
| issn | 2185-0992 |
| language | English |
| last_indexed | 2024-03-08T14:32:43Z |
| publishDate | 2019-01-01 |
| publisher | Society of Automotive Engineers of Japan, Inc. |
| record_format | Article |
| series | International Journal of Automotive Engineering |
| spelling | doaj.art-2bd2726fa54342a1a5134b01040322fa2024-01-12T06:42:40ZengSociety of Automotive Engineers of Japan, Inc.International Journal of Automotive Engineering2185-09922019-01-01101869310.20485/jsaeijae.10.1_86A Traffic Flow Simulation Framework for Learning Driver Heterogeneity from Naturalistic Driving Data using AutoencodersEkim Yurtsever0Chiyomi Miyajima1Kazuya Takeda2Nagoya University, Graduate School of Information ScienceNagoya University, Graduate School of Information ScienceNagoya University, Graduate School of Information ScienceThis paper proposes a novel data-centric framework for microscopic traffic flow simulation with intra and inter driver heterogeneity. We utilized a naturalistic driving corpus of 46 different drivers to learn and model the behavior divergence of Japanese drivers. First, ego-driver behavior signals are used to extract unique features of each driver with an auto-encoder. Then, using these features, drivers are divided into groups using unsupervised clustering algorithms. For each driver group, a feedforward neural network is trained for predicting the desired speed given the road topology. The trained network is then used in a microscopic traffic flow model for simulations. We used a macroscopic traffic survey conducted in Japan to evaluate the proposed framework. Our findings indicate that the proposed framework can simulate a realistic traffic flow with high driver heterogeneity.https://www.jstage.jst.go.jp/article/jsaeijae/10/1/10_20194087/_article/-char/ja |
| spellingShingle | Ekim Yurtsever Chiyomi Miyajima Kazuya Takeda A Traffic Flow Simulation Framework for Learning Driver Heterogeneity from Naturalistic Driving Data using Autoencoders International Journal of Automotive Engineering |
| title | A Traffic Flow Simulation Framework for Learning Driver Heterogeneity from Naturalistic Driving Data using Autoencoders |
| title_full | A Traffic Flow Simulation Framework for Learning Driver Heterogeneity from Naturalistic Driving Data using Autoencoders |
| title_fullStr | A Traffic Flow Simulation Framework for Learning Driver Heterogeneity from Naturalistic Driving Data using Autoencoders |
| title_full_unstemmed | A Traffic Flow Simulation Framework for Learning Driver Heterogeneity from Naturalistic Driving Data using Autoencoders |
| title_short | A Traffic Flow Simulation Framework for Learning Driver Heterogeneity from Naturalistic Driving Data using Autoencoders |
| title_sort | traffic flow simulation framework for learning driver heterogeneity from naturalistic driving data using autoencoders |
| url | https://www.jstage.jst.go.jp/article/jsaeijae/10/1/10_20194087/_article/-char/ja |
| work_keys_str_mv | AT ekimyurtsever atrafficflowsimulationframeworkforlearningdriverheterogeneityfromnaturalisticdrivingdatausingautoencoders AT chiyomimiyajima atrafficflowsimulationframeworkforlearningdriverheterogeneityfromnaturalisticdrivingdatausingautoencoders AT kazuyatakeda atrafficflowsimulationframeworkforlearningdriverheterogeneityfromnaturalisticdrivingdatausingautoencoders AT ekimyurtsever trafficflowsimulationframeworkforlearningdriverheterogeneityfromnaturalisticdrivingdatausingautoencoders AT chiyomimiyajima trafficflowsimulationframeworkforlearningdriverheterogeneityfromnaturalisticdrivingdatausingautoencoders AT kazuyatakeda trafficflowsimulationframeworkforlearningdriverheterogeneityfromnaturalisticdrivingdatausingautoencoders |