This content is not included in your SAE MOBILUS subscription, or you are not logged in.
Real-World Driving Features for Identifying Intelligent Driver Model Parameters
ISSN: 0148-7191, e-ISSN: 2688-3627
Published April 06, 2021 by SAE International in United States
This content contains downloadable datasetsAnnotation ability available
Event: SAE WCX Digital Summit
Driver behavior models play a significant role in representing different driving styles and the associated relationships with traffic patterns and vehicle energy consumption in simulation studies. The models often serve as a proxy for baseline human driving when assessing energy-saving strategies that alter vehicle velocity. Such models are especially important in connectivity-enabled energy-saving strategy research because they can easily adapt to changing driving conditions like posted speed limits or change in traffic light state.
While numerous driver models exist, parametric driver models provide the flexibility required to represent variability in real-world driving through different combinations of model parameters. These model parameters must be informed by a representative set of parameter values for the driver model to adequately represent a real-world driver. It stands to reason that determining the parameter values from real-world driving data would serve the purpose of representing a real-world driver. Although the published literature is replete with techniques and consequences of using real-world driving data to determine parameter values for parametric driver models, none have explored them in the context of using shorter driving features where the parameter values may change over the course of a single trip for the same driver.
In this study we consider the “intelligent driver model” (IDM) as our driver behavior model and use real-world driving data from the Transportation Secure Data Center (TSDC) maintained at the National Renewable Energy Laboratory (NREL). The TSDC includes real-world travel data from across the United States, from which NREL has created a wide range of driving routes consisting of road features such as speed limits, stop locations, and turn locations. The real-world driving data are categorized into different driving regimes and extracted into driving features. The driving features are then used to calibrate the parameter values for the IDM. The distribution of the parameters and the relationships among them are reported. The insights obtained from this study enable judicious usage of IDM or similar parametric driver models to represent baseline human driver behavior in simulations.
CitationHegde, B., O'Keefe, M., Muldoon, S., Gonder, J. et al., "Real-World Driving Features for Identifying Intelligent Driver Model Parameters," SAE Technical Paper 2021-01-0436, 2021, https://doi.org/10.4271/2021-01-0436.
Data Sets - Support Documents
|Unnamed Dataset 1|
|Unnamed Dataset 2|
|Unnamed Dataset 3|
- Zhao , J. , Chang , C.F. , Rajkumar , R. , and Gonder , J. Corroborative Evaluation of the Real-World Energy Saving Potentials of InfoRich Eco-Autonomous Driving (iREAD) System SAE Technical Paper 2020-01-0588 2020 https://doi.org/10.4271/2020-01-0588
- Vergeest , J. and Van Arem , B. The Effect of Vehicle Acceleration near Traffic Congestion Fronts IEEE Intelligent Vehicles Symposium 2012 45 50
- Hegde , B. , Rajendran , A.V. , Ahmed , Q. , and Rizzoni , G. On Quantifying the Utility of Look-Ahead Data for Energy Management IFAC-PapersOnLine 2018 10.1016/j.ifacol.2018.10.011
- Kesting , A. , and Treiber , M. Calibrating Car-Following Models by Using Trajectory Data Methodological Study Transportation Research Record 148 156 2008 10.3141/2088-16
- Chen , C. , Li , L. , Hu , J. , and Geng , C. Calibration of MITSIM and IDM Car-Following Model Based on NGSIM Trajectory Datasets Proceedings of 2010 IEEE International Conference on Vehicular Electronics and Safety (ICVES 2010) 2010 48 53
- Gupta , S. , Rajakumar Deshpande , S. , Tufano , D. , Canova , M. et al. Estimation of Fuel Economy on Real-World Routes for Next-Generation Connected and Automated Hybrid Powertrains SAE Technical Paper 2020-01-0593 2020 https://doi.org/10.4271/2020-01-0593
- Monteil , J. , Billot , R. , Sau , J. , Buisson , C. , and El Faouzi , N.-E. Calibration, Estimation, and Sampling Issues of Car-Following Parameters Transportation Research Record: Journal of the Transportation Research Board Jan. 2014 131 140 10.3141/2422-15
- James , R.M. , Hammit , B.E. , and Boyles , S.D. Methods to Obtain Representative Car-Following Model Parameters from Trajectory-Level Data for Use in Microsimulation Transportation Research Record: Journal of the Transportation Research Board 62 73 Jul. 2019 10.1177/0361198119849401.
- Chen , X. , Sun , J. , Ma , Z. , Sun , J. , and Zheng , Z. Investigating the long- and short-term driving characteristics and incorporating them into car-following models Transportation Research Part C: Emerging Technologies 102698 Aug. 2020 10.1016/j.trc.2020.102698.
- National Renewable Energy Laboratory 2019
- Hegde , B. , Ahmed , Q. , and Rizzoni , G. Velocity and energy trajectory prediction of electrified powertrain for look ahead control Applied Energy 115903 Dec. 2020 10.1016/j.apenergy.2020.115903
- Kapoor , A. and Singhal , A. A Comparative Study of K-Means, K-Means++ and Fuzzy C-Means Clustering Algorithms 3rd IEEE International Conference on Computational Intelligence & Communication Technology (CICT) 2017
- Hegde , B. , Muldoon , S.E. , O’Keefe , M. , and Gonder , J. Leveraging Real-World Driving Data for Design and Impact Evaluation of Energy Efficient Control Strategies SAE Technical Paper 2020-01-0585 2020 https://doi.org/10.4271/2020-01-0585