Driver's Speed Decision-Making Model Based on ANFIS

Abstract:

To simulate the driver's ability to deal with uncertainty and solve the unsmooth problem in the driving-status-transformation between free-traveling and car-following during the microscopic traffic simulation, the Adaptive Neuro-Fuzzy Inference System (ANFIS) was introduced to model the driver's speed decision-making behavior which integrated the free-traveling and car-following behavior. The difference between velocity and desired speed was added into the inputs of the ANFIS model besides vehicle speed, relative distance and relative velocity which commonly appeared in car-following models. In this paper, the NGSIM (Next Generation Simulation) data was used to calibrate and evaluate the model. With the analysis and pretreatment of NGSIM data, drivers reaction time was calibrated, drivers were clustered into three categories according to the level of recklessness, and the desired speed of different driver characteristic in different vehicle was approximated as the corresponding free speed. Using the processed NGSIM data, the ANFIS model was trained and the model output was validated and compared with the original data. The results showed that the ANFIS model performed well. In addition, the output of ANFIS model under car-following state was compared with that of GM model. This comparison provided a better chance to analyze the performance of the model and showed that the model simulation the driving data in a more realistic way.

• Periodical: Applied Mechanics and Materials (Volumes 488-489)
• Pages: 955-960
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.488-489.955
• Online since: January 2014
• Authors: Lian Li*, Song Yang, Wen Jing Cao
• Keywords: ANFIS Model, Car-Following Model, Driver's Speed Decision-Making, Reaction Time

References

[1] L. DongHui, S. Hu: Implement of Micro Traffic Model in Virtual Environment. Journal of Southwest Jiaotong University. Vol. 48-1(2013), pp.1-8.

[2] A. Ghaffari, A. Khodayari, S. Arvin, et al: Lane Change Trajectory Model Considering the Driver Effects Based on MANFIS. International Journal of Automotive Engineering. Vol. 2-4(2012), pp.261-275.

[3] V. Seydi Ghomsheh, M. Aliyari Shoorehdeli, M. Teshnehlab: Training ANFIS Structure with Modified PSO Algorithm. Proceedings of the 15th Mediterranean Conference on Control and Automation, July 27-29, 2007, Athens-Greece.

DOI: 10.1109/med.2007.4433927

[4] A. Khodayari, A. Ghaffari, R. Kazemi, N. Manavizadeh: Modeling and Intelligent Control Design of Car Following Behavior in Real Traffic Flow. IEEE Conference on Cybernetics and Intelligent Systems, (2010).

DOI: 10.1109/iccis.2010.5518546

[5] US Department of Transportation, NGSIM-Next Generation Simulation, ngsim. fhwa. dot. gov, (2009).

[6] L. Siwen, W. Junhua: Freeway Car-following Model and Simulation Based on Adaptive Neuro-Fuzzy Inference System. Vol. 38-7(2010), pp.1018-1022.

[7] Q. Jin: Research on Parameters Calibration and Verification of Car-following Models. Shanghai China: Shanghai JiaoTong University, (2008).

[8] W. Lurong: Application of Grey Relational Analysis on Comprehensive Evaluation about China Traffic Accidents. Mathematics in Practice and Theory. Vol. 39-22(2009), pp.68-74.

[9] K. MOTEGI, K. SHINKAI, H. Uesu, et al: Fuzzy Cluster Analysis and Its Application on International Stock Prices. Third International Conference on Innovationsin Bio-Inspired Computing and Applications, (2012).

DOI: 10.1109/ibica.2012.50

[10] W. Xiaoyuan, Z. Jinglei, X. Li: Desired speed model based on integrated calculation of human-vehicle-road-environment. Computer Engineering and Applications, Vol. 48-1(2012), pp.223-227.

[11] W. Dianhai, T. Pengfei, JIN Sheng, et al: Method of calibrating and validating car-following model. Journal of Jilin University(Engineering and Technology Edition). Vol. 41-1(2011), pp.59-65.

[12] L. Yang, Y. Duancai, T. Guojin: Hybrid genetic algorithm for solving systems of nonlinear equations. Chinese Journal of Computation Mechanics. Vol. 22-1(2005), pp.109-114.