Physics-Informed Machine Learning for Calibrating Macroscopic Traffic Flow Models

References

• Aw A, Rascle M (2000) Resurrection of “second order” models of traffic flow. SIAM J. Appl. Math. 60(3):916–938.Crossref, Google Scholar
• Bartin B, Ozbay K, Gao J, Kurkcu A (2018) Calibration and validation of large-scale traffic simulation networks: A case study. Proc. Comput. Sci. 130:844–849.Crossref, Google Scholar
• Beintema G, Toth R, Schoukens M (2021) Nonlinear state-space identification using deep encoder networks. Learning for Dynamics and Control (PMLR, New York), 241–250.Google Scholar
• Caltrans (2022) Caltrans performance measurement system. Acessed December 12, 2022, https://pems.dot.ca.gov/?logout=1.Google Scholar
• Chau M, Fu MC (2014) An overview of stochastic approximation. Fu MC, ed. Handbook of Simulation Optimization (Springer, New York), 149–178.Google Scholar
• Dervisoglu G, Gomes G, Kwon J, Horowitz R, Varaiya P (2009) Automatic calibration of the fundamental diagram and empirical observations on capacity. Proc. Transportation Res. Board 88th Annual Meeting, vol. 15 (Transportation Research Board, Washington, DC), 31–59.Google Scholar
• Di X, Shi R, Mo Z, Fu Y (2023) Physics-informed deep learning for traffic state estimation: A survey and the outlook. Algorithms (Basel) 16(6):305.Crossref, Google Scholar
• Dowling R, Skabardonis A, Carroll M, Wang Z (2004) Methodology for measuring recurrent and nonrecurrent traffic congestion. Transportation Res. Rec. 1:60–68.Crossref, Google Scholar
• Engl HW, Hanke M, Neubauer A (1996) Regularization of Inverse Problems, vol. 375 (Springer Science and Business Media, Boston).Crossref, Google Scholar
• Ferrara A, Sacone S, Siri S (2018) Freeway Traffic Modelling and Control (Springer, Berlin).Crossref, Google Scholar
• Gedon D, Wahlström N, Schön TB, Ljung L (2021) Deep state space models for nonlinear system identification. IFAC PapersOnLine 54(7):481–486.Crossref, Google Scholar
• Gomes G, Horowitz R (2006) Optimal freeway ramp metering using the asymmetric cell transmission model. Transportation Res. Part C Emerging Tech. 14(4):244–262.Crossref, Google Scholar
• GoogleMap (2022) Interstate 210. Accessed December 12, 2022, https://www.google.com/maps/place/I-210,+Los+Angeles,+CA/@34.1415053,-118.054693,13.2z/data=!4m6!3m5!1s0x80c2e82ee01a1731:0x3d8b10f8674c2e28!8m2!3d34.2428723!4d-118.3169267!16zL20vMDVoNGIx?entry=ttu.Google Scholar
• Hinton GE, Salakhutdinov RR (2006) Reducing the dimensionality of data with neural networks. Science 313(5786):504–507.Crossref, Google Scholar
• Huang J, Agarwal S (2020) Physics informed deep learning for traffic state estimation. Proc. IEEE 23rd Internat. Conf. Intelligent Transportation Systems (IEEE, New York), 1–6.Google Scholar
• Huang K, Di X, Du Q, Chen X (2020) Scalable traffic stability analysis in mixed-autonomy using continuum models. Transportation Res. Part C Emerging Tech. 111:616–630.Crossref, Google Scholar
• Jaques M, Burke M, Hospedales T (2020) Physics-as-inverse-graphics: Unsupervised physical parameter estimation from video. Proc. 8th Internat. Conf. Learn. Representations (ICLR, Appleton, WI), 1–16.Google Scholar
• Khoshyaran MM, Lebacque JP (2015) Capacity drop and traffic hysteresis as a consequence of bounded acceleration. IFAC-PapersOnLine 48(1):766–771.Crossref, Google Scholar
• Kidger P (2021) On neural differential equations. PhD thesis, University of Oxford, Oxford, UK.Google Scholar
• Kontorinaki M, Spiliopoulou A, Roncoli C, Papageorgiou M (2017) First-order traffic flow models incorporating capacity drop: Overview and real-data validation. Transportation Res. Part B Methodological 106:52–75.Crossref, Google Scholar
• Kovachki N, Li Z, Liu B, Azizzadenesheli K, Bhattacharya K, Stuart A, Anandkumar A (2021) Neural operator: Learning maps between function spaces. Preprint, submitted August 19, https://arxiv.org/abs/2108.08481.Google Scholar
• Laval JA, Leclercq L (2010) A mechanism to describe the formation and propagation of stop-and-go waves in congested freeway traffic. Philosophical Trans. Roy. Soc. A 368:4519–4541.Crossref, Google Scholar
• LeCun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proc. IEEE 86(11):2278–2324.Crossref, Google Scholar
• Lee JB, Ozbay K (2009) New calibration methodology for microscopic traffic simulation using enhanced simultaneous perturbation stochastic approximation approach. Transportation Res. Rec. 2124(1):233–240.Crossref, Google Scholar
• Lu J, Li C, Wu XB, Zhou XS (2023) Physics-informed neural networks for integrated traffic state and queue profile estimation: A differentiable programming approach on layered computational graphs. Transportation Res. Part C Emerging Tech. 153:104224.Crossref, Google Scholar
• Masci J, Meier U, Cireşan D, Schmidhuber J (2011) Stacked convolutional auto-encoders for hierarchical feature extraction. Proc. 21st Internat. Conf. Artificial Neural Networks (Springer, Berlin), 52–59.Google Scholar
• Masti D, Bemporad A (2021) Learning nonlinear state–space models using autoencoders. Automatica J. IFAC 129:109666.Crossref, Google Scholar
• Messner A, Papageorgiou M (1990) Metanet: A macroscopic simulation program for motorway networks. Traffic Engrg. Control 31(8–9):466–470.Google Scholar
• Mo Z, Fu Y, Di X (2022) Quantifying uncertainty in traffic state estimation using generative adversarial networks. Proc. IEEE 25th Internat. Conf. Intelligent Transportation Systems (IEEE, New York), 2769–2774.Google Scholar
• Mo Z, Fu Y, Xu D, Di X (2022) Trafficflowgan: Physics-informed flow based generative adversarial network for uncertainty quantification. Proc. Joint Eur. Conf. Machine Learn. Knowledge Discovery Databases (Springer, Berlin), 323–339.Google Scholar
• Mohammadian S, Zheng Z, Haque MM, Bhaskar A (2021) Performance of continuum models for realworld traffic flows: Comprehensive benchmarking. Transportation Res. Part B Methodological 147:132–167.Crossref, Google Scholar
• Mudigonda S, Ozbay K (2015a) Robust calibration of macroscopic traffic simulation models using stochastic collocation. Transportation Res. Part C Emerging Tech. 59:358–374.Crossref, Google Scholar
• Mudigonda S, Ozbay K (2015b) Robust calibration of macroscopic traffic simulation models using stochastic collocation. Transportation Res. Proc. 9:1–20.Crossref, Google Scholar
• Muñoz L, Sun X, Sun D, Gomes G, Horowitz R (2004) Methodological calibration of the cell transmission model. Proc. Amer. Control Conf., vol. 1 (IEEE, New York), 798–803.Google Scholar
• Nagel T, Huber MF (2021) Autoencoder-inspired identification of LTI systems. Proc. Eur. Control Conf. (IEEE, New York), 2352–2357.Google Scholar
• Ozbay K, Yang H, Morgul EF, Mudigonda S, Bartin B (2014) Big data and the calibration and validation of traffic simulation models. Proc. Traffic Transportation Simulation-Looking Back Looking Ahead Celebrating 50 Years Traffic Flow Theory Workshop (Transportation Research Board, Washington, DC), 92–122.Google Scholar
• Papamichail I, Kotsialos A, Margonis I, Papageorgiou M (2010) Coordinated ramp metering for freeway networks: A model-predictive hierarchical control approach. Transportation Res. Part C Emerging Tech. 18(3):311–331.Crossref, Google Scholar
• Pascanu R, Mikolov T, Bengio Y (2013) On the difficulty of training recurrent neural networks. Proc. Internat. Conf. Machine Learn. (PMLR, New York), 1310–1318.Google Scholar
• Payne H (1971) Models of Freeway Traffic and Control. Mathematical Models of Public Systems (Simulation Councils, Vista, CA).Google Scholar
• Schoukens J, Ljung L (2019) Nonlinear system identification: A user-oriented road map. IEEE Control Systems Magazine 39(6):28–99.Crossref, Google Scholar
• Sha D, Ozbay K, Ding Y (2020) Applying Bayesian optimization for calibration of transportation simulation models. Transportation Res. Rec. 2674(10):215–228.Crossref, Google Scholar
• Sha D, Gao J, Yang D, Zuo F, Ozbay K (2023) Calibrating stochastic traffic simulation models for safety and operational measures based on vehicle conflict distributions obtained from aerial and traffic camera videos. Accident Anal. Prevention 179:106878.Crossref, Google Scholar
• Shi X, Li X (2021) Constructing a fundamental diagram for traffic flow with automated vehicles: Methodology and demonstration. Transportation Res. Part B Methodological 150:279–292.Crossref, Google Scholar
• Shi R, Mo Z, Di X (2021) Physics-informed deep learning for traffic state estimation: A hybrid paradigm informed by second-order traffic models. Proc. Conf. AAAI Artificial Intelligence 35:540–547.Crossref, Google Scholar
• Shi R, Mo Z, Huang K, Di X, Du Q (2022) A physics-informed deep learning paradigm for traffic state and fundamental diagram estimation. IEEE Trans. Intelligent Transportation Systems 23(8):11688–11698.Crossref, Google Scholar
• Spiliopoulou A, Kontorinaki M, Papageorgiou M, Kopelias P (2014) Macroscopic traffic flow model validation at congested freeway off-ramp areas. Transportation Res. Part C Emerging Tech. 41:18–29.Crossref, Google Scholar
• Spiliopoulou A, Papamichail I, Papageorgiou M, Tyrinopoulos Y, Chrysoulakis J (2017) Macroscopic traffic flow model calibration using different optimization algorithms. Oper. Res. 17(1):145–164.Google Scholar
• Vincent P, Larochelle H, Bengio Y, Manzagol PA (2008) Extracting and composing robust features with denoising autoencoders. Proc. 25th Internat. Conf. Machine Learn. (Association for Computing Machinery, New York), 1096–1103.Google Scholar
• Wang B, Xia H, Nguyen T, Osher S (2022a) How does momentum benefit deep neural networks architecture design? A few case studies. Res. Math. Sci. 9(3):57.Crossref, Google Scholar
• Wang Y, Yu X, Guo J, Papamichail I, Papageorgiou M, Zhang L, Hu S, et al. (2022b) Macroscopic traffic flow modelling of large-scale freeway networks with field data verification: State-of-the-art review, benchmarking framework, and case studies using Metanet. Transportation Res. Part C Emerging Tech. 145:103904.Crossref, Google Scholar
• Whitham GB (1974) Linear and Nonlinear Waves (Wiley-Interscience, New York).Google Scholar
• Work DB, Tossavainen OP, Jacobson Q, Bayen AM (2009) Lagrangian sensing: Traffic estimation with mobile devices. Proc. Amer. Control Conf. (IEEE, New York), 1536–1543.Google Scholar
• Yang TY, Rosca J, Narasimhan K, Ramadge PJ (2022) Learning physics constrained dynamics using autoencoders. Koyejo S, Mohamed S, Agarwal A, Belgrave D, Cho K, Oh A, eds. Adv. Neural Inform. Processing Systems, vol. 35 (Curran Associates, Inc., Red Hook, NY), 17157–17172.Google Scholar
• Yuan Y, Wang Q, Yang XT (2021) Traffic flow modeling with gradual physics regularized learning. IEEE Trans. Intelligent Transportation Systems. 23(9):14649–14660Google Scholar
• Yuan Y, Zhang Z, Yang XT (2020) Highway traffic state estimation using physics regularized gaussian process: Discretized formulation. Preprint, submitted July 14, https://arxiv.org/abs/2007.07762.Google Scholar
• Yuan Y, Zhang Z, Yang XT, Zhe S (2021) Macroscopic traffic flow modeling with physics regularized gaussian process: A new insight into machine learning applications in transportation. Transportation Res. Part B Methodological 146:88–110.Crossref, Google Scholar
• Yuan Y, Van Lint J, Wilson RE, van Wageningen-Kessels F, Hoogendoorn SP (2012) Real-time Lagrangian traffic state estimator for freeways. IEEE Trans. Intelligent Transportation Systems 13(1):59–70.Crossref, Google Scholar
• Zhang HM (2002) A non-equilibrium traffic model devoid of gas-like behavior. Transportation Res. Part B Methodological 36(3):275–290.Crossref, Google Scholar
• Zhong R, Chen C, Chow AH, Pan T, Yuan F, He Z (2016) Automatic calibration of fundamental diagram for first-order macroscopic freeway traffic models. J. Adv. Transportation 50(3):363–385.Crossref, Google Scholar