Abstract
There are multibody systems whose physics are partially known owing to their complexity and nonlinearity. Therefore, motion equations are not utterly available to be utilized for the prediction, control, design, and monitoring of these systems. To alleviate this issue, this study aims at developing a hybrid modeling procedure to discover respective unidentified physics and, subsequently, provide a holistic governing model of the original mechanism. For approach development, a vehicle with unmodeled tires is thoroughly considered in this research work. Tires profoundly impact the dynamics of vehicles, influencing their handling, drivability, and ride comfort. Advanced chassis control systems used to improve vehicles’ safety, performance, and reliability also require knowledge of tire behavior. Nevertheless, tires are very challenging to model as they are very complex and nonlinear components. Although simplified models are often employed, they are incapable of fully capturing tire behaviors. Using neural networks, i.e., black-box models, of the tire represents a common alternative. However, these approaches do not work outside the training data distribution, and they need costly and hard-to-measure experimental data for training purposes. Thus, this research study proposes a hybrid method by combining partially known physics of vehicle dynamics and a neural network to compensate for the unknown physics of tires. The developed approach learns the tire dynamics automatically from vehicle responses without requiring costly measured tire forces but solely relying on signals from an inertial measuring unit. The suggested methodology is validated experimentally, providing accurate and stable results. The time-depending behaviors of tires during cornering are also discovered and reported. The developed model is generic and can handle either linear or nonlinear physics-based models. However, the linear tire model integrated into the hybrid procedure in this study limits the simulation to stationary trajectories and cannot address the physics of tires when a vehicle undergoes nonstationary maneuvers.
Similar content being viewed by others
Data Availability
The data and materials presented in this study are available on request from the corresponding author.
References
Galvani, M.: History and future of driver assistance. IEEE Instrum. Meas. Mag. 22(1), 11–16 (2019). https://doi.org/10.1109/MIM.2019.8633345
Jazar, R.: Vehicle Dynamics: Theory and Application, 3rd edn. Springer, Berlin (2017)
Pereira, C.L., Costa Neto, R.T., Loiola, B.R.: Cornering stiffness estimation using Levenberg–Marquardt approach. Inverse Probl. Sci. Eng. 29(12), 2207–2239 (2021). https://doi.org/10.1080/17415977.2021.1910683
Feng, L., Chen, W., Wu, T., Wang, H., Dai, D., Wang, D., Zhang, W.: An improved sensor system for wheel force detection with motion-force decoupling technique. Measurement 119, 205–217 (2018). https://doi.org/10.1016/j.measurement.2018.01.066
Jin, X., Yin, G., Chen, N.: Advanced estimation techniques for vehicle system dynamic state: a survey. Sensors 19(19), Article ID 4289 (2019). https://doi.org/10.3390/s19194289
Pacejka, H.B., Bakker, E.: The magic formula tyre model. Veh. Syst. Dyn. 21, 1–18 (1992). https://doi.org/10.1080/00423119208969994
Smith, G., Blundell, M.: A new efficient free-rolling tyre-testing procedure for the parameterisation of vehicle dynamics tyre models. Proc. Inst. Mech. Eng., Part D, J. Automob. Eng., 1435–1448 (2017). https://doi.org/10.1177/0954407016675216
Wang, T., Liu, Y., Ding, L., Li, J., Gao, H., Liang, Y., Sun, T.: Neural network identification of a racing car tire model. J Eng., 1–11 (2018). https://doi.org/10.1155/2018/4143794
Sousa, L.C., Ayala, H.: Nonlinear tire model approximation using machine learning for efficient model predictive control. IEEE Access 10, 107549–107562 (2022). https://doi.org/10.1109/ACCESS.2022.3212420
Xu, N., Askari, H., Huang, Y., Zhou, J., Khajepour, A.: Tire force estimation in intelligent tires using machine learning. IEEE Trans. Intell. Transp. Syst. 23(4), 3565–3574 (2022). https://doi.org/10.1109/TITS.2020.3038155
Viehweger, M., Cyrano Vaseur, C., Aalst, S., Acosta, M., Regolin, E., Alatorre, A.: Vehicle state and tyre force estimation: demonstrations and guidelines. Veh. Syst. Dyn., 675–702 (2021). https://doi.org/10.1080/00423114.2020.1714672
Kalman, R.E.: A new approach to linear filtering and prediction problems. J. Basic Eng., 35–45 (1960). https://doi.org/10.1115/1.3662552
Naets, F., Aalst, S., Boulkroune, B., Ghouti, N.E., Desmet, W.: Design and experimental validation of a stable two-stage estimator for automotive sideslip angle and tire parameters. IEEE Trans. Veh. Technol. 66(1), 9727–9742 (2017). https://doi.org/10.1109/TVT.2017.2742665
Doumiati, M., Victorino, A., Charara, A., Lechner, D.: Unscented Kalman filter for real-time vehicle lateral tire forces and sideslip angle estimation. IEEE Intell. Veh. Symp., 901–906 (2009). https://doi.org/10.1109/IVS.2009.5164399
Acosta, M., Kanarachos, S.: Tire lateral force estimation and grip potential identification using neural networks, extended Kalman filter, and recursive least squares. Neural Comput. Appl. 30, 3445–3465 (2018). https://doi.org/10.1007/s00521-017-2932-9
Nikravesh, P.: Computer-Aided Analysis of Mechanical Systems. Prentice Hall, Englewood Cliffs (1988)
Bauchau, O.A.: Flexible Multibody Dynamics. Springer, Berlin (2011)
Rahnejat, H.: Multibody dynamics: historical evolution and application. J. Mech. Eng. Sci. 214, 149–173 (2000)
Askari, E., Crevecoeur, G.: Evolutionary sparse data-driven discovery of multibody system dynamics. Multibody Syst. Dyn. 58, 197–226 (2023). https://doi.org/10.1007/s11044-023-09901-z
Kibble, T., Berkshire, W.B., Frank, H.: Classical Mechanics, 5th edn. Imperial College Press, London (2004)
Askari, E., Andersen, M.S.: On the effect of friction on tibiofemoral joint kinematics. Appl. Sci. 11(16), 7516 (2021). https://doi.org/10.3390/app11167516
Askari, E., Andersen, M.S.: An anatomy-based dynamic model of total knee arthroplasty. Nonlinear Dyn. 106, 3539–3555 (2021). https://doi.org/10.1007/s11071-021-06949-4
De Groote, W., Kikken, E., Hostens, E., Hoecke, S.V., Guillaume Crevecoeur, G.: Neural network augmented physics models for systems with partially unknown dynamics: application to slider-Crank mechanism. IEEE/ASME Trans. Mechatron. 27(1), 103–114 (2022). https://doi.org/10.1109/TMECH.2021.3058536
Haykin, S.: Neural Network, a Comprehensive Foundation, 2nd edn. Pearson Education, Upper Saddle River (1994)
Ardeh, H.A., Tupy, M., Negrut, D.: On the construction and use of surrogate models for the dynamic analysis of multibody systems. In: Volume 13: New Developments in Simulation Methods and Software for Engineering Applications; Safety Engineering, Risk Analysis and Reliability Methods; Transporta- Tion Systems, ASMEDC, vol. 13, pp. 17–26 (2009). https://doi.org/10.1115/IMECE2009-10277
Azzam, B., Schelenz, R., Roscher, B., Baseer, A., Jacobs, G.: Development of a wind turbine gear- box virtual load sensor using multibody simulation and artificial neural networks. Forsch. Ingenieurwes./Eng. Res. 85, 241–250 (2021). https://doi.org/10.1007/s10010-021-00460-3
Kahr, M., Kovács, G., Loinig, M., Brückl, H.: Condition monitoring of ball bearings based on machine learning with synthetically generated data. Sensors 22(7), 2490 (2022). https://doi.org/10.3390/s22072490
Ogunmolu, O., Gu, X., Jiang, S., Gans, N.: Nonlinear systems identification using deep dynamic neural networks (2016). arXiv:1610.01439
Nasr, A., Inkol, K.A., Bell, S., McPhee, J.: InverseMuscleNET: alternative machine learning solution to static optimization and inverse muscle modeling. Front. Comput. Neurosci. 15, 759489 (2021). https://doi.org/10.3389/fncom.2021.759489
Ye, Y., Huang, P., Sun, Y., Shi, D.: MBSNet: a deep learning model for multibody dynamics simulation and its application to a vehicle-track system. Mech. Syst. Signal Process. 157, 107716 (2021). https://doi.org/10.1016/j.ymssp.2021.107716
Mohajerin, N., Waslander, S.L.: Multistep prediction of dynamic systems with recurrent neural networks. IEEE Trans. Neural Netw. Learn. Syst. 30(11), 3370–3383 (2019). https://doi.org/10.1109/TNNLS.2019.2891257
Raissi, M.: Deep hidden physics models: deep learning of nonlinear partial differential equations. J. Mach. Learn. Res. 19, 1–24 (2018)
Schmidt, M., Lipson, H.: Distilling free-form natural laws from experimental data. Science 324(5923), 81–85 (2009). https://doi.org/10.1126/science.1165893
Brunton, S.L., Proctor, J.L., Kutz, J.N.: Discovering governing equations from data by sparse identification of nonlinear dynamical systems. Proc. Natl. Acad. Sci. 113(15), 3932–3937 (2016). https://doi.org/10.1073/pnas.1517384113
Avendaño-Valencia, L.D., Abdallah, B., Chatzi, E.: Virtual fatigue diagnostics of wake-affected wind turbine via Gaussian process regression. Renew. Energy 170, 539–561 (2021). https://doi.org/10.1016/j.renene.2021.02.003
Raissi, M., Perdilaris, P., Karniadakis, G.E.: Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J. Comput. Phys. 378, 686–707 (2019). https://doi.org/10.1016/j.jcp.2018.10.045
Punjani, A., Abbeel, P.: Deep learning helicopter dynamics models. In: Proc. IEEE Int. Conf. Robot. Autom., pp. 3223–3230 (2015). https://doi.org/10.1109/ICRA.2015.7139643
Hashemi, A., Orzechowski, G., Mikkola, A., McPhee, J.: Multibody dynamics and control using machine learning. Multibody Syst. Dyn. 58, 397–431 (2023). https://doi.org/10.1007/s11044-023-09884-x
Zhang, Z., Zhu, Y., Rai, R., Doermann, D.: PIMNet: physics-infuzed neural network for human motion prediction. IEEE Robot. Autom. Lett. 7(4), 8949–8955 (2022). https://doi.org/10.1109/LRA.2022.3188892
Erge, O., Oort, E.: Combining physics-based and data-driven modeling in well construction: hybrid fluid dynamics modeling. J. Nat. Gas Sci. Eng. 97, 104348 (2022). https://doi.org/10.1016/j.jngse.2021.104348
Rahman, M., Rasheed, A., San, O.: A hybrid analytics paradigm combining physics-based modeling and data-driven modeling to accelerate incompressible flow solvers. Fluids 3(3), 50 (2018). https://doi.org/10.3390/fluids3030050
Liu, Q., Liang, J., Ma, O.: A physics-based and data-driven hybrid modeling method for accurately simulating complex contact phenomenon. Multibody Syst. Dyn. 50, 97–117 (2020). https://doi.org/10.1007/s11044-020-09746-w
Askari, E., et al.: Micro-CT based finite element modelling and experimental characterization of the compressive mechanical properties of 3-D zirconia scaffolds for bone tissue engineering. J. Mech. Behav. Biomed. Mater. 102, 103516 (2020). https://doi.org/10.1016/j.jmbbm.2019.103516
Blakseth, S.S., Rasheed, A., Kvamsdal, T., San, O.: Combining physics-based and data-driven techniques for reliable hybrid analysis and modeling using the corrective source term approach. Appl. Soft Comput. 128, 109533 (2022). https://doi.org/10.1016/j.asoc.2022.109533
Linxia, L.L., Köttig, F.: A hybrid framework combining data-driven and model-based methods for system remaining useful life prediction. Appl. Soft Comput. 44, 191–199 (2016). https://doi.org/10.1016/j.asoc.2016.03.013
Zeng, Y., Song, D., Zhang, W., Zhou, B., Xie, M., Tang, X.: A new physics-based data-driven guideline for wear modelling and prediction of train wheels. Wear 456–457, 203355 (2020). https://doi.org/10.1016/j.wear.2020.203355
Askari, E., Andersen, M.S.: Effect of ligament properties on nonlinear dynamics and wear prediction of knee prostheses. J. Biomech. Eng. 143, 021014 (2021). https://doi.org/10.1115/1.4048707
Askari, E., Flores, P., Dabirrahmani, D., Appleyard, R.: Dynamic modeling and analysis of wear in spatial hard-on-hard couple hip replacements using multibody systems methodologies. Nonlinear Dyn. 82(1–2), 1039–1058 (2015). https://doi.org/10.1007/s11071-015-2216-9
Askari, E., Andersen, M.S.: A modification on velocity terms of Reynolds equation in a spherical coordinate system. Tribol. Int. 131, 15–23 (2019). https://doi.org/10.1016/j.triboint.2018.10.019
Flores, P., Lankarani, H.M.: Spatial rigid-multibody systems with lubricated spherical clearance joints: modeling and simulation. Nonlinear Dyn. 60, 99–114 (2010). https://doi.org/10.1007/s11071-009-9583-z
Askari, E.: Mathematical models for characterizing non-Hertzian contacts. Appl. Math. Model. 90, 432–447 (2021). https://doi.org/10.1016/j.apm.2020.08.048
Askari, E., Andersen, M.: A closed-form formulation for the conformal articulation of metal- on-polyethylene hip prostheses: contact mechanics and sliding distance. J. Eng. Med. 232(12), 1196–1208 (2018). https://doi.org/10.1177/0954411918810044
Flores, P., Ambrósio, J., Claro, J.C.P., Lankarani, H.M.: Spatial revolute joints with clearance for dynamic analysis of multibody systems. Proc. Inst. Mech. Eng., Part K, J. Multi-Body Dyn. 220(4), 257–271 (2006). https://doi.org/10.1243/1464419JMBD70
Aggarwal, C.C.: Neural Networks and Deep Learning. Springer, Berlin (2018)
Moshkelgosha, E., Askari, E., Jeong, K.H., Shafiee, A.: Fluid-structure coupling of concentric double FGM shells with different lengths. Struct. Eng. Mech. 61(2), 231–244 (2017). https://doi.org/10.12989/sem.2017.61.2.231
Quaghebeur, W., Nopens, I., De Baets, B.: Incorporating unmodelled dynamics into first-principles models through machine learning. IEEE Access 9, 22014–22022 (2021). https://doi.org/10.1109/ACCESS.2021.3055353
Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G.S., Davis, A., Dean, J., Devin, M., Ghemawat, S., Goodfellow, I., Harp, A., Irving, G., Isard, M., Jia, Y., Jozefowicz, R., Kaiser, L., Kudlur, M., Levenberg, J., Mané, D., Monga, R., Moore, S., Murray, D., Olah, C., Schuster, M., Shlens, J., Steiner, B., Sutskever, I., Talwar, K., Tucker, P., Vanhoucke, V., Vasudevan, V., Viégas, F., Vinyals, O., Warden, P., Wattenberg, M., Wicke, M., Yu, Y., Zheng, X.: (2015). TensorFlow: Large-scale machine learning on heterogeneous systems. Software available from tensorflow.org
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., et al.: Pytorch: an imperative style, high-performance 217 deep learning library. In: Proceedings of the 33rd Conference on Neural Information Processing Systems, pp. 8026–8037 (2019)
Pontryagin, L.S., Mishchenko, E., Boltyanskii, V., Gamkrelidze, R.: The Mathematical Theory of Optimal Processes. Wiley, Hoboken (1962)
Allaire, G.: A review of adjoint methods for sensitivity analysis, uncertainty quantification and optimization in numerical codes. Ing. Automob. 836, 33–36 (2015)
Giles, M., Pierce, N.: An introduction to the adjoint approach to design. Flow Turbul. Combust. 65, 393–415 (2000)
Cao, Y., Li, S., Petzold, L., Serban, R.: Adjoint sensitivity analysis for differential-algebraic equations: the adjoint DAE system and its numerical solution. SIAM J. Sci. Comput. 24, 1076–1089 (2003). https://doi.org/10.1016/S0377-0427(02)00528-9
Sengupta, B., Friston, K.J., Penny, W.D.: Efficient gradient computation for dynamical models. NeuroImage 98, 521–527 (2014). https://doi.org/10.1016/j.neuroimage.2014.04.040
Schramm, D., Hiller, M., Bardini, R.: Vehicle Dynamics: Modeling and Simulation, 2nd edn. Springer, Germany (2014)
https://www.volkswagen-newsroom.com/en/electric-vehicles-3646
Cho, W., Yoon, J., Yim, S., Koo, B., Yi, K.: Estimation of tire forces for application to vehicle stability control. IEEE Trans. Veh. Technol. 59(2), 638–649 (2010). https://doi.org/10.1109/TVT.2009.2034268
Baffet, G., Charara, A., Lechner, D.: Estimation of vehicle sideslip, tire force and wheel cornering stiffness. Control Eng. Pract. 17, 1255–1264 (2009). https://doi.org/10.1016/j.conengprac.2009.05.005
Baffet, G., Charara, A., Dherbomez, G.: An observer of tire–road forces and friction for active security vehicle systems. IEEE/ASME Trans. Mechatron. 12(6), 651–661 (2007). https://doi.org/10.1109/TMECH.2007.910099
Acosta, M., Kanarachos, S., Fitzpatrick, M.: Robust virtual sensing for vehicle agile manoeuvring: a tyre-model-less approach. IEEE Trans. Veh. Technol. 67(3), 1894–1908 (2017). https://doi.org/10.1109/TVT.2017.2767942
Lampe, N., Kortmann, K.P., Westerkamp, C.: Neural network based tire-road friction estimation using experimental data. IFAC-PapersOnLine 56(3), 397–402 (2023). https://doi.org/10.1016/j.ifacol.2023.12.056
Viehweger, M., Vaseur, C., Aalst, S., Acosta, M., Regolin, E., Alatorre, A., Desmet, W., Naets, F., Ivanov, V., Ferrara, A., Victorino, A.: Vehicle state and tyre force estimation: demonstrations and guidelines. Int. J. Veh. Mech. Mobil. 59(5), 675–702 (2021). https://doi.org/10.1080/00423114.2020.1714672
Yang, J., Chen, W., Wang, Y.: Estimate lateral tire force based on yaw moment without using tire model. Mech. Eng. 934181, 1–8 (2014). https://doi.org/10.1155/2014/934181
Hrgetic, M., Deur, J., Ivanovic, V., Tseng, E.: Vehicle sideslip angle EKF estimator based on nonlinear vehicle dynamics model and stochastic tire forces modeling. SAE Int. J. Passeng. Cars - Mech. Syst. 7(1), 86–95 (2014). https://doi.org/10.4271/2014-01-0144
Ray, L.R.: Nonlinear tire force estimation and road friction identification: simulation and experiments. Automatica 33(10), 1819–1833 (1997). https://doi.org/10.1016/S0005-1098(97)00093-9
Albinsson, A., Bruzelius, F., Jonasson, M., Jacobson, B.: Tire force estimation utilizing wheel torque measurements and validation in simulations and experiments. 12th International Symposium on Advanced Vehicle Control September 22-26 (2014)
Ray, L.R.: Nonlinear state and tire force estimation for advanced vehicle control. IEEE Trans. Control Syst. Technol. 3(1), 117–124 (1995)
Pacejka, H.B.: Tyre and Vehicle Dynamics, 2nd edn. Elsevier, Amsterdam (2006)
Wang, Y., Geng, K., Xu, L., Ren, Y., Dong, H., Yin, G.: Estimation of sideslip angle and tire cornering stiffness using fuzzy adaptive robust cubature Kalman filter. IEEE Trans. Syst. Man Cybern. Syst. 52(3), 1451–1462 (2022). https://doi.org/10.1109/TSMC.2020.3020562
Hornik, K.: Multilayer feedforward networks are universal approximators. Neural Netw. 2, 359–366 (1989). https://doi.org/10.1016/0893-6080(89)90020-8
Acknowledgements
The authors gratefully acknowledge the support of VLAIO (Flemish Innovation & Entrepreneurship) through the O&O project AI4Test (HBC.2022.0005).
Funding
This project is financed through the European Recovery and Resilience Facility (RRF).
Author information
Authors and Affiliations
Contributions
E.A. wrote and reviewed the manuscript, developed the method, validated it, wrote scripts, and D.G. wrote the manuscript and provided experiments data, and G.C. reviewed the manuscript and mentored. All authors reviewed the manuscript.
Corresponding author
Ethics declarations
Ethical approval
Not applicable.
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix A
In this appendix, the formulations associated with RNN are given.
The gradient of network output with respect to weights, computing the weight correction at each training iteration, Eq. (6), is given as follows:
in which
The gradient of network output with respect to recurrent weights is, in turn, determined as follows:
where
Appendix B
The differentiation of the hybrid model’s output with respect to the output of the neural network for both hybrid models of S1 and S2 is given in this appendix.
The derivatives associated with the S1 hybrid model are as follows:
and
In addition, those for S2 hybrid model can be written by
and
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Askari, E., Gorgoretti, D. & Crevecoeur, G. Hybrid modeling of multibody vehicles with partially known physics: discovering complex behaviors of tires. Multibody Syst Dyn (2024). https://doi.org/10.1007/s11044-024-09983-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11044-024-09983-3