Learning Exactly Linearizable Deep Dynamics Models (2311.18261v2)
Abstract: Research on control using models based on machine-learning methods has now shifted to the practical engineering stage. Achieving high performance and theoretically guaranteeing the safety of the system is critical for such applications. In this paper, we propose a learning method for exactly linearizable dynamical models that can easily apply various control theories to ensure stability, reliability, etc., and to provide a high degree of freedom of expression. As an example, we present a design that combines simple linear control and control barrier functions. The proposed model is employed for the real-time control of an automotive engine, and the results demonstrate good predictive performance and stable control under constraints.
- Control Barrier Functions: Theory and Applications. 2019 18th European Control Conference (ECC), pages 3420–3431, jun 2019. doi: 10.23919/ECC.2019.8796030.
- Integral control barrier functions for dynamically defined control laws. IEEE Control Systems Letters, 5(3):887–892, 2021. doi: 10.1109/LCSYS.2020.3006764.
- Input convex neural networks. 34th International Conference on Machine Learning, ICML 2017, 1:192–206, 2017.
- One-step neural network inversion with PDF learning and emulation. In Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005., volume 2, pages 966–971. IEEE, 2005. doi: 10.1109/IJCNN.2005.1555983.
- A nonlinear model predictive control system based on wiener piecewise linear models. Journal of Process Control, 13(7):655–666, 2003. doi: https://doi.org/10.1016/S0959-1524(02)00121-X.
- Optimal control via neural networks: A convex approach. 7th International Conference on Learning Representations, ICLR 2019, pages 1–25, 2019.
- Flatness and defect of non-linear systems: Introductory theory and examples. International Journal of Control, 61(6):1327–1361, 1995. ISSN 13665820. doi: 10.1080/00207179508921959.
- Nolinear model predictive control using hammerstein models. Journal of Process Control, 7(1):31–41, 1997. ISSN 0959-1524. doi: https://doi.org/10.1016/S0959-1524(97)80001-B.
- Sebastien Gros. Implicit non-convex model predictive control. In Handbook of Model Predictive Control, pages 305–333. Springer, 2019.
- A. Hammerstein. Nichtlineare integralgleichung nebst anwendungen. Acta Mathematica, 54:117–176, 1930.
- Ramdane Hedjar. Adaptive neural network model predictive control. International Journal of Innovative Computing, Information and Control, 9:1245–1257, 01 2013.
- Hassan K. Khalil. Nonlinear Systems. Prentice Hall, 3rd ed edition, 2002.
- Gaussian process model based predictive control. In 2004 American Control Conference, page ThA08.3, 2004.
- Maciej Ławryńczuk. Suboptimal Nonlinear Predictive Control Based on Neural Wiener Models. In Artificial Intelligence: Methodology, Systems, and Applications, pages 410–414. Springer Berlin Heidelberg, Berlin, Heidelberg, 2008. doi: 10.1007/978-3-540-85776-1_40.
- DeepMPC: Learning deep latent features for model predictive control. Robotics: Science and Systems, 11, 2015. doi: 10.15607/RSS.2015.XI.012.
- Diesel engine air path control based on neural approximation of nonlinear MPC. Control Engineering Practice, 91(April):104114, 2019. doi: 10.1016/j.conengprac.2019.104114.
- Structured Hammerstein-Wiener Model Learning for Model Predictive Control. IEEE Control Systems Letters, 6:397–402, 2022. ISSN 2475-1456. doi: 10.1109/LCSYS.2021.3077201.
- Truong X. Nghiem. Linearized Gaussian processes for fast data-driven model predictive control. In 2019 American Control Conference, pages 1629–1634, 2019.
- Eduardo Sontag. Mathematical Control Theory: Deterministic Finite-Dimensional Systems. 01 1998. doi: 10.1007/978-1-4612-0577-7.
- Renjeng Su. On the linear equivalents of nonlinear systems. Systems & Control Letters, 2(1):48–52, 1982. ISSN 0167-6911. doi: https://doi.org/10.1016/S0167-6911(82)80042-X.
- Norbert Wiener. Nonlinear problems in random theory. Wiley, 1958.
Sponsor
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.