Abstract
Semi-Lagrangian schemes for the discretization of the dynamic programming principle are based on a time discretization projected on a state-space grid. The use of a structured grid makes this approach not feasible for high-dimensional problems due to the curse of dimensionality. Here, we present a new approach for infinite horizon optimal control problems where the value function is computed using radial basis functions by the Shepard moving least squares approximation method on scattered grids. We propose a new method to generate a scattered mesh driven by the dynamics and the selection of the shape parameter in the RBF using an optimization routine. This mesh will help to localize the problem and approximate the dynamic programming principle in high dimension. Error estimates for the value function are also provided. Numerical tests for high dimensional problems will show the effectiveness of the proposed method.
Similar content being viewed by others
Data Availibility
Enquiries about code availability should be directed to the authors. No data has been used in this paper.
Notes
We remark that, although it is well-known how to compute feedback controls by semi-Lagrangian schemes, this is not trivial for high dimensional problems, as already mentioned earlier in the introduction.
Strictly positive definite functions are such that the associated matrix \(K_X\) is positive definite.
References
Alla, A., D’Elia, M., Glusa, C., Oliveria, A.H.: Control of fractional diffusion problems via dynamic programming equations. J. Peridyn. Nonlocal. Model (2023). https://doi.org/10.1007/s42102-023-00101-z
Alla, A., Falcone, M., Kalise, D.: An efficient policy iteration algorithm for dynamic programming equations. SIAM J. Sci. Comput. 3(7), 181–200 (2015)
Alla, A., Falcone, M., Saluzzi, L.: An efficient DP algorithm on a tree-structure for finite horizon optimal control problems. SIAM J. Sci. Comput. 41, A2384–A2406 (2019)
Alla, A., Falcone, M., Volkwein, S.: Error analysis for POD approximations of infinite horizon problems via the dynamic programming approach. SIAM J. Control Optim. 55, 3091–3115 (2017)
Alla, A., Saluzzi, L.: A HJB-POD approach for the control of nonlinear PDEs on a tree structure. Appl. Numer. Math. 155, 192–207 (2020)
Bardi, M., Capuzzo-Dolcetta, I.: Optimal Control and Viscosity Solutions of Hamilton–Jacobi–Bellman Equations. Birkhäuser, Basel (1997)
Benner, P., Gugercin, S., Willcox, K.: A survey of projection-based model reduction methods for parametric dynamical systems. SIAM Rev. 57, 483–531 (2015)
Bokanowski, O., Garcke, J., Griebel, M., Klompmaker, I.: An adaptive sparse grid semi-Lagrangian scheme for first order Hamilton–Jacobi Bellman equations. J. Sci. Comput. 55, 575–605 (2013)
Capuzzo Dolcetta, I., Ishii, H.: Approximate solutions of the Bellman equation of deterministic control theory. Appl. Math. Optim. 11, 161–181 (1984)
Capuzzo Dolcetta, I.: On a discrete approximation of the Hamilton–Jacobi equation of dynamic programming. Appl. Math. Optim. 10, 367–377 (1983)
Carlini, E., Falcone, M., Ferretti, R.: An efficient algorithm for Hamilton–Jacobi equations in high dimension. Comput. Vis. Sci. 7, 15–29 (2004)
Carlini, E., Ferretti, R.: A semi-Lagrangian scheme with radial basis approximation for surface reconstruction. Comput. Vis. Sci. 18, 103–112 (2017)
Chen, W., Fu, Z.-J., Chen, C.-S.: Recent advances in radial basis function collocation methods. Briefs Appl. Sci. Technol. (2014). https://doi.org/10.1007/978-3-642-39572-7
Chilan, C.M., Conway, B.A.: Optimal nonlinear control using Hamilton–Jacobi–Bellman viscosity solutions on unstructured grids. J. Guid. Control Dyn. 43, 30–38 (2020)
Darbon, J., Osher, S.J.: Splitting enables overcoming the curse of dimensionality. In: Glowinski, R., Osher, S., Yin, W. (eds.) Splitting Methods in Communication, Imaging, Science, and Engineering. Scientific Computation. Springer, Cham (2016)
Darbon, J., Osher, S.: Algorithms for overcoming the curse of dimensionality for certain Hamilton–Jacobi equations arising in control theory and elsewhere. Res. Math. Sci. 3, 19–26 (2016)
Dolgov, S., Kalise, D., Kunisch, K.: Tensor decomposition for high-dimensional Hamilton–Jacobi–Bellman equations, submitted, 2019. https://arxiv.org/pdf/1908.01533.pdf
Dolgov, S., Kalise, D., Saluzzi, L.: Data-driven tensor train gradient cross approximation for Hamilton–Jacobi–Bellman equations, submitted, 2022. arxiv:2205.05109
Ehring, T., Haasdonk, B.: Feedback control for a coupled soft tissue system by kernel surrogates. In: Proceedings of COUPLED 2021 (2021)
Ehring, T., Haasdonk, B.: Greedy sampling and approximation for realizing feedback control for high dimensional nonlinear systems. IFAC-PapersOnLine 55, 325–330 (2022)
Falcone, M., Ferretti, M.R.: Discrete time high-order schemes for viscosity solutions of Hamilton–Jacobi–Bellman equations. Numer. Math. 67, 315–344 (1994)
Falcone, M., Ferretti, R.: Semi-Lagrangian Approximation Schemes for Linear and Hamilton-Jacobi equations. SIAM (2013). https://doi.org/10.1137/1.9781611973051
Fasshauer, G.F.: Meshfree approximation methods with MATLAB (2007)
Fasshauer, G.F., McCourt, M.: Kernel-based approximation methods using MATLAB (2015)
Ferretti, G., Ferretti, R., Junge, O., Schreiber, A.: An adaptive multilevel radial basis function scheme for the HJB equation. IFAC-PapersOnLine 50, 1643–1648 (2017)
Fornberg, B., Flyer, N.: Solving PDEs with radial basis functions. Acta Numer. 24, 215–258 (2015)
Garcke, J., Kröner, A.: Suboptimal feedback control of PDEs by solving HJB equations on adaptive sparse grids. J. Sci. Comput. 70, 1–28 (2017)
Junge, O., Schreiber, A.: Dynamic programming using radial basis functions. Discrete Contin. Dyn. Syst.- Ser. A 35, 4439–4453 (2015)
Grüne, L.: An adaptive grid scheme for the discrete Hamilton–Jacobi–Bellman equation. Numerische Mathematik 75, 319–337 (1997)
Kalise, D., Kunisch, K.: Polynomial approximation of high-dimensional Hamilton–Jacobi–Bellman equations and applications to feedback control of semilinear parabolic PDEs. SIAM J. Sci. Comput. 40, A629–A652 (2018)
Kunisch, K., Volkwein, S., Xie, L.: HJB-POD based feedback design for the optimal control of evolution problems. SIAM J. Appl. Dyn. Syst. 4, 701–722 (2004)
Leveque, R.J.: Finite difference methods for ordinary and partial differential equations: steady-state and time-dependent problems. SIAM (2007). https://doi.org/10.1137/1.9780898717839
McEneaney, W.M.: Convergence rate for a curse-of-dimensionality-free method for Hamilton–Jacobi–Bellman PDEs represented as maxima of quadratic forms. SIAM J. Control Optim. 48, 2651–2685 (2009)
McEneaney, W.M.: A curse-of-dimensionality-free numerical method for solution of certain HJB PDEs. SIAM J. Control Optim. 46, 1239–1276 (2007)
Schmidt, A., Haasdonk, B.: Data-driven surrogates of value functions and applications to feedback control for dynamical systems. IFAC-PapersOnLine 51, 307–312 (2018)
Wendland, H.: Piecewise polynomial, positive definite and compactly supported radial functions of minimal degree. Adv. Comput. Math. 4, 389–396 (1995)
Wendland, H.: Scattered data approximation. Cambridge University Press, Cambridge (2004)
Funding
AA was supported by the CNPq research grant 3008414/2019-1 and by a research grant from PUC-Rio. HO was supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
No conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
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
Alla, A., Oliveira, H. & Santin, G. HJB-RBF Based Approach for the Control of PDEs. J Sci Comput 96, 25 (2023). https://doi.org/10.1007/s10915-023-02208-3
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10915-023-02208-3