Abstract
For control problems with control constraints, a local convergence rate is established for an hp-method based on collocation at the Radau quadrature points in each mesh interval of the discretization. If the continuous problem has a sufficiently smooth solution and the Hamiltonian satisfies a strong convexity condition, then the discrete problem possesses a local minimizer in a neighborhood of the continuous solution, and as either the number of collocation points or the number of mesh intervals increase, the discrete solution convergences to the continuous solution in the sup-norm. The convergence is exponentially fast with respect to the degree of the polynomials on each mesh interval, while the error is bounded by a polynomial in the mesh spacing. An advantage of the hp-scheme over global polynomials is that there is a convergence guarantee when the mesh is sufficiently small, while the convergence result for global polynomials requires that a norm of the linearized dynamics is sufficiently small. Numerical examples explore the convergence theory.
Similar content being viewed by others
Change history
16 May 2019
The original version of this article unfortunately contained an error in two equations on page 25, inside the proof of Proposition 6.1.
References
Babuška, I., Suri, M.: The h-p version of the finite element method with quasiuniform meshes, RAIRO. Modélisation Mathématique et Analyse Numérique 21, 199–238 (1987)
Babuška, I., Suri, M.: The p- and h-p version of the finite element method, an overview. Comput. Methods Appl. Mech. Eng. 80, 5–26 (1990)
Babuška, I., Suri, M.: The p and h-p version of the finite element method, basic principles and properties. SIAM Rev. 36, 578–632 (1994)
Benson, D.A., Huntington, G.T., Thorvaldsen, T.P., Rao, A.V.: Direct trajectory optimization and costate estimation via an orthogonal collocation method. J. Guid. Control Dyn. 29, 1435–1440 (2006)
Bernardi, C., Maday, Y.: Polynomial interpolation results in Sobolev spaces. J. Comput. Appl. Math. 43, 53–82 (1992)
Betts, J.T.: Sparse optimization suite. In: Applied Mathematical Analysis. LLC, Issaquah (2013)
Biegler, L.T., Zavala, V.M.: Large-scale nonlinear programming using IPOPT: an integrating framework for enterprise-wide optimization. Comput. Chem. Eng. 33, 575–582 (2008)
Bonnans, J.F., Laurent-Varin, J.: Computation of order conditions for symplectic partitioned Runge–Kutta schemes with application to optimal control. Numer. Math. 103, 1–10 (2006)
Canuto, C., Hussaini, M., Quarteroni, A., Zang, T.: Spectral Methods, Fundamentals in Single Domains. Springer, Berlin (2006)
Chen, W., Du, W., Hager, W.W., Yang, L.: Bounds for integration matrices that arise in Gauss and Radau collocation. Comput. Optim. Appl. (2019). https://doi.org/10.1007/s10589-019-00099-5
Darby, C.L., Hager, W.W., Rao, A.V.: Direct trajectory optimization using a variable low-order adaptive pseudospectral method. J. Spacecr. Rockets 48, 433–445 (2011)
Darby, C.L., Hager, W.W., Rao, A.V.: An hp-adaptive pseudospectral method for solving optimal control problems. Optim. Control Appl. Methods 32, 476–502 (2011)
Dennis, M.E., Hager, W.W., Rao, A.V.: Computational method for optimal guidance and control using adaptive Gaussian quadrature collocation. J. Guid. Control Dyn. (2019). https://doi.org/10.2514/1.G003943
Dontchev, A., Hager, W.W., Poore, A., Yang, B.: Optimality, stability, and convergence in nonlinear control. Appl. Math. Optim. 31, 297–326 (1995)
Dontchev, A.L., Hager, W.W.: Lipschitzian stability in nonlinear control and optimization. SIAM J. Control Optim. 31, 569–603 (1993)
Dontchev, A.L., Hager, W.W.: A new approach to Lipschitz continuity in state constrained optimal control. Syst. Control Lett. 35, 137–143 (1998)
Dontchev, A.L., Hager, W.W.: The Euler approximation in state constrained optimal control. Math. Comput. 70, 173–203 (2000)
Dontchev, A.L., Hager, W.W., Veliov, V.M.: Second-order Runge–Kutta approximations in constrained optimal control. SIAM J. Numer. Anal. 38, 202–226 (2000)
Elnagar, G., Kazemi, M., Razzaghi, M.: The pseudospectral Legendre method for discretizing optimal control problems. IEEE Trans. Automat. Control 40, 1793–1796 (1995)
Elnagar, G.N., Kazemi, M.A.: Pseudospectral Chebyshev optimal control of constrained nonlinear dynamical systems. Comput. Optim. Appl. 11, 195–217 (1998)
Elschner, J.: The h-p-version of spline approximation methods for Melin convolution equations. J. Integral Equ. Appl. 5, 47–73 (1993)
Fahroo, F., Ross, I.M.: Costate estimation by a Legendre pseudospectral method. J. Guid. Control Dyn. 24, 270–277 (2001)
Fahroo, F., Ross, I.M.: Direct trajectory optimization by a Chebyshev pseudospectral method. J. Guid. Control Dyn. 25, 160–166 (2002)
Garg, D., Patterson, M.A., Darby, C.L., Françolin, C., Huntington, G.T., Hager, W.W., Rao, A.V.: Direct trajectory optimization and costate estimation of finite-horizon and infinite-horizon optimal control problems using a Radau pseudospectral method. Comput. Optim. Appl. 49, 335–358 (2011)
Garg, D., Patterson, M.A., Hager, W.W., Rao, A.V., Benson, D.A., Huntington, G.T.: A unified framework for the numerical solution of optimal control problems using pseudospectral methods. Automatica 46, 1843–1851 (2010)
Gong, Q., Ross, I.M., Kang, W., Fahroo, F.: Connections between the covector mapping theorem and convergence of pseudospectral methods for optimal control. Comput. Optim. Appl. 41, 307–335 (2008)
Gui, W., Babuška, I.: The h, p and h-p versions of the finite element method in 1 dimension. Part I. The error analysis of the p-version. Numer. Math. 49, 577–612 (1986)
Gui, W., Babuška, I.: The h, p and h-p versions of the finite element method in 1 dimension. Part II. The error analysis of the h-and h-p versions. Numer. Math. 49, 613–657 (1986)
Gui, W., Babuška, I.: The h, p and h-p versions of the finite element method in 1 dimension. Part III. The adaptive h-p version. Numer. Math. 49, 659–683 (1986)
Hager, W.W.: Multiplier methods for nonlinear optimal control. SIAM J. Numer. Anal. 27, 1061–1080 (1990)
Hager, W.W.: Runge–Kutta methods in optimal control and the transformed adjoint system. Numer. Math. 87, 247–282 (2000)
Hager, W.W.: Numerical analysis in optimal control. In: Hoffmann, K.-H., Lasiecka, I., Leugering, G., Sprekels, J., Tröltzsch, F. (eds.) International Series of Numerical Mathematics, vol. 139, pp. 83–93. Birkhauser Verlag, Basel (2001)
Hager, W.W., Hou, H., Rao, A.V.: Convergence rate for a Radau collocation method applied to unconstrained optimal control (2015). arXiv:1508.03783
Hager, W.W., Hou, H., Rao, A.V.: Convergence rate for a Gauss collocation method applied to unconstrained optimal control. J. Optim. Theory Appl. 169, 801–824 (2016)
Hager, W.W., Ianculescu, G.: Dual approximations in optimal control. SIAM J. Control Optim. 22, 423–465 (1984)
Hager, W.W., Liu, J., Mohapatra, S., Rao, A.V., Wang, X.-S.: Convergence rate for a Gauss collocation method applied to constrained optimal control. SIAM J. Control Optim. 56, 1386–1411 (2018)
Hager, W.W., Liu, J., Mohapatra, S., Rao, A.V., Wang, X.-S.: A pseudospectral method for optimal control based on collocation at the gauss points. In: 2018 IEEE Conference on Decision and Control (CDC), pp. 2490–2495 (Dec 2018)
Horn, R.A., Johnson, C.R.: Matrix Analysis. Cambridge University Press, Cambridge (2013)
Kang, W.: The rate of convergence for a pseudospectral optimal control method. In: Proceeding of the 47th IEEE Conference on Decision and Control, pp. 521–527. IEEE (2008)
Kang, W.: Rate of convergence for the Legendre pseudospectral optimal control of feedback linearizable systems. J. Control Theory Appl. 8, 391–405 (2010)
Liu, F., Hager, W.W., Rao, A.V.: Adaptive mesh refinement method for optimal control using nonsmoothness detection and mesh size reduction. J. Frankl. Inst. 352, 4081–4106 (2015)
Nocedal, J., Wright, S.J.: Numerical Optimization, 2nd edn. Springer, New York (2006)
Patterson, M.A., Hager, W.W., Rao, A.V.: A \(ph\) mesh refinement method for optimal control. Optim. Control Appl. Meth. 36, 398–421 (2015)
Patterson, M.A., Rao, A.V.: GPOPS-II: A MATLAB software for solving multi-phase optimal control problems using hp-adaptive Gaussian quadrature collocation methods and sparse non-linear programming. ACM Trans. Math. Softw. 41, 1–37 (2014)
Patterson, M.A., Rao, A.V.: \({\mathbb{GPOPS-II}}\), a MATLAB software for solving multiple-phase optimal control problems using \(hp\)-adaptive Gaussian quadrature collocation methods and sparse nonlinear programming. ACM Trans. Math. Softw. 41, 1–37 (2015)
Shen, J., Tang, T., Wang, L.-L.: Spectral Methods. Springer, Berlin (2011)
Williams, P.: Jacobi pseudospectral method for solving optimal control problems. J. Guid. Control Dyn. 27, 293–297 (2004)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The original version of this article was revised: In page 25, under Proposition 6.1, the typo in two equations has been corrected.
October 24, 2017, revised March 25, 2019. The authors gratefully acknowledge support by the Office of Naval Research under Grants N00014-11-1-0068, N00014-15-1-2048 and N00014-18-1-2100, by the National Science Foundation under Grants DMS-1522629, CBET-1404767 and DMS-1819002, and by the U.S. Air Force Research Laboratory under contract FA8651-08-D-0108/0054.
Appendix: Proof of (P1) and (P2)
Appendix: Proof of (P1) and (P2)
We analyze (P1) and (P2) when \(\tau _i\), \(1 \le i \le N\), are either the Radau quadrature points analyzed in this paper, or the Gauss quadrature points studied in [36].
Lemma 10.1
For either the Gauss or Radau quadrature points, the rows of the matrix \([\mathbf{{W}}^{1/2}\mathbf{{D}}_{1:N}]^{-1}\) have Euclidean length bounded by \(\sqrt{2}\). For the Gauss quadrature points, \(\Vert \mathbf{{D}}_{1:N}^{-1}\Vert _\infty \le 2\), and \(\Vert \mathbf{{D}}_{1:N}^{-1}\Vert _\infty \) approaches 2 as N tends to infinity, while for the Radau quadrature points, \(\Vert \mathbf{{D}}_{1:N}^{-1}\Vert _\infty = 2\).
Proof
Given \(\mathbf{{p}} \in \mathbb {R}^N\), let \(p \in \mathcal{{P}}_N\) denote the polynomial that satisfies \(p(-1) = 0\) and \(p(\tau _i) = p_i\), \(1 \le i \le N\). Let \(\dot{\mathbf{{p}}} \in \mathbb {R}^N\) denote the vector with components \(\dot{p}_i = \dot{p}(\tau _i)\), and let \(\ell _j\) be the Lagrange polynomial defined by
The identity
holds since \(\dot{p} \in \mathcal{{P}}_{N-1}\) and the polynomials on each side of (10.1) are equal at the N quadrature points. Integrate (10.1) to obtain
Since \(\mathbf{{D}}\) is a differentiation matrix and \(p(-1) = 0\), it follows that \(\mathbf{{D}}_{1:N}\mathbf{{p}} = \dot{\mathbf{{p}}}\). If the vector \(\dot{\mathbf{{p}}} = \mathbf{{0}}\), then the polynomial \(\dot{p} = 0\) since \(\dot{p}\) has degree \(N-1\) and vanishes at N points. Since \(p(-1) = 0\), it follows that polynomial \(p = 0\), which implies that the vector \(\mathbf{{p}} = \mathbf{{0}}\). Hence, \(\mathbf{{D}}_{1:N}\) is invertible, and \(\mathbf{{p}} = \mathbf{{D}}^{-1}\dot{\mathbf{{p}}}\). Comparing the equality \(\mathbf{{p}} = \mathbf{{D}}^{-1}\dot{\mathbf{{p}}}\) to (10.2), we deduce that
Choose any \(s \in [-1, 1]\) and define
Observe that \((\mathbf{{D}}^{-1})_{ij} = d_j(\tau _i)\) and \(R(\tau _i)\) is the square of the Euclidean length of row i in \((\mathbf{{W}}^{1/2}\mathbf{{D}})^{-1}\). Let \(q \in \mathcal{{P}}_{N-1}\) be the polynomial defined by
Hence, by the triangle and Schwarz inequalities,
Since \(q^2 \in \mathcal{{P}}_{2N-2}\), both Radau and Gauss quadrature are exact, and
where the \(\tau _j\) are either the Radau or Gauss quadrature points and the \(\omega _j\) are the associated weights. Since \(\ell _j(\tau _i) = 1\) for \(i = j\) and \(\ell _j(\tau _i) = 0\) otherwise, it follows from the definition of q that \(q(\tau _i) = d_i(s)/\omega _i\). This substitution in (10.5) yields
Equating the expressions (10.4) and (10.6) implies that
By (10.6), \(R(s) \le 2\) for any \(s \in [-1, 1]\). In particular, \(R(\tau _i) \le 2\) for \(1 \le i \le N\). Since \(R(\tau _i)\) is the square of the Euclidean length of row i in \((\mathbf{{W}}^{1/2}\mathbf{{D}})^{-1}\), the rows of \((\mathbf{{W}}^{1/2}\mathbf{{D}})^{-1}\) have Euclidean length bounded by \(\sqrt{2}\). This result holds for both the Radau and Gauss quadrature points since since \(q^2 \in \mathcal{{P}}_{2N-2}\), and both Radau and Gauss quadrature are exact for polynomials of this degree.
If \(\mathbf{{r}}\) is a row of \(\mathbf{{D}}_{1:N}^{-1}\), then by the Schwarz inequality and the fact that the quadrature weights sum to 2 and the rows of the matrix \([\mathbf{{W}}^{1/2}\mathbf{{D}}_{1:N}]^{-1}\) have Euclidean length bounded by \(\sqrt{2}\), we have
Consequently, the absolute row sums for \(\mathbf{{D}}_{1:N}^{-1}\) are all bounded by 2, or equivalently, \(\Vert \mathbf{{D}}_{1:N}^{-1}\Vert _\infty \le 2\). Given any polynomial \(p \in \mathcal{{P}}_N\) with \(p(-1) = 0\) and \(|\dot{p}(\tau _i)| \le 1\) for \(1 \le i \le N\), it is observed in Section 9 of [36] that \(\Vert \mathbf{{D}}_{1:N}^{-1}\Vert _\infty \ge \)\(\max \{p(\tau _i) : 1 \le i \le N\}\). Take \(p(\tau ) = 1 + \tau \) to deduce that \(\Vert \mathbf{{D}}_{1:N}^{-1}\Vert _\infty \ge 1 + \tau _N\). Hence, \(1 + \tau _N \le \Vert \mathbf{{D}}_{1:N}^{-1}\Vert _\infty \le 2\). Since \(\tau _N = 1\) for the Radau points, it follows that \(\Vert \mathbf{{D}}_{1:N}^{-1}\Vert _\infty = 2\). For the Gauss points, \(\tau _N\) approaches 1 as N tends to infinity; consequently, \(\Vert \mathbf{{D}}_{1:N}^{-1}\Vert _\infty \) approaches 2 as N tends to infinity for the Gauss points. \(\square \)
Rights and permissions
About this article
Cite this article
Hager, W.W., Hou, H., Mohapatra, S. et al. Convergence rate for a Radau hp collocation method applied to constrained optimal control. Comput Optim Appl 74, 275–314 (2019). https://doi.org/10.1007/s10589-019-00100-1
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10589-019-00100-1