Numerical Resolution of optimal control problems by a Piecewise Linear continuation method

Size: px
Start display at page:

Download "Numerical Resolution of optimal control problems by a Piecewise Linear continuation method"

Transcription

1 Numerical Resolution of optimal control problems by a Piecewise Linear continuation method Pierre Martinon november 25

2

3 Shooting method and homotopy

4 Introduction We would like to solve numerically the following problems: Low thrust orbital transfer problem (bang-bang control, huge number of revolutions) Problems with singular arcs Direct methods (state / control discretization) Non linear optimisation problem Not a good choice for a large number of commutations a priori. Indirect methods (Necessary conditions) Based on Pontriaguine s Maximum Principle. Fast and accurate in favorable cases.

5 Shooting method Optimal control problem Non linear equation system Starting Problem (P) Boundary Value Problem (BVP) Initial Value Problem (IVP) Shooting function S Solve S(z) = Find a solution of (P)

6 6 Difficulties of the studied problems Orbital transfer: S is not smooth The shooting function is not differentiable (even not defined) everywhere. We need a good starting point for the simple shooting convergence. Singular arcs: S is set valued Multiple shooting: requires some knowledge of the control structure (number of singular arcs typically). How to solve these difficulties? We use a homotopic method to obtain the required information (initial point and control structure), with no a priori knowledge about the solution of the problem.

7 Homotopy and zero path Parametrize the problem (P) with λ [, ]: Family of problems (P λ ) such that - (P ) is simple enough to solve. - (P ) is the original problem (P). We define the homotopy H by H : (z, λ) S λ (z) Homotopic method (continuation) - Start from a known zero z of H(, ) - Follow the zero path of H until we reach λ = - Then we have a zero z of H(, ) = S = S.

8 8 Convergence results (cf J.Gergaud) Homotopy convergence We denote y = (x, p) R n the state-costate pair. H : [a, b] R n U R continuous and convex with respect to u. Let Γ(t, x, p) be the set of the solutions of min u U H(t, x, p, u). Under the correct assumptions (in particular Γ has compact convex nonempty values and is usc), from every sequence of solutions (y λk ) of (BVP) λk (λ k ), we can extract a subsequence (y k ) satisfying: (i) (y k ) converges uniformly to y solution of (BVP). (ii) (ẏ k ) converges weakly-* to ẏ in L n ([, t f ]). et (iii) (u k ) converges weakly-* in L n ([, t f ]).

9 Simplicial algorithm

10 Principle: piecewise linear approximation of the path We consider a homotopy H : R n+ R n. Simplex and face A simplex is the convex hull of n + 2 affinely independant points of R n+. A k-face is the convex hull of k vertices of a simplex (face for k = n + ). Triangulation A triangulation is a countable family T of simplices of R n+ satisfying: The intersection of two simplices of T is either a k-face or empty. T is locally finite. Illustration of triangulations K and J 3

11 Vertices labeling We define the labeling l by l(v i ) = H(z i, λ i ), with v i = (z i, λ i ). We define H T by affine interpolation over the vertices of the simplices of T. Completely labeled face A face [v,.., v n+ ] is completely labeled iff it contains a zero of H T, this being true under a certain perturbation (precisely, it contains a solution v ɛ of the equation H T (v) = ɛ for all sufficiently small ɛ >, with ɛ = (ɛ,.., ɛ n )). Fundamental property Every simplex has either zero, or exactly two completely labeled faces.

12 2 General simplicial algorithm Given starting simplex (completely labeled face at λ = ). Repeat until we reach λ = - Determine the second compl. labeled face of the current simplex. - Build the unique simplex sharing this face (pivot rules) new current simplex. End repeat x* x followed zero path zero of homotopy PL approximation completely labeled face transverse simplex Schematic illustration in dimension 2

13 3 Properties The followed path does not cycle It is possible to follow a path with non-maximal rank Possible adaptation to the set-valued case Convergence for a set-valued H (cf Allgower-Georg) We consider a simplicial algorithm using a selection of H as labeling and a refining triangulation of R n [, [. Under the following assumptions: (i) the generated faces remain in K [, ] (K compact). (ii) the algorithm does not go back to λ =. Then if H is usc and convex compact valued, the algorithm generates a sequence (z i, λ i ) such that λ i, and there exists a subsequence that converges to (z, ) such that H(z, ).

14 4 Examples of path following for several triangulations.8 Lambda = X = Simplx: 3.8 Lambda = X = Simplx: 33.8 Lambda = X = Simplx: 32 Lambda.6.4 Lambda.6.4 Lambda Z 2 () Z () Z 2 () Z () Z 2 () Z () Uniform triangulations K, J, D.8 Lambda = X = Simplx:.8 Lambda = X = Simplx: 53 Lambda.6.4 Lambda Z 2 ().55 Z () Z 2 () Z () Refining triangulations J 3 et J 4

15 Junction homotopy Objective: We would like to be able to find a completely labeled face for a given triangulation meshsize, at a certain level λ j. Application: Path following initialisation (first face at λ = ) Change the triangulation during the following Refine the solution at λ = (sequence of triangulations with decreasing meshsize) Principle: Intermediate homotopy at level λ j : we try to connect a well chosen affine application to H(, λ j ), using the wanted meshisze. The choice of the affine application is such that the junction homotopy is easy to initialise.

16 Adaptive triangulation Objective: improve the speed / precision of the path following Refining triangulations? Difficult to use in practice (meshsize is fixed regardless of the path followed) Idea: dynamically adapt the triangulation to the followed path At certain levels λ = λ i during the following: - we determine the new meshsize δ. - we perform a junction homotopy to find a completely labeled face of meshsize δ at the λ i level. - the following continues with the new meshsize until the next level. 2 mechanisms to choose the new meshsize: Deviation control: adapt the meshsize according to the accuracy of the following. Anisotropic deformation: try to take into account the direction of the path.

17 Deviation control Norm of H at the zeros of H T : hint of the accuracy of the following. Depending on the quality of this hint, we uniformly increase or decrease the meshsize Expectations: Illustration on a simple example Speed up the following, without degrading the accuracy too much.

18 8 Anisotropic deformation We estimate the relative weight of each dimension in the path followed since the previous level. We increase (decrease) the meshize along the dimensions in the majority (minority). δ = δ 2 δ 2 2δ 2 δ δ 2 (δ, δ 2 ) ( δ 2, 2δ 2 ) Expectations: Better accuracy, can balance the side effects of the deviation control part.

19 Solution refining Objective: improve the solution at λ =. First idea: refining triangulation before λ = to finish the following Lambda Lambda = X = Simplx: 3.3 Flaw: we encounter the difficulties related to the use of refining triangulations Z 2 () Z () Second idea: perform a sequence of junctions at λ = with triangulations of decreasing meshsize (cf Merril s restart algorithm) Generally effective, but the cost of the junctions is often high.

20 Conclusion The use of junction homotopies allows an easy initialisation of the simplicial algorithm, and performs the meshsize changes for the adaptive triangulation. Concerning the adaptive triangulation, we tried to preserve the robustness of the simplicial algorithm (no derivatives, same numerical settings for all problems)

21 Orbital transfer 2

22 Problem statement Orbital transfer problem (CNES) Initial orbit: ellipse, slight inclination (7 degrees) Final orbit: geostationnary, geosynchronous Criterion: final mass maximisation (payload) We consider low thrust propulsors Difficulties: Mass criterion bang-bang control Low thrust huge number of revolutions

23 23 Problem formulation Forces: central force + propulsor thrust { r = r µ + u r 3 m ṁ = T Max I spg u µ = Gm T Earth gravitationnal constant. T Max the maximal thrust ( u ). I sp the specific impulsion of the propulsor. For the practical resolution (cf T.Haberkorn): - we minimize the fuel consumption. - non cartesian coordinates, form orbital parameters. - integration along longitude instead of time. - final time and longitude are fixed.

24 24 Criterion homotopy Criterion energy to mass We start from a quadratic criterion (minimization of an energy ) to the consumption criterion: J λ = Min tf t λ u(t) + ( λ) u(t) 2 dt. Homotopy: corresponding shooting function for λ [, ] H : (z, λ) S λ (z) λ < : continuous control (Hamiltonian is strictly convex) λ = : bang-bang control (the original problem)

25 The following converges to λ = without any major difficulties. The number of simplices does not seem related to the thrust T Max. The shooting results are consistent with the ones obtained by 25 Path following and results Initialisation at λ = : We perform a shoot with a fixed step integration (Runge Kutta 4). This solution is sufficient for the first junction of the simplicial algorithm. Results T max Initial norm Simplices Time Objective Final norm N N N N N N

26 t f L Transfer for T Max = Newtons Solution at λ = e y P h x m e x h y STATE x x COSTATE CONTROL Evolution of the control along the path CONTROL NORM u λ= λ=.5 λ= TIME

27 RKF45 Integrators comparison: Trajectories for,,.n DOP ODEX

28 Integration stepsize with respect to the thrust T Max For each integrator, we have: Integration steps T Max C te RKF45, DOP835 and ODEX Integration steps vs T Max 5 RKF45 DOP853 ODEX STEPS 4 3 T max (N) Note: logarithmic scale on both axes Confirms the regularity of the problem with respect to T max (we already had t f T Max C te, L f T Max C te )

29 29 Trajectory and stepsize - RKF45 RKF45: embedded Runge Kutta formulas (4,5) RKF45: SWITCHING FUNCTION and INTEGRATION STEPSIZE SWITCH STEPSIZE RKF45: Integrator Stepsize h TIME Time Regular spreading of the dots, 3 values scopes for the stepsize Larger stepsize over the arcs without thrust (only L changes) Extremely small stepsize at the commutations ( 9 )

30 3 Trajectory and stepsize - DOP853 DOP853: embedded Runge Kutta formulas (8,5-3) DOP853: SWITCHING FUNCTION and INTEGRATION STEPSIZE.2 SWITCH STEPSIZE DOP853: Integrator Stepsize h TIME Time Similarity with RKF45 concerning the spreading of the dots Globally, larger stepsize than RKF45 (higher order) Still very small stepsize at the commutations

31 3 Trajectory and stepsize - ODEX ODEX: Gragg-Bulirsch-Stoer extrapolation (variable order) 4 2 ODEX: SWITCH, STEPSIZE AND ORDER SWITCH STEPSIZE ORDER ODEX: Integrator Stepsize 2 Order h TIME Time Less regular spreading of the dots Some very large steps (extrapolation, high order) Again, very small stepsize at the commutations Order is generally high ( 8), decreases at the commutations

32 N UNIFORM AND ADAPTIVE TRIANGULATIONS (K. /.),5N UNIFORM AND ADAPTIVE TRIANGULATIONS (K. /.) 5N UNIFORM AND ADAPTIVE TRIANGULATIONS (K.),2N UNIFORM AND ADAPTIVE TRIANGULATIONS (K. /.) N UNIFORM AND ADAPTIVE TRIANGULATIONS (K. /.) Adaptative triangulation Number of simplices followed along the path (uniform / adaptive) for T Max =, 5,,.5,.2 et.n 25 UNIFORM ADAPTIVE 2 8 UNIFORM ADAPTIVE 8 6 UNIFORM ADAPTIVE SPLX 5 SPLX 2 8 SPLX LAMBDA LAMBDA LAMBDA 35 3 UNIFORM ADAPTIVE 8 6 UNIFORM ADAPTIVE 5,N UNIFORM AND ADAPTIVE TRIANGULATIONS (K. /.) UNIFORM ADAPTIVE SPLX 2 5 SPLX 8 SPLX LAMBDA LAMBDA LAMBDA Less simplices followed (and smoother following) Better final accuracy (faster shooting) BUT: the cost of the junctions can sometimes offset this gain! 32

33 33 Conclusion Resolution of the probleme up to T Max =.N, results are consistent with the ones obtained with a predictor-corrector method (for an execution time 3 to 5 times longer). The 3 integrators tested manage to perform the integration, but the commutations (more than 5 at.n) require extremely small steps ( 9 ). Adaptative mode is qualitatively satisfying (less simplices, more accurate following) but sometimes suffers from the high numerical cost of the junctions.

34 Singular arcs

35 35 Introduction Singular arc: Interval (non reduced to a point) over which the minimisation of H does not determine u uniquely. Differential inclusion: ϕ the dynamic of the state-costate pair y = (x, p). Γ the expression of the optimal control, Γ is set valued and we have { ẏ Φ(y) = ϕ(y, Γ(y)) ae over [t, t (BVP) f ] Boundary conditions Resolution with multiple shooting: Typically requires some knowledge of the singular structure (number, approximate location and nature of the arcs).

36 Problem : optimal fishing (P ) Min ( c x(t) E) u(t) U max dt ẋ(t) = rx(t) ( x(t) u(t) t [, ] x() = 7. 6 x() free k ) u(t) U max E =, c = , r =.7, k = , U max = 2. 6 Switching function and optimal control: ψ(t) = c x(t) E p(t), u (t) = if ψ(t) > u (t) = if ψ(t) < u (t) [, ] if ψ(t) =. Over a singular arc: via ψ = ( ) using = k r c 2 ( c x p) Umax x c k p + 2px k 2px2 k 2

37 Problem 2: quadratic regulator (P 2 ) 5 Min 2 (x 2(t) + x 2 2 (t)) dt x (t) = x 2 (t) x 2 (t) = u(t) u(t) t [, 5] x() = (, ) x(5) free. Switching function and optimal control: { u ψ(t) = p 2 (t), (t) = sign p 2 (t) if ψ(t) u (t) [, ] if ψ(t) =. Over a singular arc: via ψ = u sing (t) = x (t)

38 Homotopy on the criterion: quadratic perturbation Add a quadratic perturbation to the criterion (λ [, ]) We regularize these problems by adding a u 2 term to the criterion c (note: x E < ). (J λ ) Min (J λ 2 ) Min 2 ( c x E ) (u ( λ) u 2 ) U max dt 5 (x 2 + x 2 2 ) + ( λ)u 2 dt λ < : the Hamiltonian is strictly convex, u is a continuous function. λ = : we have the two original problems.

39 Path following Initialisation at λ = without any difficulties, the following begins well. Evolution of the control and switching function? CONTROL EVOLUTION SWITCHING FUNCTION EVOLUTION U.7 ψ TIME TIME (P ): u and ψ for λ =,.5,.75,.9 and.95 We can predict the appearance of a singular arc when λ : - the control u remains non extremal - ψ gets closer to

40 But the following degrades brutally when λ comes closer to Numerical instability when ψ is close to..8 CROSSING TURNING BACK SWITCHING FUNCTION.9 CONTROL EVOLUTION ψ.2 U.6.5 CROSSING TURNING BACK TIME TIME λ = - Two control structures - (P ) Depending on the exit sign of ψ: 2 different bang-bang controls. When λ, 2 structures at the vertices of the simplices. The algorithm converges to λ =, but loses the singular structure. At λ = : Discontinuity of S near the singular arc.

41 Discretized BVP formulation We discretize the equations of the BVP: Euler or Trapezoid formulas ( homotopy is convex valued). (x i, p i ) unknowns, u i obtained by usual NC. Matching conditions at t i : continuity of x, p { xi (x Match i + h x t (t i, x i, p i, ui )) cond p i (p i + h p t (t i, x i, p i, ui )) Note: the discretisation is limited by the problem size. Results Converges without difficulties to λ =. Good approximation of x, p. Detection of the singular structure with ψ. Matching: residual errors related to the u i over the singular arcs.

42 42 Precise resolution Multiple shooting adapted to the singular case Uses the algebraic expression of the singular control u sing. The boundaries of the singular arcs are part of the unknowns. Initialisation with the information provided by the homotopies. 7 x 7 X P U X P.5 U Psi SWITCHING FUNCTION X P2 Psi SWITCHING FUNCTION Accurate solution, quasi instant convergence. We clearly find the suspected singular arcs. Requires a good starting point (correct structure).

43 Quality of the approximate solutions? Comparison of the formulations - Simple shooting (λ =.95 and.925, before the structure loss) - Discretized BVP (5 and 2 points, Trapezoid with refinement) - Reference solution x 7 X P Psi.5..2 CONTROL 5 SWITCHING FUNCTION ψ.3 STRUCTURED SINGLE (λ=.95).4 DISCRETIZED TIME STATE X STATE X COSTATE P COSTATE P2 2 Psi.5.5 CONTROL SWITCHING FUNCTION ψ STRUCTURED SINGLE (λ=.925) DISCRETIZED TIME The discretized solution is quite close (except for singular control...)

44 Discretized control formulation Inspired by direct methods: - the discretized control is part of the unknowns, x, p are integrated. - conditions over the u i : u i Γ(t i, x i, p i ) (needs formalising...). 7 x 7 X P.5 U State x Costate p Control u Psi ψ State x Costate p2 Psi Switching function ψ Discretized control (RK4-5 points) - (P ) and (P 2 ) Good approximation of x, p, ψ. Approximation of the singular control without the algebraic expression. 44

45 Comparison with a direct method Discretization of x and u Optimal control problem non linear optimisation problem Resolution with KNITRO (barrier sub-problems, SQP). 7 x 7 State Control State Control X time U time X X time.5 U.5 5 time time Direct method (RK4 - points) - (P ) and (P 2 ) We find again the singular structure. The solution has some noise over the singular arcs (commutations). 45

46 46 Conclusion The two homotopies (simple shooting and discretized BVP) converge. Simple shooting: guesses the singular structure but loses it at λ =. Discretized BVP formulation: good approximation of x, p and ψ Precise resolution by a multiple shooting method. Discretized control formulation: needs formalising, but interesting first numerical results. Direct method: consistent results for a similar execution time.

47 Conclusions and Prospects

48 48 General conclusions For the problems we studied (orbital transfer and singular arcs), the homotopy allowed us to determine the control structure and obtain a satisfying initial point, with no a priori knowledge. Thanks to this information, we can then solve the problems by the shooting methods. The simplicial algorithm and the shooting methods have been implemented in a single program (Simplicial), used for all numerical experiments (except direct methods). External codes: HYBRD, RKF45 (Shampine/Watts), DOP853 (Hairer/Wanner), ODEX (Hairer/Wanner)

49 49 Prospects Simplicial algorithm: - improve the cost of junctions for adaptative mode and solution refining - introduce speed-up mechanisms (cf Newton / secant) Work on the shooting method itself: - adapt the integration (detection of commutations) - better computation of the derivatives (IND) Further development of the discretized formulations: - more evolved discretisation schemes (symplectic? cf Bonnans/Laurent-Varin) - better formalisation of the discretized control formulation More thorough comparison with direct methods State constraints (formulation, homotopy?)

ODEs occur quite often in physics and astrophysics: Wave Equation in 1-D stellar structure equations hydrostatic equation in atmospheres orbits

ODEs occur quite often in physics and astrophysics: Wave Equation in 1-D stellar structure equations hydrostatic equation in atmospheres orbits Solving ODEs General Stuff ODEs occur quite often in physics and astrophysics: Wave Equation in 1-D stellar structure equations hydrostatic equation in atmospheres orbits need workhorse solvers to deal

More information

Introduction to Optimization Problems and Methods

Introduction to Optimization Problems and Methods Introduction to Optimization Problems and Methods wjch@umich.edu December 10, 2009 Outline 1 Linear Optimization Problem Simplex Method 2 3 Cutting Plane Method 4 Discrete Dynamic Programming Problem Simplex

More information

Value function and optimal trajectories for a control problem with supremum cost function and state constraints

Value function and optimal trajectories for a control problem with supremum cost function and state constraints Value function and optimal trajectories for a control problem with supremum cost function and state constraints Hasnaa Zidani ENSTA ParisTech, University of Paris-Saclay joint work with: A. Assellaou,

More information

Parameterization of triangular meshes

Parameterization of triangular meshes Parameterization of triangular meshes Michael S. Floater November 10, 2009 Triangular meshes are often used to represent surfaces, at least initially, one reason being that meshes are relatively easy to

More information

Spline Methods Draft. Tom Lyche and Knut Mørken

Spline Methods Draft. Tom Lyche and Knut Mørken Spline Methods Draft Tom Lyche and Knut Mørken 24th May 2002 2 Contents 1 Splines and B-splines an introduction 3 1.1 Convex combinations and convex hulls..................... 3 1.1.1 Stable computations...........................

More information

Parameterization. Michael S. Floater. November 10, 2011

Parameterization. Michael S. Floater. November 10, 2011 Parameterization Michael S. Floater November 10, 2011 Triangular meshes are often used to represent surfaces, at least initially, one reason being that meshes are relatively easy to generate from point

More information

Lecture 0: Reivew of some basic material

Lecture 0: Reivew of some basic material Lecture 0: Reivew of some basic material September 12, 2018 1 Background material on the homotopy category We begin with the topological category TOP, whose objects are topological spaces and whose morphisms

More information

Spline Methods Draft. Tom Lyche and Knut Mørken

Spline Methods Draft. Tom Lyche and Knut Mørken Spline Methods Draft Tom Lyche and Knut Mørken January 5, 2005 2 Contents 1 Splines and B-splines an Introduction 3 1.1 Convex combinations and convex hulls.................... 3 1.1.1 Stable computations...........................

More information

Point-Set Topology II

Point-Set Topology II Point-Set Topology II Charles Staats September 14, 2010 1 More on Quotients Universal Property of Quotients. Let X be a topological space with equivalence relation. Suppose that f : X Y is continuous and

More information

Shape Modeling and Geometry Processing

Shape Modeling and Geometry Processing 252-0538-00L, Spring 2018 Shape Modeling and Geometry Processing Discrete Differential Geometry Differential Geometry Motivation Formalize geometric properties of shapes Roi Poranne # 2 Differential Geometry

More information

Interior point algorithm for the optimization of a space shuttle re-entry trajectory

Interior point algorithm for the optimization of a space shuttle re-entry trajectory Interior point algorithm for the optimization of a space shuttle re-entry trajectory. Julien Laurent-Varin, CNES-INRIA-ONERA Common work with Julien Laurent-Varin J. F. Bonnans INRIA, N. Bérend ONERA,

More information

Spline Methods Draft. Tom Lyche and Knut Mørken. Department of Informatics Centre of Mathematics for Applications University of Oslo

Spline Methods Draft. Tom Lyche and Knut Mørken. Department of Informatics Centre of Mathematics for Applications University of Oslo Spline Methods Draft Tom Lyche and Knut Mørken Department of Informatics Centre of Mathematics for Applications University of Oslo January 27, 2006 Contents 1 Splines and B-splines an Introduction 1 1.1

More information

Lower bounds on the barrier parameter of convex cones

Lower bounds on the barrier parameter of convex cones of convex cones Université Grenoble 1 / CNRS June 20, 2012 / High Performance Optimization 2012, Delft Outline Logarithmically homogeneous barriers 1 Logarithmically homogeneous barriers Conic optimization

More information

weighted minimal surface model for surface reconstruction from scattered points, curves, and/or pieces of surfaces.

weighted minimal surface model for surface reconstruction from scattered points, curves, and/or pieces of surfaces. weighted minimal surface model for surface reconstruction from scattered points, curves, and/or pieces of surfaces. joint work with (S. Osher, R. Fedkiw and M. Kang) Desired properties for surface reconstruction:

More information

Finite difference methods

Finite difference methods Finite difference methods Siltanen/Railo/Kaarnioja Spring 8 Applications of matrix computations Applications of matrix computations Finite difference methods Spring 8 / Introduction Finite difference methods

More information

Finite Element Methods

Finite Element Methods Chapter 5 Finite Element Methods 5.1 Finite Element Spaces Remark 5.1 Mesh cells, faces, edges, vertices. A mesh cell is a compact polyhedron in R d, d {2,3}, whose interior is not empty. The boundary

More information

AM205: lecture 2. 1 These have been shifted to MD 323 for the rest of the semester.

AM205: lecture 2. 1 These have been shifted to MD 323 for the rest of the semester. AM205: lecture 2 Luna and Gary will hold a Python tutorial on Wednesday in 60 Oxford Street, Room 330 Assignment 1 will be posted this week Chris will hold office hours on Thursday (1:30pm 3:30pm, Pierce

More information

Lecture notes for Topology MMA100

Lecture notes for Topology MMA100 Lecture notes for Topology MMA100 J A S, S-11 1 Simplicial Complexes 1.1 Affine independence A collection of points v 0, v 1,..., v n in some Euclidean space R N are affinely independent if the (affine

More information

Adaptive-Mesh-Refinement Pattern

Adaptive-Mesh-Refinement Pattern Adaptive-Mesh-Refinement Pattern I. Problem Data-parallelism is exposed on a geometric mesh structure (either irregular or regular), where each point iteratively communicates with nearby neighboring points

More information

DIRECT SEQUENTIAL DYNAMIC OPTIMIZATION WITH AUTOMATIC SWITCHING STRUCTURE DETECTION. Martin Schlegel, Wolfgang Marquardt 1

DIRECT SEQUENTIAL DYNAMIC OPTIMIZATION WITH AUTOMATIC SWITCHING STRUCTURE DETECTION. Martin Schlegel, Wolfgang Marquardt 1 DIRECT SEQUENTIAL DYNAMIC OPTIMIZATION WITH AUTOMATIC SWITCHING STRUCTURE DETECTION Martin Schlegel, Wolfgang Marquardt 1 Lehrstuhl für Prozesstechnik, RWTH Aachen University D 52056 Aachen, Germany Abstract:

More information

Ordinary Differential Equations

Ordinary Differential Equations Next: Partial Differential Equations Up: Numerical Analysis for Chemical Previous: Numerical Differentiation and Integration Subsections Runge-Kutta Methods Euler's Method Improvement of Euler's Method

More information

Manifolds. Chapter X. 44. Locally Euclidean Spaces

Manifolds. Chapter X. 44. Locally Euclidean Spaces Chapter X Manifolds 44. Locally Euclidean Spaces 44 1. Definition of Locally Euclidean Space Let n be a non-negative integer. A topological space X is called a locally Euclidean space of dimension n if

More information

Simplicial Hyperbolic Surfaces

Simplicial Hyperbolic Surfaces Simplicial Hyperbolic Surfaces Talk by Ken Bromberg August 21, 2007 1-Lipschitz Surfaces- In this lecture we will discuss geometrically meaningful ways of mapping a surface S into a hyperbolic manifold

More information

Cell-Like Maps (Lecture 5)

Cell-Like Maps (Lecture 5) Cell-Like Maps (Lecture 5) September 15, 2014 In the last two lectures, we discussed the notion of a simple homotopy equivalences between finite CW complexes. A priori, the question of whether or not a

More information

However, this is not always true! For example, this fails if both A and B are closed and unbounded (find an example).

However, this is not always true! For example, this fails if both A and B are closed and unbounded (find an example). 98 CHAPTER 3. PROPERTIES OF CONVEX SETS: A GLIMPSE 3.2 Separation Theorems It seems intuitively rather obvious that if A and B are two nonempty disjoint convex sets in A 2, then there is a line, H, separating

More information

Lecture 2 Unstructured Mesh Generation

Lecture 2 Unstructured Mesh Generation Lecture 2 Unstructured Mesh Generation MIT 16.930 Advanced Topics in Numerical Methods for Partial Differential Equations Per-Olof Persson (persson@mit.edu) February 13, 2006 1 Mesh Generation Given a

More information

Geometric structures on 2-orbifolds

Geometric structures on 2-orbifolds Geometric structures on 2-orbifolds Section 1: Manifolds and differentiable structures S. Choi Department of Mathematical Science KAIST, Daejeon, South Korea 2010 Fall, Lectures at KAIST S. Choi (KAIST)

More information

over The idea is to construct an algorithm to solve the IVP ODE (9.1)

over The idea is to construct an algorithm to solve the IVP ODE (9.1) Runge- Ku(a Methods Review of Heun s Method (Deriva:on from Integra:on) The idea is to construct an algorithm to solve the IVP ODE (9.1) over To obtain the solution point we can use the fundamental theorem

More information

Fully discrete Finite Element Approximations of Semilinear Parabolic Equations in a Nonconvex Polygon

Fully discrete Finite Element Approximations of Semilinear Parabolic Equations in a Nonconvex Polygon Fully discrete Finite Element Approximations of Semilinear Parabolic Equations in a Nonconvex Polygon Tamal Pramanick 1,a) 1 Department of Mathematics, Indian Institute of Technology Guwahati, Guwahati

More information

arxiv: v1 [math.na] 20 Sep 2016

arxiv: v1 [math.na] 20 Sep 2016 arxiv:1609.06236v1 [math.na] 20 Sep 2016 A Local Mesh Modification Strategy for Interface Problems with Application to Shape and Topology Optimization P. Gangl 1,2 and U. Langer 3 1 Doctoral Program Comp.

More information

Contents. I Basics 1. Copyright by SIAM. Unauthorized reproduction of this article is prohibited.

Contents. I Basics 1. Copyright by SIAM. Unauthorized reproduction of this article is prohibited. page v Preface xiii I Basics 1 1 Optimization Models 3 1.1 Introduction... 3 1.2 Optimization: An Informal Introduction... 4 1.3 Linear Equations... 7 1.4 Linear Optimization... 10 Exercises... 12 1.5

More information

over The idea is to construct an algorithm to solve the IVP ODE (8.1)

over The idea is to construct an algorithm to solve the IVP ODE (8.1) Runge- Ku(a Methods Review of Heun s Method (Deriva:on from Integra:on) The idea is to construct an algorithm to solve the IVP ODE (8.1) over To obtain the solution point we can use the fundamental theorem

More information

Optimal Adaptive Feedback Control of a Network Buffer.

Optimal Adaptive Feedback Control of a Network Buffer. 25 American Control Conference June 8-1, 25. Portland, OR, USA ThA3. Optimal Adaptive Feedback Control of a Network Buffer. V. Guffens, G. Bastin Centre for Systems Engineering and Applied Mechanics (CESAME)

More information

SIMPLICIAL ENERGY AND SIMPLICIAL HARMONIC MAPS

SIMPLICIAL ENERGY AND SIMPLICIAL HARMONIC MAPS SIMPLICIAL ENERGY AND SIMPLICIAL HARMONIC MAPS JOEL HASS AND PETER SCOTT Abstract. We introduce a combinatorial energy for maps of triangulated surfaces with simplicial metrics and analyze the existence

More information

Lecture 2 - Introduction to Polytopes

Lecture 2 - Introduction to Polytopes Lecture 2 - Introduction to Polytopes Optimization and Approximation - ENS M1 Nicolas Bousquet 1 Reminder of Linear Algebra definitions Let x 1,..., x m be points in R n and λ 1,..., λ m be real numbers.

More information

CS205b/CME306. Lecture 9

CS205b/CME306. Lecture 9 CS205b/CME306 Lecture 9 1 Convection Supplementary Reading: Osher and Fedkiw, Sections 3.3 and 3.5; Leveque, Sections 6.7, 8.3, 10.2, 10.4. For a reference on Newton polynomial interpolation via divided

More information

Programming, numerics and optimization

Programming, numerics and optimization Programming, numerics and optimization Lecture C-4: Constrained optimization Łukasz Jankowski ljank@ippt.pan.pl Institute of Fundamental Technological Research Room 4.32, Phone +22.8261281 ext. 428 June

More information

EARLY INTERIOR-POINT METHODS

EARLY INTERIOR-POINT METHODS C H A P T E R 3 EARLY INTERIOR-POINT METHODS An interior-point algorithm is one that improves a feasible interior solution point of the linear program by steps through the interior, rather than one that

More information

Inverse and Implicit functions

Inverse and Implicit functions CHAPTER 3 Inverse and Implicit functions. Inverse Functions and Coordinate Changes Let U R d be a domain. Theorem. (Inverse function theorem). If ϕ : U R d is differentiable at a and Dϕ a is invertible,

More information

SPH: Towards the simulation of wave-body interactions in extreme seas

SPH: Towards the simulation of wave-body interactions in extreme seas SPH: Towards the simulation of wave-body interactions in extreme seas Guillaume Oger, Mathieu Doring, Bertrand Alessandrini, and Pierre Ferrant Fluid Mechanics Laboratory (CNRS UMR6598) Ecole Centrale

More information

New formulations of the semi-lagrangian method for Vlasov-type equations

New formulations of the semi-lagrangian method for Vlasov-type equations New formulations of the semi-lagrangian method for Vlasov-type equations Eric Sonnendrücker IRMA Université Louis Pasteur, Strasbourg projet CALVI INRIA Nancy Grand Est 17 September 2008 In collaboration

More information

The Borsuk-Ulam theorem- A Combinatorial Proof

The Borsuk-Ulam theorem- A Combinatorial Proof The Borsuk-Ulam theorem- A Combinatorial Proof Shreejit Bandyopadhyay April 14, 2015 1 Introduction The Borsuk-Ulam theorem is perhaps among the results in algebraic topology having the greatest number

More information

Mars Pinpoint Landing Trajectory Optimization Using Sequential Multiresolution Technique

Mars Pinpoint Landing Trajectory Optimization Using Sequential Multiresolution Technique Mars Pinpoint Landing Trajectory Optimization Using Sequential Multiresolution Technique * Jisong Zhao 1), Shuang Li 2) and Zhigang Wu 3) 1), 2) College of Astronautics, NUAA, Nanjing 210016, PRC 3) School

More information

Advanced Operations Research Techniques IE316. Quiz 1 Review. Dr. Ted Ralphs

Advanced Operations Research Techniques IE316. Quiz 1 Review. Dr. Ted Ralphs Advanced Operations Research Techniques IE316 Quiz 1 Review Dr. Ted Ralphs IE316 Quiz 1 Review 1 Reading for The Quiz Material covered in detail in lecture. 1.1, 1.4, 2.1-2.6, 3.1-3.3, 3.5 Background material

More information

Simple shooting-projection method for numerical solution of two-point Boundary Value Problems

Simple shooting-projection method for numerical solution of two-point Boundary Value Problems Simple shooting-projection method for numerical solution of two-point Boundary Value Problems Stefan M. Filipo Ivan D. Gospodino a Department of Programming and Computer System Application, University

More information

Numerical Optimization

Numerical Optimization Convex Sets Computer Science and Automation Indian Institute of Science Bangalore 560 012, India. NPTEL Course on Let x 1, x 2 R n, x 1 x 2. Line and line segment Line passing through x 1 and x 2 : {y

More information

OBSTACLE AVOIDANCE FOR MOBILE ROBOTS USING SWITCHING SURFACE OPTIMIZATION. Mauro Boccadoro Magnus Egerstedt Yorai Wardi

OBSTACLE AVOIDANCE FOR MOBILE ROBOTS USING SWITCHING SURFACE OPTIMIZATION. Mauro Boccadoro Magnus Egerstedt Yorai Wardi OBSTACLE AVOIDANCE FOR MOBILE ROBOTS USING SWITCHING SURFACE OPTIMIZATION Mauro Boccadoro Magnus Egerstedt Yorai Wardi boccadoro@diei.unipg.it Dipartimento di Ingegneria Elettronica e dell Informazione

More information

The 3D DSC in Fluid Simulation

The 3D DSC in Fluid Simulation The 3D DSC in Fluid Simulation Marek K. Misztal Informatics and Mathematical Modelling, Technical University of Denmark mkm@imm.dtu.dk DSC 2011 Workshop Kgs. Lyngby, 26th August 2011 Governing Equations

More information

Chapter 6. Semi-Lagrangian Methods

Chapter 6. Semi-Lagrangian Methods Chapter 6. Semi-Lagrangian Methods References: Durran Chapter 6. Review article by Staniford and Cote (1991) MWR, 119, 2206-2223. 6.1. Introduction Semi-Lagrangian (S-L for short) methods, also called

More information

Mathematics in Orbit

Mathematics in Orbit Mathematics in Orbit Dan Kalman American University Slides and refs at www.dankalman.net Outline Basics: 3D geospacial models Keyhole Problem: Related Rates! GPS: space-time triangulation Sensor Diagnosis:

More information

Math 225 Scientific Computing II Outline of Lectures

Math 225 Scientific Computing II Outline of Lectures Math 225 Scientific Computing II Outline of Lectures Spring Semester 2003 I. Interpolating polynomials Lagrange formulation of interpolating polynomial Uniqueness of interpolating polynomial of degree

More information

CS 450 Numerical Analysis. Chapter 7: Interpolation

CS 450 Numerical Analysis. Chapter 7: Interpolation Lecture slides based on the textbook Scientific Computing: An Introductory Survey by Michael T. Heath, copyright c 2018 by the Society for Industrial and Applied Mathematics. http://www.siam.org/books/cl80

More information

16.410/413 Principles of Autonomy and Decision Making

16.410/413 Principles of Autonomy and Decision Making 16.410/413 Principles of Autonomy and Decision Making Lecture 17: The Simplex Method Emilio Frazzoli Aeronautics and Astronautics Massachusetts Institute of Technology November 10, 2010 Frazzoli (MIT)

More information

1.2 Numerical Solutions of Flow Problems

1.2 Numerical Solutions of Flow Problems 1.2 Numerical Solutions of Flow Problems DIFFERENTIAL EQUATIONS OF MOTION FOR A SIMPLIFIED FLOW PROBLEM Continuity equation for incompressible flow: 0 Momentum (Navier-Stokes) equations for a Newtonian

More information

Panagiotis Tsiotras. Dynamics and Control Systems Laboratory Daniel Guggenheim School of Aerospace Engineering Georgia Institute of Technology

Panagiotis Tsiotras. Dynamics and Control Systems Laboratory Daniel Guggenheim School of Aerospace Engineering Georgia Institute of Technology Panagiotis Tsiotras Dynamics and Control Systems Laboratory Daniel Guggenheim School of Aerospace Engineering Georgia Institute of Technology ICRAT 12 Tutotial on Methods for Optimal Trajectory Design

More information

The Level Set Method. Lecture Notes, MIT J / 2.097J / 6.339J Numerical Methods for Partial Differential Equations

The Level Set Method. Lecture Notes, MIT J / 2.097J / 6.339J Numerical Methods for Partial Differential Equations The Level Set Method Lecture Notes, MIT 16.920J / 2.097J / 6.339J Numerical Methods for Partial Differential Equations Per-Olof Persson persson@mit.edu March 7, 2005 1 Evolving Curves and Surfaces Evolving

More information

An Image Curvature Microscope

An Image Curvature Microscope An Jean-Michel MOREL Joint work with Adina CIOMAGA and Pascal MONASSE Centre de Mathématiques et de Leurs Applications, Ecole Normale Supérieure de Cachan Séminaire Jean Serra - 70 ans April 2, 2010 Jean-Michel

More information

Chapter 3. Bootstrap. 3.1 Introduction. 3.2 The general idea

Chapter 3. Bootstrap. 3.1 Introduction. 3.2 The general idea Chapter 3 Bootstrap 3.1 Introduction The estimation of parameters in probability distributions is a basic problem in statistics that one tends to encounter already during the very first course on the subject.

More information

COMPLETE METRIC ABSOLUTE NEIGHBORHOOD RETRACTS

COMPLETE METRIC ABSOLUTE NEIGHBORHOOD RETRACTS COMPLETE METRIC ABSOLUTE NEIGHBORHOOD RETRACTS WIES LAW KUBIŚ Abstract. We characterize complete metric absolute (neighborhood) retracts in terms of existence of certain maps of CW-polytopes. Using our

More information

21. Efficient and fast numerical methods to compute fluid flows in the geophysical β plane

21. Efficient and fast numerical methods to compute fluid flows in the geophysical β plane 12th International Conference on Domain Decomposition Methods Editors: Tony Chan, Takashi Kako, Hideo Kawarada, Olivier Pironneau, c 2001 DDM.org 21. Efficient and fast numerical methods to compute fluid

More information

William P. Thurston. The Geometry and Topology of Three-Manifolds

William P. Thurston. The Geometry and Topology of Three-Manifolds William P. Thurston The Geometry and Topology of Three-Manifolds Electronic version 1.1 - March 2002 http://www.msri.org/publications/books/gt3m/ This is an electronic edition of the 1980 notes distributed

More information

Lecture 18: Groupoids and spaces

Lecture 18: Groupoids and spaces Lecture 18: Groupoids and spaces The simplest algebraic invariant of a topological space T is the set π 0 T of path components. The next simplest invariant, which encodes more of the topology, is the fundamental

More information

Section 5 Convex Optimisation 1. W. Dai (IC) EE4.66 Data Proc. Convex Optimisation page 5-1

Section 5 Convex Optimisation 1. W. Dai (IC) EE4.66 Data Proc. Convex Optimisation page 5-1 Section 5 Convex Optimisation 1 W. Dai (IC) EE4.66 Data Proc. Convex Optimisation 1 2018 page 5-1 Convex Combination Denition 5.1 A convex combination is a linear combination of points where all coecients

More information

Aspects of Convex, Nonconvex, and Geometric Optimization (Lecture 1) Suvrit Sra Massachusetts Institute of Technology

Aspects of Convex, Nonconvex, and Geometric Optimization (Lecture 1) Suvrit Sra Massachusetts Institute of Technology Aspects of Convex, Nonconvex, and Geometric Optimization (Lecture 1) Suvrit Sra Massachusetts Institute of Technology Hausdorff Institute for Mathematics (HIM) Trimester: Mathematics of Signal Processing

More information

Applied Lagrange Duality for Constrained Optimization

Applied Lagrange Duality for Constrained Optimization Applied Lagrange Duality for Constrained Optimization Robert M. Freund February 10, 2004 c 2004 Massachusetts Institute of Technology. 1 1 Overview The Practical Importance of Duality Review of Convexity

More information

Exact discrete Morse functions on surfaces. To the memory of Professor Mircea-Eugen Craioveanu ( )

Exact discrete Morse functions on surfaces. To the memory of Professor Mircea-Eugen Craioveanu ( ) Stud. Univ. Babeş-Bolyai Math. 58(2013), No. 4, 469 476 Exact discrete Morse functions on surfaces Vasile Revnic To the memory of Professor Mircea-Eugen Craioveanu (1942-2012) Abstract. In this paper,

More information

Outline Introduction Problem Formulation Proposed Solution Applications Conclusion. Compressed Sensing. David L Donoho Presented by: Nitesh Shroff

Outline Introduction Problem Formulation Proposed Solution Applications Conclusion. Compressed Sensing. David L Donoho Presented by: Nitesh Shroff Compressed Sensing David L Donoho Presented by: Nitesh Shroff University of Maryland Outline 1 Introduction Compressed Sensing 2 Problem Formulation Sparse Signal Problem Statement 3 Proposed Solution

More information

Computation of the Constrained Infinite Time Linear Quadratic Optimal Control Problem

Computation of the Constrained Infinite Time Linear Quadratic Optimal Control Problem Computation of the Constrained Infinite Time Linear Quadratic Optimal Control Problem July 5, Introduction Abstract Problem Statement and Properties In this paper we will consider discrete-time linear

More information

60 2 Convex sets. {x a T x b} {x ã T x b}

60 2 Convex sets. {x a T x b} {x ã T x b} 60 2 Convex sets Exercises Definition of convexity 21 Let C R n be a convex set, with x 1,, x k C, and let θ 1,, θ k R satisfy θ i 0, θ 1 + + θ k = 1 Show that θ 1x 1 + + θ k x k C (The definition of convexity

More information

Mutation-linear algebra and universal geometric cluster algebras

Mutation-linear algebra and universal geometric cluster algebras Mutation-linear algebra and universal geometric cluster algebras Nathan Reading NC State University Mutation-linear ( µ-linear ) algebra Universal geometric cluster algebras The mutation fan Universal

More information

Basics of Combinatorial Topology

Basics of Combinatorial Topology Chapter 7 Basics of Combinatorial Topology 7.1 Simplicial and Polyhedral Complexes In order to study and manipulate complex shapes it is convenient to discretize these shapes and to view them as the union

More information

Orientation of manifolds - definition*

Orientation of manifolds - definition* Bulletin of the Manifold Atlas - definition (2013) Orientation of manifolds - definition* MATTHIAS KRECK 1. Zero dimensional manifolds For zero dimensional manifolds an orientation is a map from the manifold

More information

Mathematics for chemical engineers

Mathematics for chemical engineers Mathematics for chemical engineers Drahoslava Janovská Department of mathematics Winter semester 2013-2014 Numerical solution of ordinary differential equations Initial value problem Outline 1 Introduction

More information

By taking products of coordinate charts, we obtain charts for the Cartesian product of manifolds. Hence the Cartesian product is a manifold.

By taking products of coordinate charts, we obtain charts for the Cartesian product of manifolds. Hence the Cartesian product is a manifold. 1 Manifolds A manifold is a space which looks like R n at small scales (i.e. locally ), but which may be very different from this at large scales (i.e. globally ). In other words, manifolds are made by

More information

Horizontal Flight Dynamics Simulations using a Simplified Airplane Model and Considering Wind Perturbation

Horizontal Flight Dynamics Simulations using a Simplified Airplane Model and Considering Wind Perturbation Horizontal Flight Dynamics Simulations using a Simplified Airplane Model and Considering Wind Perturbation Dan N. DUMITRIU*,1,2, Andrei CRAIFALEANU 2, Ion STROE 2 *Corresponding author *,1 SIMULTEC INGINERIE

More information

Robots are built to accomplish complex and difficult tasks that require highly non-linear motions.

Robots are built to accomplish complex and difficult tasks that require highly non-linear motions. Path and Trajectory specification Robots are built to accomplish complex and difficult tasks that require highly non-linear motions. Specifying the desired motion to achieve a specified goal is often a

More information

Integer Programming Theory

Integer Programming Theory Integer Programming Theory Laura Galli October 24, 2016 In the following we assume all functions are linear, hence we often drop the term linear. In discrete optimization, we seek to find a solution x

More information

Surfaces: notes on Geometry & Topology

Surfaces: notes on Geometry & Topology Surfaces: notes on Geometry & Topology 1 Surfaces A 2-dimensional region of 3D space A portion of space having length and breadth but no thickness 2 Defining Surfaces Analytically... Parametric surfaces

More information

Using Subspace Constraints to Improve Feature Tracking Presented by Bryan Poling. Based on work by Bryan Poling, Gilad Lerman, and Arthur Szlam

Using Subspace Constraints to Improve Feature Tracking Presented by Bryan Poling. Based on work by Bryan Poling, Gilad Lerman, and Arthur Szlam Presented by Based on work by, Gilad Lerman, and Arthur Szlam What is Tracking? Broad Definition Tracking, or Object tracking, is a general term for following some thing through multiple frames of a video

More information

CS 6210 Fall 2016 Bei Wang. Review Lecture What have we learnt in Scientific Computing?

CS 6210 Fall 2016 Bei Wang. Review Lecture What have we learnt in Scientific Computing? CS 6210 Fall 2016 Bei Wang Review Lecture What have we learnt in Scientific Computing? Let s recall the scientific computing pipeline observed phenomenon mathematical model discretization solution algorithm

More information

INTRODUCTION TO FINITE ELEMENT METHODS

INTRODUCTION TO FINITE ELEMENT METHODS INTRODUCTION TO FINITE ELEMENT METHODS LONG CHEN Finite element methods are based on the variational formulation of partial differential equations which only need to compute the gradient of a function.

More information

Lecture 2 September 3

Lecture 2 September 3 EE 381V: Large Scale Optimization Fall 2012 Lecture 2 September 3 Lecturer: Caramanis & Sanghavi Scribe: Hongbo Si, Qiaoyang Ye 2.1 Overview of the last Lecture The focus of the last lecture was to give

More information

1. Introduction. performance of numerical methods. complexity bounds. structural convex optimization. course goals and topics

1. Introduction. performance of numerical methods. complexity bounds. structural convex optimization. course goals and topics 1. Introduction EE 546, Univ of Washington, Spring 2016 performance of numerical methods complexity bounds structural convex optimization course goals and topics 1 1 Some course info Welcome to EE 546!

More information

Convexity: an introduction

Convexity: an introduction Convexity: an introduction Geir Dahl CMA, Dept. of Mathematics and Dept. of Informatics University of Oslo 1 / 74 1. Introduction 1. Introduction what is convexity where does it arise main concepts and

More information

An introduction to Topological Data Analysis through persistent homology: Intro and geometric inference

An introduction to Topological Data Analysis through persistent homology: Intro and geometric inference Sophia-Antipolis, January 2016 Winter School An introduction to Topological Data Analysis through persistent homology: Intro and geometric inference Frédéric Chazal INRIA Saclay - Ile-de-France frederic.chazal@inria.fr

More information

TOPOLOGY, DR. BLOCK, FALL 2015, NOTES, PART 3.

TOPOLOGY, DR. BLOCK, FALL 2015, NOTES, PART 3. TOPOLOGY, DR. BLOCK, FALL 2015, NOTES, PART 3. 301. Definition. Let m be a positive integer, and let X be a set. An m-tuple of elements of X is a function x : {1,..., m} X. We sometimes use x i instead

More information

Understanding Concepts of Optimization and Optimal Control with WORHP Lab

Understanding Concepts of Optimization and Optimal Control with WORHP Lab Understanding Concepts of Optimization and Optimal Control with WORHP Lab M. Knauer, C. Büskens Zentrum für Universität Bremen 6th International Conference on Astrodynamics Tools and Techniques 14 th -

More information

Joint Feature Distributions for Image Correspondence. Joint Feature Distribution Matching. Motivation

Joint Feature Distributions for Image Correspondence. Joint Feature Distribution Matching. Motivation Joint Feature Distributions for Image Correspondence We need a correspondence model based on probability, not just geometry! Bill Triggs MOVI, CNRS-INRIA, Grenoble, France http://www.inrialpes.fr/movi/people/triggs

More information

Optimal boundary control of a tracking problem for a parabolic distributed system with open-loop control using evolutionary algorithms

Optimal boundary control of a tracking problem for a parabolic distributed system with open-loop control using evolutionary algorithms Optimal boundary control of a tracking problem for a parabolic distributed system with open-loop control using evolutionary algorithms Russel J STONIER Faculty of Informatics and Communication, Central

More information

Simplicial Complexes: Second Lecture

Simplicial Complexes: Second Lecture Simplicial Complexes: Second Lecture 4 Nov, 2010 1 Overview Today we have two main goals: Prove that every continuous map between triangulable spaces can be approximated by a simplicial map. To do this,

More information

Parameter Estimation in Differential Equations: A Numerical Study of Shooting Methods

Parameter Estimation in Differential Equations: A Numerical Study of Shooting Methods Parameter Estimation in Differential Equations: A Numerical Study of Shooting Methods Franz Hamilton Faculty Advisor: Dr Timothy Sauer January 5, 2011 Abstract Differential equation modeling is central

More information

Unstructured Mesh Generation for Implicit Moving Geometries and Level Set Applications

Unstructured Mesh Generation for Implicit Moving Geometries and Level Set Applications Unstructured Mesh Generation for Implicit Moving Geometries and Level Set Applications Per-Olof Persson (persson@mit.edu) Department of Mathematics Massachusetts Institute of Technology http://www.mit.edu/

More information

05 - Surfaces. Acknowledgements: Olga Sorkine-Hornung. CSCI-GA Geometric Modeling - Daniele Panozzo

05 - Surfaces. Acknowledgements: Olga Sorkine-Hornung. CSCI-GA Geometric Modeling - Daniele Panozzo 05 - Surfaces Acknowledgements: Olga Sorkine-Hornung Reminder Curves Turning Number Theorem Continuous world Discrete world k: Curvature is scale dependent is scale-independent Discrete Curvature Integrated

More information

Simplicial Global Optimization

Simplicial Global Optimization Simplicial Global Optimization Julius Žilinskas Vilnius University, Lithuania September, 7 http://web.vu.lt/mii/j.zilinskas Global optimization Find f = min x A f (x) and x A, f (x ) = f, where A R n.

More information

1 Euler characteristics

1 Euler characteristics Tutorials: MA342: Tutorial Problems 2014-15 Tuesday, 1-2pm, Venue = AC214 Wednesday, 2-3pm, Venue = AC201 Tutor: Adib Makroon 1 Euler characteristics 1. Draw a graph on a sphere S 2 PROBLEMS in such a

More information

4. Definition: topological space, open set, topology, trivial topology, discrete topology.

4. Definition: topological space, open set, topology, trivial topology, discrete topology. Topology Summary Note to the reader. If a statement is marked with [Not proved in the lecture], then the statement was stated but not proved in the lecture. Of course, you don t need to know the proof.

More information

Lecture 12: Feasible direction methods

Lecture 12: Feasible direction methods Lecture 12 Lecture 12: Feasible direction methods Kin Cheong Sou December 2, 2013 TMA947 Lecture 12 Lecture 12: Feasible direction methods 1 / 1 Feasible-direction methods, I Intro Consider the problem

More information

Optimal control and applications in aerospace

Optimal control and applications in aerospace Optimal control and applications in aerospace Collaborations: Emmanuel Trélat (LJLL), Max Cerf (Airbus) Laboratoire Jacques-Louis Lions Université Pierre et Marie Curie zhu@ann.jussieu.fr - Why aerospace

More information

Lagrangian methods and Smoothed Particle Hydrodynamics (SPH) Computation in Astrophysics Seminar (Spring 2006) L. J. Dursi

Lagrangian methods and Smoothed Particle Hydrodynamics (SPH) Computation in Astrophysics Seminar (Spring 2006) L. J. Dursi Lagrangian methods and Smoothed Particle Hydrodynamics (SPH) Eulerian Grid Methods The methods covered so far in this course use an Eulerian grid: Prescribed coordinates In `lab frame' Fluid elements flow

More information

The Cyclic Cycle Complex of a Surface

The Cyclic Cycle Complex of a Surface The Cyclic Cycle Complex of a Surface Allen Hatcher A recent paper [BBM] by Bestvina, Bux, and Margalit contains a construction of a cell complex that gives a combinatorial model for the collection of

More information