Combining Direct and Indirect Methods in Optimal Control: Range Maximization of a Hang Glider View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

1993

AUTHORS

Roland Bulirsch , Edda Nerz , Hans Josef Pesch , Oskar von Stryk

ABSTRACT

When solving optimal control problems, indirect methods such as multiple shooting suffer from difficulties in finding an appropriate initial guess for the adjoint variables. For, this initial estimate must be provided for the iterative solution of the multipoint boundary-value problems arising from the necessary conditions of optimal control theory. Direct methods such as direct collocation do not suffer from this problem, but they generally yield results of lower accuracy and their iteration may even terminate with a non-optimal solution. Therefore, both methods are combined in such a way that the direct collocation method is at first applied to a simplified optimal control problem where all inequality constraints are neglected as long as the resulting problem is still well-defined. Because of the larger domain of convergence of the direct method, an approximation of the optimal solution of this problem can be obtained easier. The fusion between direct and indirect methods is then based on a relationship between the Lagrange multipliers of the underlying nonlinear programming problem to be solved by the direct method and the adjoint variables appearing in the necessary conditions which form the boundary-value problem to be solved by the indirect method. Hence, the adjoint variables, too, can be estimated from the approximation obtained by the direct method. This first step then facilitates the subsequent extension and competition of the model by homotopy techniques and the solution of the arising boundary-value problems by the indirect multiple shooting method. Proceeding in this way, the high accuracy and reliability of the multiple shooting method, especially the precise computation of the switching structure and the possibility to verify many necessary conditions, is preserved while disadvantages caused by the sensitive dependence on an appropriate estimate of the solution are considerably cut down. This procedure is described in detail for the numerical solution of the maximum-range trajectory optimization problem of a hang glider in an upwind which provides an example for a control problem where appropriate initial estimates for the adjoint variables are hard to find. More... »

PAGES

273-288

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-0348-7539-4_20

DOI

http://dx.doi.org/10.1007/978-3-0348-7539-4_20

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1004585554


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/01", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Mathematical Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0102", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Applied Mathematics", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0103", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Numerical and Computational Mathematics", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Mathematisches Institut, Technische Universit\u00e4t M\u00fcnchen, Postfach 20 24 20, D-8000, M\u00fcnchen 2, Germany", 
          "id": "http://www.grid.ac/institutes/grid.6936.a", 
          "name": [
            "Mathematisches Institut, Technische Universit\u00e4t M\u00fcnchen, Postfach 20 24 20, D-8000, M\u00fcnchen 2, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Bulirsch", 
        "givenName": "Roland", 
        "id": "sg:person.0767272315.31", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0767272315.31"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Mathematisches Institut, Technische Universit\u00e4t M\u00fcnchen, Postfach 20 24 20, D-8000, M\u00fcnchen 2, Germany", 
          "id": "http://www.grid.ac/institutes/grid.6936.a", 
          "name": [
            "Mathematisches Institut, Technische Universit\u00e4t M\u00fcnchen, Postfach 20 24 20, D-8000, M\u00fcnchen 2, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Nerz", 
        "givenName": "Edda", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Mathematisches Institut, Technische Universit\u00e4t M\u00fcnchen, Postfach 20 24 20, D-8000, M\u00fcnchen 2, Germany", 
          "id": "http://www.grid.ac/institutes/grid.6936.a", 
          "name": [
            "Mathematisches Institut, Technische Universit\u00e4t M\u00fcnchen, Postfach 20 24 20, D-8000, M\u00fcnchen 2, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Pesch", 
        "givenName": "Hans Josef", 
        "id": "sg:person.014543723551.22", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014543723551.22"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Mathematisches Institut, Technische Universit\u00e4t M\u00fcnchen, Postfach 20 24 20, D-8000, M\u00fcnchen 2, Germany", 
          "id": "http://www.grid.ac/institutes/grid.6936.a", 
          "name": [
            "Mathematisches Institut, Technische Universit\u00e4t M\u00fcnchen, Postfach 20 24 20, D-8000, M\u00fcnchen 2, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "von Stryk", 
        "givenName": "Oskar", 
        "id": "sg:person.014313516176.44", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014313516176.44"
        ], 
        "type": "Person"
      }
    ], 
    "datePublished": "1993", 
    "datePublishedReg": "1993-01-01", 
    "description": "When solving optimal control problems, indirect methods such as multiple shooting suffer from difficulties in finding an appropriate initial guess for the adjoint variables. For, this initial estimate must be provided for the iterative solution of the multipoint boundary-value problems arising from the necessary conditions of optimal control theory. Direct methods such as direct collocation do not suffer from this problem, but they generally yield results of lower accuracy and their iteration may even terminate with a non-optimal solution. Therefore, both methods are combined in such a way that the direct collocation method is at first applied to a simplified optimal control problem where all inequality constraints are neglected as long as the resulting problem is still well-defined. Because of the larger domain of convergence of the direct method, an approximation of the optimal solution of this problem can be obtained easier. The fusion between direct and indirect methods is then based on a relationship between the Lagrange multipliers of the underlying nonlinear programming problem to be solved by the direct method and the adjoint variables appearing in the necessary conditions which form the boundary-value problem to be solved by the indirect method. Hence, the adjoint variables, too, can be estimated from the approximation obtained by the direct method. This first step then facilitates the subsequent extension and competition of the model by homotopy techniques and the solution of the arising boundary-value problems by the indirect multiple shooting method. Proceeding in this way, the high accuracy and reliability of the multiple shooting method, especially the precise computation of the switching structure and the possibility to verify many necessary conditions, is preserved while disadvantages caused by the sensitive dependence on an appropriate estimate of the solution are considerably cut down. This procedure is described in detail for the numerical solution of the maximum-range trajectory optimization problem of a hang glider in an upwind which provides an example for a control problem where appropriate initial estimates for the adjoint variables are hard to find.", 
    "editor": [
      {
        "familyName": "Bulirsch", 
        "givenName": "R.", 
        "type": "Person"
      }, 
      {
        "familyName": "Miele", 
        "givenName": "A.", 
        "type": "Person"
      }, 
      {
        "familyName": "Stoer", 
        "givenName": "J.", 
        "type": "Person"
      }, 
      {
        "familyName": "Well", 
        "givenName": "K.", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-0348-7539-4_20", 
    "inLanguage": "en", 
    "isAccessibleForFree": false, 
    "isPartOf": {
      "isbn": [
        "978-3-0348-7541-7", 
        "978-3-0348-7539-4"
      ], 
      "name": "Optimal Control", 
      "type": "Book"
    }, 
    "keywords": [
      "boundary value problem", 
      "optimal control problem", 
      "multiple shooting method", 
      "adjoint variables", 
      "control problem", 
      "shooting method", 
      "indirect multiple shooting method", 
      "direct method", 
      "multipoint boundary value problem", 
      "necessary condition", 
      "optimal control theory", 
      "direct collocation method", 
      "trajectory optimization problem", 
      "appropriate initial estimates", 
      "nonlinear programming problem", 
      "appropriate initial guess", 
      "hang glider", 
      "initial estimates", 
      "non-optimal solutions", 
      "optimal control", 
      "direct collocation", 
      "multiple shooting", 
      "homotopy technique", 
      "collocation method", 
      "control theory", 
      "inequality constraints", 
      "optimization problem", 
      "iterative solution", 
      "numerical solution", 
      "Lagrange multipliers", 
      "programming problem", 
      "initial guess", 
      "optimal solution", 
      "sensitive dependence", 
      "switching structure", 
      "precise computation", 
      "approximation", 
      "indirect method", 
      "range maximization", 
      "appropriate estimates", 
      "large domain", 
      "problem", 
      "solution", 
      "high accuracy", 
      "adjoint", 
      "estimates", 
      "iteration", 
      "convergence", 
      "collocation", 
      "computation", 
      "guess", 
      "variables", 
      "low accuracy", 
      "accuracy", 
      "multipliers", 
      "theory", 
      "maximization", 
      "constraints", 
      "upwind", 
      "extension", 
      "subsequent extension", 
      "dependence", 
      "first step", 
      "model", 
      "conditions", 
      "glider", 
      "detail", 
      "way", 
      "technique", 
      "structure", 
      "reliability", 
      "domain", 
      "step", 
      "shooting", 
      "procedure", 
      "results", 
      "control", 
      "difficulties", 
      "possibility", 
      "disadvantages", 
      "competition", 
      "fusion", 
      "relationship", 
      "method", 
      "example"
    ], 
    "name": "Combining Direct and Indirect Methods in Optimal Control: Range Maximization of a Hang Glider", 
    "pagination": "273-288", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1004585554"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-0348-7539-4_20"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-0348-7539-4_20", 
      "https://app.dimensions.ai/details/publication/pub.1004585554"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-06-01T22:36", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20220601/entities/gbq_results/chapter/chapter_61.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-0348-7539-4_20"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/978-3-0348-7539-4_20'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-0348-7539-4_20'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-0348-7539-4_20'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-0348-7539-4_20'


 

This table displays all metadata directly associated to this object as RDF triples.

184 TRIPLES      23 PREDICATES      112 URIs      104 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-0348-7539-4_20 schema:about anzsrc-for:01
2 anzsrc-for:0102
3 anzsrc-for:0103
4 schema:author Nc5817d4bfcbf4b57b6a21991cf05f325
5 schema:datePublished 1993
6 schema:datePublishedReg 1993-01-01
7 schema:description When solving optimal control problems, indirect methods such as multiple shooting suffer from difficulties in finding an appropriate initial guess for the adjoint variables. For, this initial estimate must be provided for the iterative solution of the multipoint boundary-value problems arising from the necessary conditions of optimal control theory. Direct methods such as direct collocation do not suffer from this problem, but they generally yield results of lower accuracy and their iteration may even terminate with a non-optimal solution. Therefore, both methods are combined in such a way that the direct collocation method is at first applied to a simplified optimal control problem where all inequality constraints are neglected as long as the resulting problem is still well-defined. Because of the larger domain of convergence of the direct method, an approximation of the optimal solution of this problem can be obtained easier. The fusion between direct and indirect methods is then based on a relationship between the Lagrange multipliers of the underlying nonlinear programming problem to be solved by the direct method and the adjoint variables appearing in the necessary conditions which form the boundary-value problem to be solved by the indirect method. Hence, the adjoint variables, too, can be estimated from the approximation obtained by the direct method. This first step then facilitates the subsequent extension and competition of the model by homotopy techniques and the solution of the arising boundary-value problems by the indirect multiple shooting method. Proceeding in this way, the high accuracy and reliability of the multiple shooting method, especially the precise computation of the switching structure and the possibility to verify many necessary conditions, is preserved while disadvantages caused by the sensitive dependence on an appropriate estimate of the solution are considerably cut down. This procedure is described in detail for the numerical solution of the maximum-range trajectory optimization problem of a hang glider in an upwind which provides an example for a control problem where appropriate initial estimates for the adjoint variables are hard to find.
8 schema:editor N1112d8ed22bf4fb5acc719c9b7acb6f7
9 schema:genre chapter
10 schema:inLanguage en
11 schema:isAccessibleForFree false
12 schema:isPartOf Nf54070ed7f2848deb76cda71e411c1f6
13 schema:keywords Lagrange multipliers
14 accuracy
15 adjoint
16 adjoint variables
17 appropriate estimates
18 appropriate initial estimates
19 appropriate initial guess
20 approximation
21 boundary value problem
22 collocation
23 collocation method
24 competition
25 computation
26 conditions
27 constraints
28 control
29 control problem
30 control theory
31 convergence
32 dependence
33 detail
34 difficulties
35 direct collocation
36 direct collocation method
37 direct method
38 disadvantages
39 domain
40 estimates
41 example
42 extension
43 first step
44 fusion
45 glider
46 guess
47 hang glider
48 high accuracy
49 homotopy technique
50 indirect method
51 indirect multiple shooting method
52 inequality constraints
53 initial estimates
54 initial guess
55 iteration
56 iterative solution
57 large domain
58 low accuracy
59 maximization
60 method
61 model
62 multiple shooting
63 multiple shooting method
64 multipliers
65 multipoint boundary value problem
66 necessary condition
67 non-optimal solutions
68 nonlinear programming problem
69 numerical solution
70 optimal control
71 optimal control problem
72 optimal control theory
73 optimal solution
74 optimization problem
75 possibility
76 precise computation
77 problem
78 procedure
79 programming problem
80 range maximization
81 relationship
82 reliability
83 results
84 sensitive dependence
85 shooting
86 shooting method
87 solution
88 step
89 structure
90 subsequent extension
91 switching structure
92 technique
93 theory
94 trajectory optimization problem
95 upwind
96 variables
97 way
98 schema:name Combining Direct and Indirect Methods in Optimal Control: Range Maximization of a Hang Glider
99 schema:pagination 273-288
100 schema:productId Nb424dcc2922f47d2ae46a5877bb574b7
101 Nc9440ae12d4545248ca6446ef3fd6cb4
102 schema:publisher N05f9ed2181ad4bc1a60bfeef079264f3
103 schema:sameAs https://app.dimensions.ai/details/publication/pub.1004585554
104 https://doi.org/10.1007/978-3-0348-7539-4_20
105 schema:sdDatePublished 2022-06-01T22:36
106 schema:sdLicense https://scigraph.springernature.com/explorer/license/
107 schema:sdPublisher Neec714434f3541078780a3f063167cc0
108 schema:url https://doi.org/10.1007/978-3-0348-7539-4_20
109 sgo:license sg:explorer/license/
110 sgo:sdDataset chapters
111 rdf:type schema:Chapter
112 N05f9ed2181ad4bc1a60bfeef079264f3 schema:name Springer Nature
113 rdf:type schema:Organisation
114 N1112d8ed22bf4fb5acc719c9b7acb6f7 rdf:first Ndc83eff9399d458ca43c20c273d29cba
115 rdf:rest Ne14934f0dc5f4c1fa93c58774fa13547
116 N475cd6cf5e3246879dc74c7948320727 rdf:first sg:person.014313516176.44
117 rdf:rest rdf:nil
118 N4e84215d78a54754b19b1654f95829fe rdf:first Nddab82eb4dd641d68f8629c5d5ca2e67
119 rdf:rest N9acba20e92a549ad9f4f0528610c7bea
120 N620c859c37c94d0e8603f1fb8558648c schema:familyName Stoer
121 schema:givenName J.
122 rdf:type schema:Person
123 N7f07ef361a574603af361338a70420c9 schema:familyName Miele
124 schema:givenName A.
125 rdf:type schema:Person
126 N9acba20e92a549ad9f4f0528610c7bea rdf:first sg:person.014543723551.22
127 rdf:rest N475cd6cf5e3246879dc74c7948320727
128 Nb424dcc2922f47d2ae46a5877bb574b7 schema:name dimensions_id
129 schema:value pub.1004585554
130 rdf:type schema:PropertyValue
131 Nc5817d4bfcbf4b57b6a21991cf05f325 rdf:first sg:person.0767272315.31
132 rdf:rest N4e84215d78a54754b19b1654f95829fe
133 Nc9440ae12d4545248ca6446ef3fd6cb4 schema:name doi
134 schema:value 10.1007/978-3-0348-7539-4_20
135 rdf:type schema:PropertyValue
136 Ndc83eff9399d458ca43c20c273d29cba schema:familyName Bulirsch
137 schema:givenName R.
138 rdf:type schema:Person
139 Nddab82eb4dd641d68f8629c5d5ca2e67 schema:affiliation grid-institutes:grid.6936.a
140 schema:familyName Nerz
141 schema:givenName Edda
142 rdf:type schema:Person
143 Ne14934f0dc5f4c1fa93c58774fa13547 rdf:first N7f07ef361a574603af361338a70420c9
144 rdf:rest Nf63c012547a8477fb3b0438cabae7fa0
145 Neec714434f3541078780a3f063167cc0 schema:name Springer Nature - SN SciGraph project
146 rdf:type schema:Organization
147 Nf25a9ef8238a454caa679c396355e283 rdf:first Nf8df981bde7d48d386b6099c75da00af
148 rdf:rest rdf:nil
149 Nf54070ed7f2848deb76cda71e411c1f6 schema:isbn 978-3-0348-7539-4
150 978-3-0348-7541-7
151 schema:name Optimal Control
152 rdf:type schema:Book
153 Nf63c012547a8477fb3b0438cabae7fa0 rdf:first N620c859c37c94d0e8603f1fb8558648c
154 rdf:rest Nf25a9ef8238a454caa679c396355e283
155 Nf8df981bde7d48d386b6099c75da00af schema:familyName Well
156 schema:givenName K.
157 rdf:type schema:Person
158 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
159 schema:name Mathematical Sciences
160 rdf:type schema:DefinedTerm
161 anzsrc-for:0102 schema:inDefinedTermSet anzsrc-for:
162 schema:name Applied Mathematics
163 rdf:type schema:DefinedTerm
164 anzsrc-for:0103 schema:inDefinedTermSet anzsrc-for:
165 schema:name Numerical and Computational Mathematics
166 rdf:type schema:DefinedTerm
167 sg:person.014313516176.44 schema:affiliation grid-institutes:grid.6936.a
168 schema:familyName von Stryk
169 schema:givenName Oskar
170 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014313516176.44
171 rdf:type schema:Person
172 sg:person.014543723551.22 schema:affiliation grid-institutes:grid.6936.a
173 schema:familyName Pesch
174 schema:givenName Hans Josef
175 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014543723551.22
176 rdf:type schema:Person
177 sg:person.0767272315.31 schema:affiliation grid-institutes:grid.6936.a
178 schema:familyName Bulirsch
179 schema:givenName Roland
180 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0767272315.31
181 rdf:type schema:Person
182 grid-institutes:grid.6936.a schema:alternateName Mathematisches Institut, Technische Universität München, Postfach 20 24 20, D-8000, München 2, Germany
183 schema:name Mathematisches Institut, Technische Universität München, Postfach 20 24 20, D-8000, München 2, Germany
184 rdf:type schema:Organization
 




Preview window. Press ESC to close (or click here)


...