{"id":{"repo_id":"freiburg-diss","oai_identifier":"oai:freidok.uni-freiburg.de:1632"},"canonical_url":"https://search.dev.ndltd.org/etd/freiburg-diss/oai:freidok.uni-freiburg.de:1632","repository":{"repo_id":"freiburg-diss","name":"University of Freiburg","base_url":"https://freidok.uni-freiburg.de/oai/oai2.php"},"display":{"title":"Mapping with mobile robots","abstract":"Today, mapping is largely considered the most difficult perceptual
problem in robotics. Different problems like the statistical
dependence, the correspondence problem, or dynamic elements makes
mapping so hard. The basic technique for mapping is incremental scan
matching. We present a method, which maximizes the likelihood of a
scan, given a motion model and the map built so far. This approach is
the constitutive techniques for further methods which addresses
special topics of the previous described problems. In this thesis we
describe following methods:
We present a technique which combines a people tracker in the mapping
process to filter measurements caused by walking people. Additional
to the fact that maps contains less spurious objects, the accuracy of
the results can be increased.
We present a second technique which is able to filter measurements
caused by dynamic objects. This time dynamic objects are not limited
to walking persons, every dynamic object can be filtered out if the
place of this object was seen as both: as free space as well as
occupied space. An EM-technique is used to optimize the mapping
results and in several practical experiments we will show that this
approach can effectively filter beams of different dynamic objects
like humans, cars etc.
A famous problem in mapping is the so called closing loop problem.
When closing the cycle, the robot has to find out where it is relative
to its previously built map. This problem is complicated by the fact
that at the time of cycle closing, the robot's accumulated pose error
might be unboundedly large. We describe an efficient version of the
Rao-Blackwellized mapping approach. Using a Rao-Blackwellized
particle filter is a good techniques for estimating the posterior. In
combination with the scan matching technique we can decrease the
number of particles, so that it can be executed online.
The Rao-Blackwellized mapping approach suffers from the inherent
problem of all particle filters: the particle depletion problem. We
present a new algorithm for data association in SLAM. In essence, our
approach searches the combinatorial tree of possible data association
decisions. The search is lazy: only when an alternative assignment
shows promise will it be evaluated. We tested our approach, and
consistently found that it produces accurate maps, even if for maps
with many large cycles.
It can be difficult to distinguish places with data from range sensors
as measurements can look the same at different positions. RFID tags
have to nice property to be unique, they can report their
identification numbers which are easy to distinguish. Unfortunately
they don't provide any distance information, so that estimating the
location of the RFID tags is a difficult problem. We present an
approach to generate maps of RFID tags with mobile robots. We present
a sensor model that allows us to compute the likelihood of tag
detections given the relative position of the tag with respect to the
robot. Additionally we describe how to compute a posterior about the
position of a tag after the trajectory and the map is generated with a
highly accurate Rao-Blackwellized mapping algorithm for laser range
scans.
We propose a technique for simultaneous scan registration and scan
deformation for modeling nonrigid objects. The deformation is made
possible through the definition of (soft) links between neighboring
scan points, whose configuration is calculated during registration.
To tackle the resulting optimization problem efficiently, we describe
a hierarchical optimization techniques that operate on thinned graphs.
Experimental results obtained using a mobile robot illustrate the
viability of this approach.
Finally, we present two examples for complete systems which are able
to map their environment. The first system designed to autonomously
explore and acquire 3D maps of abandoned mines is a 1,500 pound
vehicle, nicknamed \"Groundhog\". The core of the Groundhog navigation
system is comprised of a software package that solves the SLAM problem
by acquiring 2D maps. Our research demonstrates that the autonomous
acquisition of maps of abandoned mines is indeed feasible with
autonomous robotic systems. The second system is an instrumented
helicopter platform for 3D ground modeling. A real-time algorithm is
developed that integrates pose estimates from multiple sensors with
range data, acquired by a 2D laser range finders oriented
perpendicular to the vehicle's flight direction. The algorithm uses a
fast optimization technique to generate maps in real-time.","abstract_html":"Today, mapping is largely considered the most difficult perceptual <br>problem in robotics. Different problems like the statistical <br>dependence, the correspondence problem, or dynamic elements makes <br>mapping so hard. The basic technique for mapping is incremental scan <br>matching. We present a method, which maximizes the likelihood of a <br>scan, given a motion model and the map built so far. This approach is <br>the constitutive techniques for further methods which addresses <br>special topics of the previous described problems. In this thesis we <br>describe following methods: <br> <br>We present a technique which combines a people tracker in the mapping <br>process to filter measurements caused by walking people. Additional <br>to the fact that maps contains less spurious objects, the accuracy of <br>the results can be increased. <br> <br>We present a second technique which is able to filter measurements <br>caused by dynamic objects. This time dynamic objects are not limited <br>to walking persons, every dynamic object can be filtered out if the <br>place of this object was seen as both: as free space as well as <br>occupied space. An EM-technique is used to optimize the mapping <br>results and in several practical experiments we will show that this <br>approach can effectively filter beams of different dynamic objects <br>like humans, cars etc. <br> <br>A famous problem in mapping is the so called closing loop problem. <br>When closing the cycle, the robot has to find out where it is relative <br>to its previously built map. This problem is complicated by the fact <br>that at the time of cycle closing, the robot's accumulated pose error <br>might be unboundedly large. We describe an efficient version of the <br>Rao-Blackwellized mapping approach. Using a Rao-Blackwellized <br>particle filter is a good techniques for estimating the posterior. In <br>combination with the scan matching technique we can decrease the <br>number of particles, so that it can be executed online. <br> <br>The Rao-Blackwellized mapping approach suffers from the inherent <br>problem of all particle filters: the particle depletion problem. We <br>present a new algorithm for data association in SLAM. In essence, our <br>approach searches the combinatorial tree of possible data association <br>decisions. The search is lazy: only when an alternative assignment <br>shows promise will it be evaluated. We tested our approach, and <br>consistently found that it produces accurate maps, even if for maps <br>with many large cycles. <br> <br>It can be difficult to distinguish places with data from range sensors <br>as measurements can look the same at different positions. RFID tags <br>have to nice property to be unique, they can report their <br>identification numbers which are easy to distinguish. Unfortunately <br>they don't provide any distance information, so that estimating the <br>location of the RFID tags is a difficult problem. We present an <br>approach to generate maps of RFID tags with mobile robots. We present <br>a sensor model that allows us to compute the likelihood of tag <br>detections given the relative position of the tag with respect to the <br>robot. Additionally we describe how to compute a posterior about the <br>position of a tag after the trajectory and the map is generated with a <br>highly accurate Rao-Blackwellized mapping algorithm for laser range <br>scans. <br> <br>We propose a technique for simultaneous scan registration and scan <br>deformation for modeling nonrigid objects. The deformation is made <br>possible through the definition of (soft) links between neighboring <br>scan points, whose configuration is calculated during registration. <br>To tackle the resulting optimization problem efficiently, we describe <br>a hierarchical optimization techniques that operate on thinned graphs. <br>Experimental results obtained using a mobile robot illustrate the <br>viability of this approach. <br> <br>Finally, we present two examples for complete systems which are able <br>to map their environment. The first system designed to autonomously <br>explore and acquire 3D maps of abandoned mines is a 1,500 pound <br>vehicle, nicknamed "Groundhog". The core of the Groundhog navigation <br>system is comprised of a software package that solves the SLAM problem <br>by acquiring 2D maps. Our research demonstrates that the autonomous <br>acquisition of maps of abandoned mines is indeed feasible with <br>autonomous robotic systems. The second system is an instrumented <br>helicopter platform for 3D ground modeling. A real-time algorithm is <br>developed that integrates pose estimates from multiple sensors with <br>range data, acquired by a 2D laser range finders oriented <br>perpendicular to the vehicle's flight direction. The algorithm uses a <br>fast optimization technique to generate maps in real-time.","abstract_has_math":false,"creators":["Hähnel, Dirk"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Burgard, Wolfram"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-07-24T02:22:22Z","subjects":["Mobile Roboter","Zustandsschätzung","Kartenbau","mobile robots","state estimation","map building"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://freidok.uni-freiburg.de/data/1632","outbound_label":"Repository record","outbound_source":"source_url"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Burgard, Wolfram"]},{"key":"dc:creator","label":"Author","values":["Hähnel, Dirk"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:type","label":"Dc Type","values":["DoctoralThesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Mobile Roboter","Zustandsschätzung","Kartenbau","mobile robots","state estimation","map building"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Today, mapping is largely considered the most difficult perceptual
problem in robotics. Different problems like the statistical
dependence, the correspondence problem, or dynamic elements makes
mapping so hard. The basic technique for mapping is incremental scan
matching. We present a method, which maximizes the likelihood of a
scan, given a motion model and the map built so far. This approach is
the constitutive techniques for further methods which addresses
special topics of the previous described problems. In this thesis we
describe following methods:
We present a technique which combines a people tracker in the mapping
process to filter measurements caused by walking people. Additional
to the fact that maps contains less spurious objects, the accuracy of
the results can be increased.
We present a second technique which is able to filter measurements
caused by dynamic objects. This time dynamic objects are not limited
to walking persons, every dynamic object can be filtered out if the
place of this object was seen as both: as free space as well as
occupied space. An EM-technique is used to optimize the mapping
results and in several practical experiments we will show that this
approach can effectively filter beams of different dynamic objects
like humans, cars etc.
A famous problem in mapping is the so called closing loop problem.
When closing the cycle, the robot has to find out where it is relative
to its previously built map. This problem is complicated by the fact
that at the time of cycle closing, the robot's accumulated pose error
might be unboundedly large. We describe an efficient version of the
Rao-Blackwellized mapping approach. Using a Rao-Blackwellized
particle filter is a good techniques for estimating the posterior. In
combination with the scan matching technique we can decrease the
number of particles, so that it can be executed online.
The Rao-Blackwellized mapping approach suffers from the inherent
problem of all particle filters: the particle depletion problem. We
present a new algorithm for data association in SLAM. In essence, our
approach searches the combinatorial tree of possible data association
decisions. The search is lazy: only when an alternative assignment
shows promise will it be evaluated. We tested our approach, and
consistently found that it produces accurate maps, even if for maps
with many large cycles.
It can be difficult to distinguish places with data from range sensors
as measurements can look the same at different positions. RFID tags
have to nice property to be unique, they can report their
identification numbers which are easy to distinguish. Unfortunately
they don't provide any distance information, so that estimating the
location of the RFID tags is a difficult problem. We present an
approach to generate maps of RFID tags with mobile robots. We present
a sensor model that allows us to compute the likelihood of tag
detections given the relative position of the tag with respect to the
robot. Additionally we describe how to compute a posterior about the
position of a tag after the trajectory and the map is generated with a
highly accurate Rao-Blackwellized mapping algorithm for laser range
scans.
We propose a technique for simultaneous scan registration and scan
deformation for modeling nonrigid objects. The deformation is made
possible through the definition of (soft) links between neighboring
scan points, whose configuration is calculated during registration.
To tackle the resulting optimization problem efficiently, we describe
a hierarchical optimization techniques that operate on thinned graphs.
Experimental results obtained using a mobile robot illustrate the
viability of this approach.
Finally, we present two examples for complete systems which are able
to map their environment. The first system designed to autonomously
explore and acquire 3D maps of abandoned mines is a 1,500 pound
vehicle, nicknamed \"Groundhog\". The core of the Groundhog navigation
system is comprised of a software package that solves the SLAM problem
by acquiring 2D maps. Our research demonstrates that the autonomous
acquisition of maps of abandoned mines is indeed feasible with
autonomous robotic systems. The second system is an instrumented
helicopter platform for 3D ground modeling. A real-time algorithm is
developed that integrates pose estimates from multiple sensors with
range data, acquired by a 2D laser range finders oriented
perpendicular to the vehicle's flight direction. The algorithm uses a
fast optimization technique to generate maps in real-time.","Einer der entscheidenden Faktoren bei der Konstruktion eines
vollständig autonomen mobilen Roboters ist das Erstellen von Karten.
Diese Arbeit bietet Lösungsmöglichkeiten für unterschiedliche
Fragestellungen, die beim Kartenbau auftreten. Üblicherweise
beinhalten Karten nur die Beschreibungen von statischen Objekten in
der Umgebung, wie zum Beispiel Wände, Gebäude oder Straßen.
Dynamische Elemente, umhergehende Menschen oder ähnliches sind im
Allgemeinen nicht erwünscht. Ziel ist es daher, die Messungen, die
durch diese dynamischen Objekte verursacht werden, zu erkennen und zu
filtern. Neben der Reduzierung von unerwünschten Gegenständen in der
Karte kann dadurch auch die Genauigkeit der Karte erhöht werden. In
zahlreichen Experimenten wird demonstriert, dass die gefundenen
Methoden effektiv Messungen von diversen dynamischen Objekten, wie zum
Beispiel von Personen oder Autos detektieren und filtern können. Der
inkrementelle Vergleich von Abstandsbildern, das so genannte
\"Incremental Scan Matching\", ist ein elementarer Bestandteil vieler
hier gezeigter Ansätze. Diese Methode ermöglicht die Korrektur
lokaler Fehler und erlaubt die Erzeugung örtlich konsistenter Karten.
Kleinere Ungenauigkeiten in der Positionsbestimmung können sich
allerdings über die Zeit akkumulieren, und es kann daher leicht
passieren, dass eine globale Inkonsistenz eintritt, sobald ein vorher
gesehener Bereich wieder betreten wird. Eine weitere Schwierigkeit
beim Kartenbau ist demzufolge häufig das korrekte Schließen von
Zyklen. Hierbei muss der Roboter den akkumulierten Positionsfehler
erkennen und korrigieren, auch wenn dieser Fehler erheblich sein
sollte. Für diese Problematik werden zwei unterschiedliche Lösungen
vorgestellt. Die Erste basiert auf dem Einsatz von Rao-Blackwellized
Partikelfiltern, der allerdings die Notwendigkeit von erheblichen
Berechnungen impliziert. Es wird gezeigt, dass durch eine Kombination
mit dem vorgenannten Scan Matching, eine Realisierung in Echtzeit
möglich wird. Daneben wird ein Verfahren vorgestellt, das
Kartenabschnitte kombinatorisch verwaltet und Beziehungen zwischen den
Abschnitten nachträglich ändern kann, sofern ein Fehler erkannt wird.
In Experimenten wird gezeigt, dass mit Hilfe dieser Methoden Karten
von Umgebungen mit zahlreichen, auch verschachtelten, Schleifen
erstellt werden können. Neben des Problems der Erstellung von
geometrischen Karten werden in der vorliegenden Arbeit Techniken zur
Ermittlung von Position und Form einzelner Objekte vorgestellt. Zur
eindeutigen Identifizierung der Objekte werden in der Praxis
RFID-Systeme immer populärer. Die genaue Ortung der so genannten
RFID-Marken ist jedoch schwierig, da sie ihrerseits keine
Abstandsinformationen liefern und der Detektionsbereich der Antennen
mehrere Meter betragen kann. Es wird ein Verfahren gezeigt, das mit
Hilfe von mobilen Robotern detaillierte Karten dieser RFID-Marken
erstellt. Eine andere Technik ermöglicht es, aus Teilansichten eines
deformierbaren Objekts ein komplettes Modell zu generieren. Zur
Berechnung der Deformation werden die genauen Konfigurationen von
beweglichen Verbindungen zwischen benachbarten Messpunkten
herangezogen. Um das resultierende Optimierungsproblem zu
vereinfachen, wird eine hierarchische Darstellung benutzt.
Experimentelle Resultate zeigen die Realisierbarkeit des Ansatzes.
Schließlich werden, als Anwendungen der dargestellten Techniken, zwei
komplexe Systeme vorgestellt, die in der Lage sind, autonom ihre
Umgebung zu modellieren. Das erste System, das entwickelt wurde, um
dreidimensionale Karten von stillgelegten Minen zu erstellen, ist ein
800 kg schwerer Roboter mit dem Namen \"Groundhog\". Das
Basisnavigationssystem beinhaltet ein Softwarepaket, das das Problem
des Kartenbaus mittels zweidimensionaler Karten löst. Daneben wurde
ein mit unterschiedlichen Sensoren ausgestatteter Modellhelikopter
entwickelt, der eine dreidimensionale Karte eines Geländes erzeugen
kann. Ein zweidimensionales Lasermesssytem, das senkrecht zur
Flugrichtung orientiert ist, erlaubt es, in Echtzeit diese Karten zu
generieren."]},{"key":"dc:format.medium","label":"Dc Format Medium","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Mapping with mobile robots","Kartenbau durch mobile Roboter"]}]}],"canonical_facts":{"dc:contributor":["Burgard, Wolfram"],"dc:creator":["Hähnel, Dirk"],"dc:description.abstract":["Today, mapping is largely considered the most difficult perceptual
problem in robotics. Different problems like the statistical
dependence, the correspondence problem, or dynamic elements makes
mapping so hard. The basic technique for mapping is incremental scan
matching. We present a method, which maximizes the likelihood of a
scan, given a motion model and the map built so far. This approach is
the constitutive techniques for further methods which addresses
special topics of the previous described problems. In this thesis we
describe following methods:
We present a technique which combines a people tracker in the mapping
process to filter measurements caused by walking people. Additional
to the fact that maps contains less spurious objects, the accuracy of
the results can be increased.
We present a second technique which is able to filter measurements
caused by dynamic objects. This time dynamic objects are not limited
to walking persons, every dynamic object can be filtered out if the
place of this object was seen as both: as free space as well as
occupied space. An EM-technique is used to optimize the mapping
results and in several practical experiments we will show that this
approach can effectively filter beams of different dynamic objects
like humans, cars etc.
A famous problem in mapping is the so called closing loop problem.
When closing the cycle, the robot has to find out where it is relative
to its previously built map. This problem is complicated by the fact
that at the time of cycle closing, the robot's accumulated pose error
might be unboundedly large. We describe an efficient version of the
Rao-Blackwellized mapping approach. Using a Rao-Blackwellized
particle filter is a good techniques for estimating the posterior. In
combination with the scan matching technique we can decrease the
number of particles, so that it can be executed online.
The Rao-Blackwellized mapping approach suffers from the inherent
problem of all particle filters: the particle depletion problem. We
present a new algorithm for data association in SLAM. In essence, our
approach searches the combinatorial tree of possible data association
decisions. The search is lazy: only when an alternative assignment
shows promise will it be evaluated. We tested our approach, and
consistently found that it produces accurate maps, even if for maps
with many large cycles.
It can be difficult to distinguish places with data from range sensors
as measurements can look the same at different positions. RFID tags
have to nice property to be unique, they can report their
identification numbers which are easy to distinguish. Unfortunately
they don't provide any distance information, so that estimating the
location of the RFID tags is a difficult problem. We present an
approach to generate maps of RFID tags with mobile robots. We present
a sensor model that allows us to compute the likelihood of tag
detections given the relative position of the tag with respect to the
robot. Additionally we describe how to compute a posterior about the
position of a tag after the trajectory and the map is generated with a
highly accurate Rao-Blackwellized mapping algorithm for laser range
scans.
We propose a technique for simultaneous scan registration and scan
deformation for modeling nonrigid objects. The deformation is made
possible through the definition of (soft) links between neighboring
scan points, whose configuration is calculated during registration.
To tackle the resulting optimization problem efficiently, we describe
a hierarchical optimization techniques that operate on thinned graphs.
Experimental results obtained using a mobile robot illustrate the
viability of this approach.
Finally, we present two examples for complete systems which are able
to map their environment. The first system designed to autonomously
explore and acquire 3D maps of abandoned mines is a 1,500 pound
vehicle, nicknamed \"Groundhog\". The core of the Groundhog navigation
system is comprised of a software package that solves the SLAM problem
by acquiring 2D maps. Our research demonstrates that the autonomous
acquisition of maps of abandoned mines is indeed feasible with
autonomous robotic systems. The second system is an instrumented
helicopter platform for 3D ground modeling. A real-time algorithm is
developed that integrates pose estimates from multiple sensors with
range data, acquired by a 2D laser range finders oriented
perpendicular to the vehicle's flight direction. The algorithm uses a
fast optimization technique to generate maps in real-time.","Einer der entscheidenden Faktoren bei der Konstruktion eines
vollständig autonomen mobilen Roboters ist das Erstellen von Karten.
Diese Arbeit bietet Lösungsmöglichkeiten für unterschiedliche
Fragestellungen, die beim Kartenbau auftreten. Üblicherweise
beinhalten Karten nur die Beschreibungen von statischen Objekten in
der Umgebung, wie zum Beispiel Wände, Gebäude oder Straßen.
Dynamische Elemente, umhergehende Menschen oder ähnliches sind im
Allgemeinen nicht erwünscht. Ziel ist es daher, die Messungen, die
durch diese dynamischen Objekte verursacht werden, zu erkennen und zu
filtern. Neben der Reduzierung von unerwünschten Gegenständen in der
Karte kann dadurch auch die Genauigkeit der Karte erhöht werden. In
zahlreichen Experimenten wird demonstriert, dass die gefundenen
Methoden effektiv Messungen von diversen dynamischen Objekten, wie zum
Beispiel von Personen oder Autos detektieren und filtern können. Der
inkrementelle Vergleich von Abstandsbildern, das so genannte
\"Incremental Scan Matching\", ist ein elementarer Bestandteil vieler
hier gezeigter Ansätze. Diese Methode ermöglicht die Korrektur
lokaler Fehler und erlaubt die Erzeugung örtlich konsistenter Karten.
Kleinere Ungenauigkeiten in der Positionsbestimmung können sich
allerdings über die Zeit akkumulieren, und es kann daher leicht
passieren, dass eine globale Inkonsistenz eintritt, sobald ein vorher
gesehener Bereich wieder betreten wird. Eine weitere Schwierigkeit
beim Kartenbau ist demzufolge häufig das korrekte Schließen von
Zyklen. Hierbei muss der Roboter den akkumulierten Positionsfehler
erkennen und korrigieren, auch wenn dieser Fehler erheblich sein
sollte. Für diese Problematik werden zwei unterschiedliche Lösungen
vorgestellt. Die Erste basiert auf dem Einsatz von Rao-Blackwellized
Partikelfiltern, der allerdings die Notwendigkeit von erheblichen
Berechnungen impliziert. Es wird gezeigt, dass durch eine Kombination
mit dem vorgenannten Scan Matching, eine Realisierung in Echtzeit
möglich wird. Daneben wird ein Verfahren vorgestellt, das
Kartenabschnitte kombinatorisch verwaltet und Beziehungen zwischen den
Abschnitten nachträglich ändern kann, sofern ein Fehler erkannt wird.
In Experimenten wird gezeigt, dass mit Hilfe dieser Methoden Karten
von Umgebungen mit zahlreichen, auch verschachtelten, Schleifen
erstellt werden können. Neben des Problems der Erstellung von
geometrischen Karten werden in der vorliegenden Arbeit Techniken zur
Ermittlung von Position und Form einzelner Objekte vorgestellt. Zur
eindeutigen Identifizierung der Objekte werden in der Praxis
RFID-Systeme immer populärer. Die genaue Ortung der so genannten
RFID-Marken ist jedoch schwierig, da sie ihrerseits keine
Abstandsinformationen liefern und der Detektionsbereich der Antennen
mehrere Meter betragen kann. Es wird ein Verfahren gezeigt, das mit
Hilfe von mobilen Robotern detaillierte Karten dieser RFID-Marken
erstellt. Eine andere Technik ermöglicht es, aus Teilansichten eines
deformierbaren Objekts ein komplettes Modell zu generieren. Zur
Berechnung der Deformation werden die genauen Konfigurationen von
beweglichen Verbindungen zwischen benachbarten Messpunkten
herangezogen. Um das resultierende Optimierungsproblem zu
vereinfachen, wird eine hierarchische Darstellung benutzt.
Experimentelle Resultate zeigen die Realisierbarkeit des Ansatzes.
Schließlich werden, als Anwendungen der dargestellten Techniken, zwei
komplexe Systeme vorgestellt, die in der Lage sind, autonom ihre
Umgebung zu modellieren. Das erste System, das entwickelt wurde, um
dreidimensionale Karten von stillgelegten Minen zu erstellen, ist ein
800 kg schwerer Roboter mit dem Namen \"Groundhog\". Das
Basisnavigationssystem beinhaltet ein Softwarepaket, das das Problem
des Kartenbaus mittels zweidimensionaler Karten löst. Daneben wurde
ein mit unterschiedlichen Sensoren ausgestatteter Modellhelikopter
entwickelt, der eine dreidimensionale Karte eines Geländes erzeugen
kann. Ein zweidimensionales Lasermesssytem, das senkrecht zur
Flugrichtung orientiert ist, erlaubt es, in Echtzeit diese Karten zu
generieren."],"dc:format.medium":["application/pdf"],"dc:subject":["Mobile Roboter","Zustandsschätzung","Kartenbau","mobile robots","state estimation","map building"],"dc:title":["Mapping with mobile robots","Kartenbau durch mobile Roboter"],"dc:type":["DoctoralThesis"]},"updated_at":"2026-07-24T02:22:22Z"}