Vorwort | 5 |
Inhaltsverzeichnis | 7 |
Advanced Data Logging in RoboCup | 11 |
1 Introduction | 11 |
2 Related Work | 13 |
3 Shared World Model | 13 |
4 Camera Framework and PTU | 13 |
4.1 Observer | 14 |
4.2 Rendering Framework | 15 |
5 Camera Pose Framework | 15 |
5.1 Camera Pose Calculation | 16 |
6 Summary and Future Work | 17 |
References | 18 |
Data Association for Visual Multi-target Tracking Under Splits, Merges and Occlusions | 19 |
1 Introduction | 19 |
2 Overall System Description | 20 |
3 Time Forward Prediction | 21 |
4 Data Association Between Detections and Tracks | 22 |
5 Handling of Split and Merge Effects and Reconstruction of Compatible Object Measurements | 23 |
5.1 Determination of the Point-to-Track Affiliation Probabilities | 24 |
5.2 Point Cloud Based Reconstruction of the Track’s Position and Orientation | 24 |
5.3 Grid-Based Reconstruction of the Track’s Extent | 25 |
6 State, Existence and Observability Innovation | 25 |
7 Experimental Results | 26 |
8 Conclusion | 26 |
References | 26 |
Fusing LIDAR and Vision for Autonomous Dirt Road Following – Incorporating a Visual Feature into the Tentacles Approach | 27 |
1 Introduction | 27 |
2 Driving with Tentacles | 28 |
2.1 Tentacle Evaluation and Selection | 28 |
3 Evaluating Tentacles by Visual Appearances | 29 |
3.1 Perspective Mapping and Gaze Control | 29 |
3.2 Determine Visual Quality | 30 |
4 Rating Tentacles by Color Intensity Feature | 31 |
5 Results and Conclusions | 33 |
6 Acknowledgements | 34 |
References | 34 |
Improved Time-to-Contact Estimation by Using Information from Image Sequences | 35 |
1 Introduction | 35 |
2 Time-to-Contact Calculation | 36 |
2.1 Theory of Time-to-Contact Calculation | 36 |
2.2 TTC Estimation Using Model Equations | 37 |
3 Adaptive Parameter Estimation and Its Effect | 38 |
3.1 Iterative Parameter Estimation | 38 |
3.2 Results of the Proposed Estimation Procedure | 38 |
4 Extension of the Approach to General Movements | 39 |
4.1 Forward Movement with Changing Velocity | 40 |
4.2 Simple Rotational Movement | 40 |
4.3 Cornering | 41 |
4.4 Results | 41 |
5 Conclusions | 42 |
References | 42 |
Monocular Obstacle Detection for Real-World Environments | 43 |
1 Introduction and RelatedWork | 43 |
2 Monocular Scene Reconstruction | 44 |
2.1 State Representation | 45 |
2.2 Feature Tracking | 45 |
2.3 Descriptor Update | 46 |
2.4 Feature Initialization | 47 |
3 Obstacle Detection | 47 |
4 Results | 47 |
5 Conclusion and Future Work | 50 |
References | 50 |
Stereo-Based vs. Monocular 6-DoF Pose Estimation Using Point Features: A Quantitative Comparison | 51 |
1 Introduction | 51 |
2 Accuracy Considerations | 53 |
3 6-DoF Pose Estimation | 54 |
4 Experimental Evaluation | 55 |
5 Discussion and Outlook | 57 |
Acknowledgment | 57 |
References | 58 |
Probabilistisches Belegtheitsfilter zur Schätzungdynamischer Umgebungen unter Verwendung multipler Bewegungsmodelle | 59 |
1 Einleitung | 59 |
2 Probabilistische Belegheitsfilterung mit Gruppen | 60 |
2.1 Zellfeldrepräsentation | 60 |
2.2 Dekomposition der Verbundwahrscheinlichkeit | 61 |
2.3 Modellierung der Gruppen | 62 |
2.4 Berechnung der Filtergleichungen | 63 |
3 Evaluierung | 65 |
4 Zusammenfassung und Ausblick | 66 |
5 Danksagung | 66 |
Literaturverzeichnis | 66 |
A Computational Model of Human Table Tennis for Robot Application | 67 |
1 Introduction | 67 |
1.1 What Can We Learn from Human Table Tennis? | 67 |
1.2 A Review of Robot Table Tennis | 68 |
2 Modelling Human Table Tennis | 69 |
2.1 Movement Phases | 69 |
2.2 Movement Primitive Selection and Parametrization | 69 |
2.3 Movement Generation | 70 |
3 Computational Realization of the Model | 70 |
3.1 Overview | 70 |
3.2 Determining the Goal Parameters | 71 |
3.3 Trajectory Generation | 72 |
4 Evaluations | 72 |
4.1 Simulated Setup | 73 |
4.2 Performance of the Model | 73 |
5 Conclusion | 73 |
References | 74 |
A Vision-Based Trajectory Controller for Autonomous Cleaning Robots | 75 |
1 Introduction | 75 |
2 Biologically Inspired Visual Navigation | 76 |
2.1 Local Visual Homing | 76 |
2.2 Topological Navigation | 77 |
3 Trajectory Controller | 77 |
4 Experiments | 78 |
4.1 Setup | 78 |
4.2 Results | 80 |
5 Summary and Outlook | 81 |
References | 82 |
Automatic Take Off, Hovering and Landing Control for Miniature Helicopters with Low-Cost Onboard Hardware | 83 |
1 Introduction | 83 |
2 Features of the Wii Remote Infrared Camera | 84 |
3 The UAV System | 85 |
4 Retrieving the Pose by Means of Camera and IMU | 86 |
4.1 Pattern Analysis | 86 |
5 Flight Control | 86 |
5.1 Height Controller | 87 |
5.2 Roll/Pitch and Yaw Controller | 87 |
6 Experimental Results | 87 |
7 Conclusion and Future Work | 89 |
References | 90 |
Foot Function in Spring Mass Running | 91 |
1 Introduction | 91 |
2 Methods | 92 |
2.1 Spring-Mass-Model with Foot-Segment | 92 |
2.2 Analyzing Periodic Solutions | 92 |
2.3 Dimensionless Ankle Stiffness | 93 |
3 Results | 94 |
4 Discussion | 95 |
4.1 Leg Stiffness Modulation | 96 |
4.2 Leg Lengthening | 97 |
4.3 Shortcommings of the Model | 97 |
References | 98 |
From Walking to Running | 99 |
1 Introduction | 99 |
2 Methods | 100 |
3 Results | 102 |
4 Discussion | 104 |
Acknowledgments | 106 |
References | 106 |
Generisches Verfahren zur präzisen Pfadverfolgung für Serienfahrzeuggespanne | 107 |
1 Einleitung | 107 |
2 Stand der Technik | 108 |
3 Kinematisches Modell | 109 |
4 Das zweistufige Regelungsverfahren | 110 |
4.1 Bestimmung des Soll-Radius | 110 |
4.2 Bestimmung des Soll-Lenkwinkels | 112 |
5 Evaluierung des Verfahrens | 113 |
6 Zusammenfassung | 114 |
Literaturverzeichnis | 114 |
Learning New Basic Movements for Robotics | 115 |
1 Introduction | 115 |
2 Learning Methods for Motor Primitives | 116 |
2.1 Imitation Learning for Dynamical Motor Primitives | 117 |
2.2 Reinforcement Learning with PoWER | 117 |
3 Robot Evaluation | 118 |
3.1 Discrete Movement: Ball-in-a-Cup | 119 |
3.2 Rhythmic Movement with Start-up Phase: Ball-Paddling | 120 |
4 Conclusion | 121 |
References | 122 |
Nonlinear Landing Control for Quadrotor UAVs | 123 |
1 Introduction | 123 |
2 The Vehicle Control System | 124 |
3 Automatic Landing on a Mobile Platform | 126 |
3.1 Altitude Control | 126 |
3.2 Nonlinear 2D-Tracking Controller | 127 |
4 Simulation and Experimental Results | 129 |
5 Conclusion and Future Works | 130 |
6 References | 130 |
Oscillation Analysis in Behavior-Based Robot Architectures | 131 |
1 Introduction | 131 |
2 Development Framework | 132 |
3 Oscillation Analysis | 133 |
4 Application and Results | 135 |
5 Conclusion and Future Work | 137 |
Acknowledgments | 138 |
References | 138 |
Variable Joint Elasticities in Running | 139 |
1 Introduction | 139 |
2 Materials and Methods | 140 |
2.1 Experimental Data | 140 |
2.2 Dynamic Joint Stiffness Analysis | 140 |
3 Results | 142 |
4 Discussion | 144 |
Acknowledgements | 146 |
References | 146 |
3D-Partikelfilter SLAM | 147 |
1 Einleitung | 147 |
2 Das partikelfilterbasierte SLAM Verfahren | 148 |
2.1 Ziehen neuer Positionen | 149 |
2.2 Gewichtung der Partikel | 150 |
2.3 Resampling | 150 |
3 Reduktion des Speicherplatzes | 151 |
4 Experimentelle Ergebnisse und Evaluation | 152 |
5 Schlussfolgerung und Ausblick | 153 |
Literaturverzeichnis | 154 |
Absolute High-Precision Localisation of an Unmanned Ground Vehicle by Using Real-Time Aerial Video Imagery for Geo-referenced Orthophoto Registration | 155 |
1 Introduction | 155 |
2 Robotic Platforms | 156 |
3 Cooperative UGV/UAV Localisation | 157 |
3.1 Video Tracking of UGV | 157 |
3.2 Pitch and Roll Compensation | 158 |
3.3 Orthophoto Registration | 159 |
4 Results | 160 |
5 Conclusion | 161 |
References | 162 |
An Improved Sensor Model on Appearance Based SLAM | 163 |
1 Introduction | 163 |
2 Appearance-Based SLAM Approach with RBPF | 165 |
2.1 RBPF with Local and Global Graph Models | 165 |
2.2 Graph Matching | 166 |
2.3 Adaptive Sensor Model | 166 |
2.4 Dynamic Changes | 168 |
3 Experiments and Results | 168 |
4 Conclusion and Future Work | 169 |
References | 170 |
Monte Carlo Lokalisierung FahrerloserTransportfahrzeuge mit drahtlosen Sensornetzwerken | 171 |
1 Einf¨uhrung | 171 |
2 Funklokalisierung | 172 |
2.1 Stand der Technik | 172 |
2.2 Funklokalisierung mit dem nanoLOC System | 173 |
3 Monte Carlo Funklokalisierung | 173 |
3.1 Anfangsverteilung der Partikel | 174 |
3.2 Sensormodell der Distanzmessung | 174 |
3.3 Experimentelle Ergebnisse | 176 |
4 Zusammenfassung | 177 |
Literaturverzeichnis | 178 |
Using a Physics Engine to Improve Probabilistic Object Localization | 179 |
1 Introduction | 179 |
2 State of the Art | 180 |
3 Theoretical Concept | 180 |
3.1 Rule Set Joint State Update | 181 |
4 Implementation of the Proposed System | 181 |
4.1 Probabilistic Models | 181 |
4.2 The Trigger Function | 182 |
4.3 Implementation of the Rule Set Joint State Update | 183 |
4.4 The Physical Model: Bullet Physic Engine | 183 |
5 Experimental Results | 184 |
6 Conclusions and Future Works | 185 |
Acknowledgment | 186 |
References | 186 |
Visual Self-Localization with Tiny Images | 187 |
1 Introduction | 187 |
2 Related Work | 188 |
3 Robots | 189 |
4 Global Image Features | 189 |
4.1 Weighted Gradient Orientation Histograms | 189 |
4.2 Weighted Grid Integral Invariants | 190 |
4.3 Color/Grayscale Grid Histograms | 190 |
4.4 Pixelwise Image Comparison | 190 |
5 Localization Process | 190 |
6 Experimental Results | 191 |
7 Conclusion | 194 |
References | 194 |
Coordinated Path Following for Mobile Robots | 195 |
1 Introduction | 195 |
2 Problem Statement | 196 |
3 Controller Design | 197 |
4 Results | 199 |
4.1 Simulations | 199 |
4.2 Real-World Experiments | 200 |
5 Conclusions and Future Work | 201 |
References | 201 |
Kooperative Bewegungsplanung zur Unfallvermeidung im Straßenverkehr mit der Methode der elastischen Bänder | 203 |
1 Einleitung | 203 |
2 Anwendungsszenario | 204 |
3 Kooperative Bewegungsplanung | 204 |
3.1 Problemstellung | 204 |
3.2 Ansätze aus der Robotik | 205 |
4 Die Methode der elastischen Bänder | 206 |
4.1 Grundlagen | 206 |
4.2 Modellierung der Kräfte | 206 |
4.3 Numerische Kräfteminimierung | 208 |
4.4 Interpolation der Bewegung | 208 |
5 Simulationsergebnisse | 208 |
6 Fazit und Ausblick | 210 |
Danksagung | 210 |
Literaturverzeichnis | 210 |
Perception of Environment Properties Relevant for Off-road Navigation | 211 |
1 Introduction | 211 |
2 Controlling Properties of the Environment | 212 |
2.1 Positve Obstacles | 212 |
2.2 Negative Obstacles, Water and Ground | 213 |
3 Representation and Navigation Strategy | 215 |
4 Classification Methods Realized on Ravon | 215 |
5 Conclusion and Future Work | 217 |
Acknowledgements | 217 |
References | 218 |
Aufbau des humanoiden Roboters BART III | 219 |
1 Einleitung | 219 |
2 Mechanischer Aufbau | 220 |
2.1 übersicht | 220 |
2.2 Konstruktion der Gelenke | 221 |
2.3 Komponenten | 221 |
3 Elektrischer Aufbau | 222 |
3.1 Antriebe | 222 |
3.2 Sensoren | 223 |
3.3 SmartPower-Module | 224 |
3.4 Leitrechner | 225 |
4 Erste Gehversuche | 225 |
5 Zusammenfassung | 226 |
6 Literatur | 226 |
Development of Micro UAV Swarms | 227 |
1 Introduction | 227 |
2 Related Work | 227 |
3 Platform | 228 |
3.1 Flight Platform | 229 |
3.2 Ground Control Station | 229 |
4 Towards Autonomy | 231 |
5 Simulation and Evaluation | 232 |
6 Application Scenarios | 233 |
7 Conclusions | 234 |
8 References | 234 |
Die sechsbeinige Laufmaschine LAURON IVc | 235 |
1 Einleitung | 235 |
2 Systemüberblick | 236 |
3 Verhaltensbasierte Steuerung | 238 |
4 Lokalisation und Umweltmodellierung | 238 |
5 Navigation | 239 |
6 Semantische Missionssteuerung | 240 |
7 Zusammenfassung und Ausblick | 241 |
Literaturverzeichnis | 242 |
Dynamic Bayesian Network Library Ein C++ Framework für Berechnungen auf dynamischen Bayes’schen Netzen | 243 |
1 Stand der Technik | 243 |
1.1 Bayes’sche Netze | 243 |
1.2 Andere C++ Bibliotheken zur probabilistischen Inferenz | 244 |
2 Konzept | 245 |
2.1 Struktur der DBNL | 245 |
2.2 Statische Netze | 246 |
2.3 Dynamische Netze | 247 |
2.4 Anfragen | 247 |
2.5 Inferenz | 248 |
3 Evaluation | 249 |
4 Ausblick | 249 |
Danksagung | 250 |
Literaturverzeichnis | 250 |
Modellgetriebene Softwareentwicklung für Robotiksysteme | 251 |
1 Einleitung | 251 |
2 Motivation und Stand der Technik | 252 |
3 SmartSoft | 253 |
4 Der modellgetriebene Ansatz | 254 |
4.1 Platform Independent Model (PIM) | 255 |
4.2 Platform Specific Model (PSM) | 256 |
4.3 Codegenerierung aus dem PSM in die PSI | 256 |
5 Beispiel | 257 |
6 Zusammenfassung und Ausblick | 258 |
Literaturverzeichnis | 258 |
Situation Analysis and Adaptive Risk Assessment for Intersection Safety Systems in Advanced Assisted Driving | 259 |
1 Introduction | 259 |
2 Intersection Modeling | 260 |
3 Behavior Modeling | 260 |
4 Behavior Prediction | 262 |
5 Adaptive Risk Assessment | 264 |
6 Results | 266 |
6.1 System and User Test Results | 267 |
7 Conclusion | 267 |
References | 267 |
Transparente protokollierbare Kommunikation zwischen Funktionen kognitiver Systeme | 269 |
1 Einleitung | 269 |
2 Stand der Technik | 270 |
3 Transparente Kommunikation f¨ur kognitive Systeme | 271 |
4 Effiziente Protokollierung der Kommunikation | 272 |
4.1 Aufzeichnungsmethode | 272 |
4.2 Aufzeichnungsformat | 273 |
4.3 Einsatz einer Aufzeichnung | 274 |
5 Ergebnisse und Anwendungen | 275 |
6 Danksagung | 276 |
Literaturverzeichnis | 276 |
Walking Humanoid Robot Lola An Overview of Hard- and Software | 277 |
1 Introduction | 277 |
2 Mechanical and Electronics System Design | 278 |
2.1 Structural Components | 278 |
2.2 Joint Design | 279 |
2.3 Force/Torque Sensors | 279 |
2.4 Inertial Measurement Unit | 280 |
2.5 Electronics Architecture | 280 |
3 Simulation | 281 |
4 Control Aspects | 282 |
4.1 Trajectory Generation | 282 |
4.2 Stabilizing Control | 283 |
5 Conclusions and Future Work | 284 |
References | 284 |