Ziegler, Leon: The attentive robot companion: learning spatial information from observation and verbal interaction. 2015
Inhalt
- List of Tables
- List of Figures
- Glossary
- 1 Introduction
- 1.1 Robot Companions in the Home
- 1.2 From Vista Space to Environmental Space
- 1.3 Research Questions
- 1.4 Scenario & System Foundation
- 1.5 Outline
- 2 Analysis and Statement of the Research Problem
- 2.1 Functional Requirements
- 2.2 The Choice of Scope
- 2.3 Knowledge Representation
- 2.4 Applying a Situation Model in Interaction
- 2.5 Summary: Contribution of this Thesis
- 3 Partitioning the Workspace
- 3.1 The Geometric Foundation
- 3.2 Detecting Roles in Articulated Scenes
- 3.2.1 Benefits of Observed Functional Roles
- 3.2.2 The Challenge in Segmenting a Scene
- 3.2.3 The Articulated Scene Model
- 3.3 Anchoring and Integrating Egocentric Models
- 3.3.1 A Twofold Spatial Representation
- 3.3.2 Registering a Scene Model with the Current View
- 3.3.3 Generating a Valid Model for the Current View
- 3.3.4 Applications Exploiting the Model's Potential
- 3.4 Focusing the Robot's Attention
- 3.4.1 The Interlocutor's Viewing Direction
- 3.4.2 Detecting Interaction Spaces for Manipulation
- 3.4.3 Repositioning for Observation
- 3.5 Evaluation
- 3.6 Summary
- 4 Applying Semantics
- 4.1 Furniture Categorization
- 4.1.1 Implicit Shape Model
- 4.1.2 Ray-Based Hough Space Voting
- 4.1.3 Evaluation
- 4.1.4 Summary & Discussion
- 4.2 Classification of Household Objects
- 4.3 Room Categorization
- 4.4 Summary
- 5 Perception and Communication
- 5.1 Benefits of Combining Perception and Communication
- 5.2 Reference Frame Selection in Human Communication
- 5.3 A Probabilistic Model
- 5.3.1 Visual Analysis
- 5.3.2 Maintaining and Updating the Spatial Network
- 5.3.3 Resolving Conflicts
- 5.3.4 Adaptation to Personal Preferences
- 5.3.5 Application in Human-Robot Interaction
- 5.4 Evaluation
- 5.5 Summary
- 6 Discussion & Conclusion
- Bibliography
- Appendices
- A Situation cases for ASM evaluation
- B Results from Multi-View ASM Evaluation
- C Models for 3D ISM Training
- D Results from Evaluation of Household Object Classification
- E Questionnaire for RSM evaluation