de
en
Schliessen
Detailsuche
Bibliotheken
Projekt
Impressum
Datenschutz
zum Inhalt
Detailsuche
Schnellsuche:
OK
Ergebnisliste
Titel
Titel
Inhalt
Inhalt
Seite
Seite
Im Dokument suchen
Emmerich, Christian: Efficient and intuitive teaching of redundant robots in task and configuration space. 2016
Inhalt
Acknowledgment
Abstract
List of Figures
Introduction
Motivation: Flexible Production in Industry 4.0
Problem Statement: Teaching of Redundant Robots is Difficult
FlexIRob@Harting: Kinesthetic Teaching in Confined Spaces is Difficult
Contribution and Goal of Thesis
Outline
Kinesthetic Teaching of Redundant Robots in Confined Spaces
Inverse Kinematics for Redundant Robots
Kinesthetic Teaching as an Intuitive Programming Interface
Demonstrating in Configuration Space
Demonstrating in Task Space
Proposed Method
Kinesthetic Teaching of Redundancy Resolutions
Hierarchical Control
Assisted Programming in Task Space
Implementation
Discussion and Related Work
Learning, Encoding and Generalizing Redundancy Resolutions
Interaction Model
Learning Redundancy Resolutions with Neural Networks
Global Learning with Random Projections
Local Learning with Local Linear Maps (LLM)
Generalization Capabilities
Task Space Accuracy vs. Null-space Constraints
Number of Training Areas and the Reachable Workspace
Model Selection and the Number of Training Samples
Discussion of Results
Teaching Redundancy Resolutions
Interaction Model
Teaching Implicit Constraints
Constraints Modeling in Joint Space
Implicit Scene Model in Task Space
FlexIRob@Harting: Teaching Redundancy Resolutions
General User Experience
Illustration of Training Data for Different User Experiences
Analysis of the Participants Teaching Success
Quality of the Provided Training Data
Discussion and Related Work
Assisted Programming in Task Space
Interaction Model
FlexIRob@Harting: Assisted Programming
Analysis of the Teaching Experience
Analysis of the Teaching Success
Discussion of Results
Assistance Blending for Teaching Redundancy Resolutions
Interaction Model
Online Sequential Extreme Learning Machine
Learning a Confidence Model from Training Data
Discussion and Related Work
Autonomous Path Planning in Confined Spaces
Interaction Model
Review: Instantaneous Topological Maps (ITM)
Hybrid ITM with Connectivity Bootstrapping
Validation of Edges using a Joint Space Criterion
Bootstrapping Data for Improved Connectivity
Results
Discussion and Related Work
Incremental Teaching of a Simulated 9-DoF manipulator
Interaction Model
Experimental Setup
Results
Discussion
Conclusion
FlexIRob - A Flexible Interactive Robot Prototype
System Overview
The KUKA Lightweight Robot IV
Implementation of Interaction Model
Software Abstractions for Learning From Demonstrations
Middleware
FlexIRob@Harting - A User Study on pHRI
Study Design and Interaction Model
Warm-up phase
Configuration Phase
Wire-loop game
Subjective Assessment of Interaction
Questionnaire Design
Data Assessment and Analysis
Clustering of the Participants' Selected Redundancy Resolutions
Related References by the Author
References