de
en
Schliessen
Detailsuche
Bibliotheken
Projekt
Impressum
Datenschutz
zum Inhalt
Detailsuche
Schnellsuche:
OK
Ergebnisliste
Titel
Titel
Inhalt
Inhalt
Seite
Seite
Im Dokument suchen
Vollmer, Anna-Lisa ; Hemion, Nikolas J.: A User Study on Robot Skill Learning Without a Cost Function: Optimization of Dynamic Movement Primitives via Naive User Feedback. In: Frontiers in Robotics and AI. Jg.5. 2018
Inhalt
1.1.1. Machine Learning With Human Teachers
1.1.2. Human-Robot Interaction With Machine Learners
1.2. Contribution and Outline
2. Materials and Methods
2.1. System
2.1.1. Robot
2.1.2. Setup
2.1.3. Ball and Cup
2.1.4. Learning Algorithm
2.2. Optimization—External Camera Setup
2.3. Optimization—Naive Users
2.3.1. Participants
2.3.2. Experimental Setup
2.3.3. Course of the Experiment
3. Experimental Results
3.1. System Performance
3.2. User Teaching Behavior
3.2.1. Questionnaire and Interview
3.2.1.1. Distance from ball to cup
3.2.1.2. Momentum
3.2.1.3. Comparative ratings
3.2.1.4. Spontaneous ratings
3.2.1.5. Individual strategies
3.2.2. Correlation With Ground Truth
3.2.3. Score Data
3.2.3.1. Hits receive always 5 stars
3.2.3.2. Rating on a global scale
3.2.3.3. Rating on a local scale
3.2.3.4. Giving the same score multiple times
4. Discussion
4.1. Usability of/Difficulties With the Current System
4.2. Outlook
Ethics Statement
Author Contributions
Funding
Supplementary Material
References