Project

Generative deep learning for contingent behaviour production on social robots

Code
01N01018
Duration
01 January 2018 → 31 December 2021
Funding
Regional and community funding: Special Research Fund
Research disciplines
  • Natural sciences
    • Artificial intelligence not elsewhere classified
  • Social sciences
    • Cognitive science and intelligent systems not elsewhere classified
    • Social psychology not elsewhere classified
  • Engineering and technology
    • Control systems, robotics and automation not elsewhere classified
    • Design theories and methods not elsewhere classified
    • Mechatronics and robotics not elsewhere classified
    • Computer theory not elsewhere classified
Keywords
Social robots Deep Learning machine learning Human-Robot Interaction
 
Project description

Social robots are robots which interact with people in a natural manner. Autonomously generating contingent verbal and nonverbal behaviour is currently impossible. We use generative Deep Learning to observe video recordings of people and learn which behaviours are appropriate for the robot,
with the aim of having the robot produce matching verbal and non-verbal behaviour.