Improving Co-Design with Imitation Learning

Scope: Master’s Project/Master’s thesis

Advisor: Dr. Kevin S. Luck (kevin.s.luck@aalto.fi)

Added: 31.August 2022

Topic: Co-Designing Behaviour and Morphology of Robots with Deep Learning has gained more attention in recent years as a complex and difficult problem in the space of reinforcement learning. The vision is to develop AI systems capable of evolving robots and uncovering novel robot designs. This research thesis will investigate further how the use of imitation learning methods [2] and algorithms can be used to improve existing Co-Design algorithms, such as [1]. The aim of this thesis is to develop systems developing both the body and mind of robots. For further information, please contact the supervisor.

Minimum knowledge:

  • Python
  • Has taken basic courses in Machine Learning

Preferred knowledge:

  • Has taken advanced courses in Mahcine Learning or Reinforcement Learning
  • Has taken any robotis course

Thesis student will:

  • Gain a deeper understanding of imitation learning and reinforcement learning
  • Conduct a research project under guidance of the supervisor

Related Literature: