PhD Defense: Reinforcement Learning from Simulation for Real Robot Manipulation

27 January 2022

Abstract: A broad application of robots in both industry and everyday life needs an agile and stable control strategy when faced with variable circumstance. Among all the everyday tasks, object pushing and grasping are two basic manipulations. The flexibility and scalability of the neural networks promote the end-to-end learning methods, which can do the perception and control together. The main content of this thesis is the study of pushing and dexterous grasping with deep reinforcement learning.

To participate in the PhD defense, please contact the Informatics Study Office via email. The Zoom link will then be mailed to you shortly before the meeting.

Time: 16:00