Motivation
To improve the absolute accuracy of robot arms visual servoing approaches are applied. The goal is to achieve a high 6DoF pose accuracy of a sample mounted at the robot endeffector. A deep reinforcement learning robot model that uses feedback from visual data in order to compensate motion errors of the robot has to be developed.
Topics
- Visual servoing
- Deep reinforcement learning
- Robot simulation
Skills
- Python skills are preferable
- Computer vision skills are preferable
- Deep learning skills are preferable







