Sebastian Koch, December 2010
The interest in robots that are able to act in unstructured environments is increasing.
The control of a robot in such an unstructured environment has to rely on sense and perception. Using visual perception provides comprehensive possibilities with regards to robust and safe manipulation of objects. This thesis presents a new approach for visual robot control (visual servoing). Existing approaches are extended in the way that depth and position information of objects is estimated during the movement of the robot. This is done by tracking an object throughout the movement.
As a result, it becomes possible to manipulate objects in the environment of a robot with little previous knowledge. In addition, a force sensor is used for reliable manipulation in unstructured environments.