Master Thesis presentation by Rajath Sridhara

Examiner: Yasemin Bekiroglu

Opponents: Sharath Chandra Ashok Kumar and Emil Erikmats

Abstract:
Visual Servoing is a control approach that uses visual data to control the motion of a robot. The vision data may be acquired from a camera that is mounted directly on a robot manipulator or on a mobile robot. This thesis aims to improve upon recent works that make use of visual servoing and extend them to mobile manipulation tasks. A mobile manipulator was modelled in Coppeliasim and combined with two different Visual Servoing tasks - Image based Visual Servoing and Point cloud Based Visual servoing. Experiments were carried out for different reaching tasks and the resulting trajectory was analysed for accuracy and the ability of the robot to complete the task. The proposed approach can be further extended to track moving objects in unstructured surroundings.

Welcome!

Rajath and Yasemin