Master Thesis

Visual servoing for fine grasping of deformable objects

Work default illustration

Supervisor/s

Information

  • If you are interested in the proposal, please contact with the supervisors.

Description

The Perception and Manipulation Group investigates in service robotics and has two TIAGo mobile manipulation robots and a laboratory simulating an apartment. This project wants to investigate different grasping strategies to allow a robot to manipulate flexible materials, in particular, clothing. First, the traditional approach that relies on hand-eye calibration and classical kinematics will be assessed. Later, using a camera embedded in the robot hand, different visual servoing (2D, 2D1/2...) will be evaluated. The experiments will be performed using a TIAGo mobile manipulation robot and specialized robotic hands. The task will be related to folding or unfolding garments in a human assistive scenario.

Objectives


Develop precise grasping strategies for garments when lying on a planar surface. Develop manipulation actions useful for the folding/unfolding tasks.

Learning oportunities

  • Learn visual servoing methods
  • ROS robotic framework
  • Gather experience in robot use and Gazebo simulator
  • Produce a scientific publication depending on the results

Background

  • Interest in robotics
  • Some experience in programming languages like C, C++ or Python

The work is under the scope of the following projects:

  • CLOTHILDE: Cloth manipulation learning from demonstration (web)