Abstract

In this work, we present an efficient 3D object recognition and pose estimation approach for grasping procedures in cluttered and occluded environments. In contrast to common appearance-based approaches, we rely solely on 3D geometry information. Our method is based on a robust geometric descriptor, a hashing technique and an efficient, localized RANSAC-like sampling strategy. We assume that each object is represented by a model consisting of a set of points with corresponding surface normals. Our method simultaneously recognizes multiple model instances and estimates their pose in the scene. A variety of tests shows that the proposed method performs well on noisy, cluttered and unsegmented range scans in which only small parts of the objects are visible. The main procedure of the algorithm has a linear time complexity resulting in a high recognition speed which allows a direct integration of the method into a continuous manipulation task. The experimental validation with a 7-degrees-of-freedom Cartesian impedance controlled robot shows how the method can be used for grasping objects from a complex random stack. This application demonstrates how the integration of computer vision and soft-robotics leads to a robotic system capable of acting in unstructured and occluded environments.

Videos

The object recognition method is integrated into a robotic system which is used to clean up a table full of grocery items piled on top of each other. First, one cleanup at a normal speed is shown, i.e., without the object recognition time having been cut away. Next, two cleanups are shown at a 20x speed-up.

Several table cleanups are shown at a 20x speed-up.

This video shows the robot reaction to strongly misaligned grasps and misaligned placements of an object using different stiffness values and tip velocities with and without collision detection.

Additional Information

If you have any questions/comments/suggestions do not hesitate to contact Chavdar Papazov. This work was done in cooperation with the Human-Friendly Robotics Team which is part of the Department of Robotic Systems at the German Aerospace Center.

Researchers at the following labs are using our 3D geometry matching software:

• Columbia University Robotics Group at Columbia University in New York, USA.
• The Media Computing Group at RWTH Aachen University, Germany.

People

• Chavdar Papazov, M.Sc. (Robotics and Embedded Systems)
• Prof. Dr.-Ing. Darius Burschka (Robotics and Embedded Systems)
• Dr.-Ing. Sami Haddadin (German Aerospace Center)
• Dipl.-Inf. (FH) Sven Parusel (German Aerospace Center)
• Kai Krieger (German Aerospace Center)

Acknowledgement

This work has been partially funded by the European Commission's Seventh Framework Programme as part of the project GRASP.

Related Publications