- Articulated object handling(drawer and door)[2]
Problem description:
Object recognition and pose estimation plays an important role in many robotic fields. Real-time depth images are widely used for solving these problems because of the substantial information contents [1], especially with the Kinect sensor which can provide well calibrated RGB and depth information at 30Hz. Furthermore, many objects are not rigid since they have moving parts such as drawers, doors and scissor. Understanding the spatial movements of parts of such objects is essential for service robots to allow them to plan relevant actions such as door-opening trajectories. A fundamental challenge is how to recognize these kinds of articulated objects and retrieve the kinematic joint state using the single depth image.
As the goal of this thesis, articulated object models with kinematic joint(such as rigid and/or prismatic) should be learned based on spatial change in depth images [2] [3]. Based on the compilation with 3D geometric features [1], the articulated object should be recognized and its pose should be estimated, and finally the state of the articulated object’s kinematic joint should be retrieved.
Tasks:
- Development and implementation of an algorithm for articulated object model learning using the spatial change of depth image;
- 3D geometric features integration for modeling and recognition in real scene;
- Optimization for real-time application and experimental evaluation.
Bibliography:
[1] Rusu,Radu Bogdan, Bradski Gary, Thibaux Romain, and Hsu John, ”Fast 3D Recognition and Pose Using the Viewpoint Feature Histogram”, Proceedings of the 23rd IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 10/2010,Taipei, Taiwan.
[2] Jürgen Sturm , Kurt Konolige , Cyrill Stachniss , Wolfram Burgard , ”Vision-based Detection for Learning Articulation Models of Cabinet Doors and Drawers in Household Environments”, In Proc. of the IEEE International Conference on Robotics and Automation (ICRA), 2010.
[3] Dov Katz, Yuri Pyuro, Oliver Brock, ”Learning to Manipulate Articulated Objects in Unstructured Environments Using a Grounded Relational Representation”, Robotics: Science and Systems, 2008.
Prerequisites:
- Good programming skills in C++
Helpful but not required:
Contact: