Matei Ciocarlie
Dr. Matei Ciocarlie is a Research Scientist at Willow Garage, focusing on grasping, manipulation and perception. He is currently developing algorithms for grasping known or novel objects using the PR2 robot. His main interest is in reliable performance in unstructured, human environments. To achieve this goal, he is investigating multiple directions, such as object recognition for grasping known objects, recognition of graspable features in sensor data, and passively adaptive hand designs that allow part of the effort of finding a stable grasp to be performed at a mechanical, rather than computational level.
Quick jump: [Contact][Research interests][Internship Opportunities][Bio][Publications]
Contact
Email: matei@willowgarage.com
- Grasping and manipulation
- grasp analysis and quality metrics
- grasp planning using various object models and representations
- real-life sensor data processing for grasping and manipulation
- eigengrasps, postural synergies and low-dimensional hand posture subspaces
- underactuated hands
- passively adaptive and compliant hand mechanisms
- Dynamic simulators and virtual environments
- rigid body dynamics
- contact models, joint and contact constraints, LCP formulations
- collision detection engines
- finite element analysis
- Perception and modeling
- object modeling and recognition
- acquisition and rendering of large 3D models (e.g. buildings)
- using depth cameras for manipulation tasks
We are looking for exceptional interns for a wide range of exciting projects in robotic grasping, manipulation and perception. Positions are available and applications are reviewed year-round. For more details, please see our internships page.
Matei earned his B.S.Eng. from the Polytechnic University of Bucharest, then completed his Ph.D. in Computer Science at Columbia University in New York, where he was a member of the Robotics Group. His thesis, "Low-dimensional Robotic Grasping: Eigengrasp Subspaces and Adaptive Underactuation", focuses on the following central question: how do we reduce the complexity of robotic grasping without compromising versatility? The thesis proposes using low-dimensional hand posture subspaces, that can act as planning spaces (for dexterous robotic hands), interaction spaces (for hand prosthetics) and design spaces (for adaptive underactuated hands). Matei's dissertation was the winner of the 2010 Robotdalen Scientific Award.
Starting in 2003, Matei has also been one of the main developers of the GraspIt! simulator for robotic grasping, designed and implemented in the Robotics Lab at Columbia University. This open-source simulator is a tool for research on autonomous grasping algorithms, robotic hand designs, interaction methods for hand prostheses, and other topics relating to robotic grasping and manipulation.
As a member of the robotics research community, Matei has contributed to the organization of events such as the 2009 and 2010 Mobile Manipulation Challenges, part of the IJCAI and ICRA conferences, served on the Program Committee of conferences such as RSS and SIMPAR, and also served as a reviewer and panelist for the National Science Foundation. Since 2009, Matei is a Research Scientist at Willow Garage, looking to add new capabilities to the PR2 robot and to ROS, the open source Robot Operating System.
View Matei Ciocarlie's publications