Ehsan Noohi












Human – Robot Cooperative Manipulation



While human-robot interaction has been receiving a lot of attention recently, the physical interaction between the human and the robot has remained one of the most challenging parts of this interaction.

My work is focused on human-robot cooperative manipulation. There exists interesting works that propose algorithms for a leader robot or a follower one. The goal of my research, however, is to enable the robot to perform cooperatively in such a way that the human partner feels a natural interaction, similar to a human-human cooperative manipulation.




Related Journal Paper:


[1] E. Noohi, M. Zefran and James L Patton, “A Model for Human–Human Collaborative Object Manipulation and Its Application to Human–Robot Interaction”. IEEE Transaction on Robotics, Vol. 32, Issue 4, August 2016, pp. 880-896.



Related Conference Papers:


[3] E. Noohi and M. Zefran, “Modeling the Interaction Force During a Haptically-Coupled Cooperative Manipulation”, IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), Columbia University, New York, NY, August 26 - 31, 2016.


[2] E. Noohi, S. Parastegari and M. Zefran, “Computational Model for Dyadic and Bimanual Reaching Movements”, IEEE World Haptic Conference (WHC’15), Northwestern University, Evanston, IL, June 22-25, 2015.


[1] E. Noohi and M. Zefran, “Quantitative Measures of Cooperation for a Dyadic Physical Interaction Task,” in International Conference on Humanoid Robots (Humanoids’14), Madrid, Spain, Nov. 18-20, 2014.






Wheeled-tip Robotic Hand



I proposed a new robotic hand structure, called “Wheeled-tip”. A wheeled-tip robotic hand is constructed by adding active wheels (balls) to an ordinary robotic hand (manipulator), at the tip of robot fingers. The wheels can grasp objects and manipulate them by rolling over their surfaces, while conserving the grasp stability.

I studied the kinematic modeling and motion planning of a planar (2-D) wheeled-tip hand. In this case, the robot manipulates arbitrarily-shaped objects by rolling over their boundaries and their corners. These videos demonstrate the extended workspace of wheeled-tip robots compared to ordinary robotic hands.

In contrast to the planar wheeled-tip robots, satisfying the non-holonomic constraints and grasp stability at the same time is a great challenge in a 3-D wheeled-tip robotic hand. Thus, extending the motion planning and control of the robot to a 3-D setup is yet to be done as the future work.





Related Journal Papers:


[2] E. Noohi, H. Moradi, S. Parastegari and M. Nili Ahmadabadi, “Object Manipulation Using Unlimited Rolling Contacts: 2D Kinematic Modeling and Motion Planning”. IEEE Transaction on Robotics, Vol. 31, Issue 3, 2015.


[1] E. Noohi, H. Moradi, N. Noori and M. Nili Ahmadabadi, “Manipulation of Polygonal Objects with Two Wheeled-tip Fingers: Planning in Presence of Contact Position Error”. Elsevier Robotics and Autonomous Systems, Vol. 59, Issue 1, January 2011, pp. 44-55.



Related Conference Papers:


[3] S. Parastegari, M. Nili Ahmadabadi, E. Noohi and H. Moradi, “Wheeled-tip Object Manipulation: Modeling and Motion Planning of Throwing an Object”, In Proceedings of IEEE International Conference on Robotics and Biomimetics (ROBIO’12), pages 1800-1805, Dec. 11-14 2012.


[2] N. Noori, E. Noohi, H. Moradi, A.H. Bakhtiary, and M. Nili Ahmadabadi. “A Probabilistic Roadmap Based Planning Algorithm for Wheeled-tip Robots Manipulating Polygonal Objects”. In Proceedings of ASME/IEEE International Conference on Advanced Intelligent Mechatronics (AIM’09), pages 1040–1046, Singapore, July 14-17, 2009.


[1] E. Noohi, H. Moradi, and M. Nili Ahmadabadi. “Manipulation Using Wheeled Tips: Benefits and Challenges”. In Proceedings of 39th International Symposium on Robotics (ISR’08), pages 442–447, Seoul, South Korea, October 15-17, 2008.






Wheel-based Climbing Robot



Many climbing robots are designed and adapted for the specific surface that they are supposed to climb. Our pole climbing robot is a good example of such robots. Its structure allows the robot to hug the pole (regardless of its diameter) and move along it. The motivating application for this project was automation of cleaning process of the highway light poles. Other applications for this wheeled-based structure includes inspection of the suspension bridge cables.

In this project, first, I introduced a non-holonomic model for the robot. Then, based on this model, I proposed an algorithm that enabled the robot to perform any desired maneuver on a pole by combining a set of the primitive movements. This video shows two basic movements, implemented on the robot.





Related Journal Paper:


[1] E. Noohi, S. Mahdavi, A. Baghani and M. Nili Ahmadabadi, “Wheel-Based Climbing Robot: Modeling and Control”. Advanced Robotics, Vol. 24, No. 8-9, May 2010, pp. 1313-1343.



Related Conference Papers:


[2] S. Mahdavi, E. Noohi and M. Nili Ahmadabadi, “Basic Movement of a Nonholonomic Wheelbased Pole Climbing Robot”. In Proceedings of IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM’07), ETH Zurich, Switzerland, Sept 4–7 2007.


[1] S. Mahdavi, E. Noohi, and M. Nili Ahmadabadi, “Path Planning of the Nonholonomic Pole Climbing Robot UT-PCR”. In Proceedings of IEEE International Conference on Robotics and Biomimetics (ROBIO’06), pages 1517-1522, Kunming, China, Dec. 17-20 2006.