Robotic grasping is a topic well studied since the 80s, however a solution to this problem is still far from being found. Recently, robotic grasping is taking advantage of neuroscientific findings suggesting that the human brain is controlling the hand muscles all together (synergistically) rather than one by one. Both hardware and software implementations focus mostly on the final position of the digits, ignoring the dynamic behaviour of the manipulated object generated by finger approaching time during the actual grasping phase. The student will build on top previous research  and will develop a software grasping controller that will rely on variable speed of fingers while approaching the object. Results will be studied mostly in simulation on the iCub robot .
To successfully complete this project, good C++ programming skills and knowledge of object oriented design principles are required.
 Giuseppe Cotugno, Vishwanathan Mohan, Kaspar Althoefer, Thrishantha Nanayakkara, “Simplifying Grasping Complexity through Generalization of Kinaesthetically Learned Synergies”, in IEEE International Conference on Robotics and Automation ( ICRA), 2014 (PDF available on thrish.org)
 http://wiki.icub.org/wiki/Manual – Video: iCub – Humanoid Platform: http://youtu.be/ZcTwO2dpX8A