The Italian Institute of Technology (IIT) was created with the objective of promoting Italy’s technological development and higher education in science and technology. IIT is making an effort to implement its scientific programme in this context, with the specific goal of integration between basic scientific research and the development of technical applications.
The research is carried out in highly innovative scientific fields, which represent the most advanced frontiers in modern technology; their contents are widely applicable in multiple areas, from medicine to industry, from computer science to robotics, to life sciences and nanobiotechnology.
In the PANDORA project, IIT participates with its Department of Advanced Robotics, which comprises more than 65 staff at all research levels from PhD (25) to Full Professors (5). Research activities in the department concentrate on an innovative, often biologically inspired, multidisciplinary approach to robot design and control, and the development of novel robotic components and technologies. It combines activities from both the hard (mechanical/ electrical design and fabrication, sensor systems, actuation development etc.) and soft (control, computer software, human factors etc) systems areas of robotics.
IIT will lead WP3 (Work Package 3 – Skill Learning) in the PANDORA project. This builds on a portfolio of current work, including recent EU funded research projects such as RobotCub, Viactors, Octopus and Amarsi. IIT will further work in WP5 to transition the skill learning techniques to Icteneu AUV and to the iterative development of the chain cleaning and valve turning demonstrations.
The Department of Advanced Robotics has an extensive expertise in Advanced Actuation and Energy/Power Systems, Interfaces and Interaction Technologies, Sensing, Intelligence and Control, and Structures, Materials and Mechanisms, where the merging of diverse technologies including both traditional (hard-bodied) robots with the new generations of hybrid systems combining engineering principles such as speed, robustness, accuracy, and endurance with biologically inspired concepts to provide robots with enriched capabilities.
The Learning and Interaction Group of the Advanced Robotics Dept, which will mainly be involved in PANDORA, has extensive experience in robot learning by imitation and reinforcement, through the use of probabilistic models to extract task constraints from skill demonstrations, with various applications ranging from torque-controlled manipulators to humanoids.
For more information, please visit:
Learning and Interaction Group web homepage