The success of our research program depends on the close integration of different algorithms and theories from different fields. We require knowledge from fields such as: perception, control, machine learning and human-robot interaction. The development of our semi-autonomous robot will require many different mathematical formalisms that need to be integrated in a coherent theory of semi-autonomous 3rd-Hand robots grounded on solid and sound formal approaches.
Attach:inria.png Δ INRIA is a public science and technology institution established in 1967 dedicated to fundamental and applied research in information and communication science and technology (ICST), distributed over 8 centers with a workforce of 3800 researchers. It also plays a major role in technology transfer by fostering training through research, disseminating scientific and technical information, and participating in international programs. Inria is involved in standardization committees such as the IETF, ISO and the W3C. The researchers at Inria published over 4,800 articles in 2010 and are behind over 270 active patents and 105 start-ups. In 2010, Inria’s budget came to 252.5 million euros, 26% of which represented its own resources.
Attach:mlopes.jpg Δ Manuel Lopes (coordinator) is a permanent researcher at INRIA. Before, he was a lecturer at the University of Plymouth UK and a researcher at Instituto Superior T´ecnico Portugal and VTT Technical Research Centre of Finland. He studied robotics and theoretical computer science at Instituto Superior Tecnico Portugal, and received his Ph.D. degree on robot learning. His research focus is to develop more adaptable and easy to use robots. For this he has made several contributions on: learning dynamical models including interactions with objects, learning from demonstration, adaptive robot control and human-robot interaction. He participated in several international research projects: RobotCub, Mirror, Contact, Handle, First-MM. He has more than 30 papers in international journals and conferences, and gave several invited lectures in international conferences. He has chaired several scientific workshops and summer schools on robot learning and performed senior editorial duties at high-levels and conferences.
Other Members: Thibaut Munzer, Yoan Mollard, Alison Piastri (finantial)
Attach:tu_darmstadt.png Δ The Technische Universität Darmstadt (TU Darmstadt) is one of Germany’s leading technical universities, and also Germany’s first fully autonomous university. TU Darmstadt) has a state funded budget of 270 million Euros (2010, incl. building funds) and currently participates in 68 FP7 projects.
Attach:iasLogo.jpg ΔThe Intelligent Autonomous Systems (IAS) institute of TU Darmstadt) is considered one of the strongest robot learning groups in Europe with expertise ranging from the development of novel machine learning methods (e.g., novel reinforcement learning approaches, policy search, imitation learning, regression approaches, etc.) over autonomous robotics (e.g., robot learning architectures, motor skill representation, acquisition & refinement, grasping, manipulation, nonlinear control, operational space control, robot table tennis, legged locomotion) up to the design of biomimetic motor control systems and brain-robot interfaces. IAS members are well-known researchers both in the machine learning and the robotics community. IAS currently participates in the EU projects GeRT (2010–2013), CompLACS (2011–2015) and CoDyCo (2013–2017).
Attach:jan.jpg Δ Jan Peters is a full professor of Computer Science, head of IAS as well as Senior Research Scientist at the Max Planck Institute for Intelligent Systems. Jan Peters holds four masters degrees in computer science, electrical and mechanical engineering from TUM, USC & FernUni Hagen as well as a Ph.D. from USC. He has received numerous awards, the most recent awards include the 2013 Early Career Award of the IEEE Robotics & Automation Society and the 2013 INNS Young Investigator Award.
Other members: Guilherme Maeda, Rudolf Lioutikov, Olivier Kroemer, Heni Ben-Amor
Founded in 1669, the University of Innsbruck today is a global Top-200 university and the second-largest university in Austria. About 28.000 students are currently enrolled in 15 faculties covering most scientific disciplines. The Institute of Computer Science is part of the Faculty of Mathematics, Informatics and Physics (MIP). UIBK Computer Science was founded in 2001 and currently counts 7 professors. It constitutes a highly productive research environment with a very strong proportion of researchers and doctoral students to undergraduate students, currently generates about 5 Million Euros in external funding per year, and has already produced about 10 start-up companies. The Intelligent and Interactive Systems group at UIBK has experience in visual learning, including closed-loop autonomous learning, and in particular in 2D and 3D object modeling for robot interaction. A strong focus is on incrementally-learnable, probabilistic object models that can incorporate diverse types of scene observations. Most current work is motivated by interactive robotics applications with the aim of equipping robots with the ability to learn about objects and their properties by explorative interaction under various levels of heuristic or human guidance, which are directly brought to bear in the 3rd-Hand project. Object modeling and grasping methods developed by Prof. Piater’s team have been successfully deployed on diverse, high-profile robotic platforms, including the ARMAR-III humanoid robot at the Karlsruhe Institute of Technology, a platform for highly autonomous exploratory learning at the University of Southern Denmark, and the Robot Learning Lab at the Max Planck Institute for Biological Cybernetics.
Attach:jpiater.jpg Δ Justus Piater is a professor of computer science at the University of Innsbruck. He holds a Ph.D. from the University of Massachusetts Amherst, where he held a Fulbright Graduate Student fellowship. After two years at INRIA Rhˆone-Alpes funded by a Marie-Curie Individual fellowship and eight years as a professor at the University of Li`ege, Belgium, including one year as a visiting scientist at the Max Planck Institute for Biological Cybernetics in T¨ubingen, Germany, Prof. Piater moved to Innsbruck in 2010, where he founded the Intelligent and Interactive Systems group at the Institute of Computer Science. He was a principal investigator in the European projects PACO-PLUS and SignSpeak, as well as in four large-scale Belgian/Walloon-Region collaborative projects. He currently is a principal investigator in the FP7 projects Xperience, IntellAct and the upcoming PaCMan project, as well as in the FP7-ECHORD Experiment LearnBiP, in which he holds lead roles in object and scene modeling, grasping, manipulation and affordance learning. Other Members: Özgür Erkent, Dadhichi Shukla
The University of Stuttgart, founded in 1829, is a research-intensive university which has its focus on Engineering and Natural Sciences, with a unique profile that concentrates on closely networking these disciplines with the Humanities and Social Sciences. Currently it has 19; 000 students and more than 60 study programs. The Machine Learning and Robotics Lab, headed by Marc Toussaint, was newly launched in January 2013. The group’s focus is on the use of machine learning methods in robotics and Reinforcement Learning applications, in particular with the goal of robotic systems that learn how to manipulate their environment in a goal-directed manner. Specific areas of expertise include: Relational Reinforcement Learning including exploration, learning and planning in relational domains, multi-agent planning, active learning and exploratio, Stochastic Optimal Control, general robotic motion generation/optimization, and Planning as Inference in general.
Attach:mtoussaint.jpg Δ Marc Toussaint is leading the Machine Learning and Robotics Lab at University Stuttgart since 2013. Before this, he was leading research groups on the same topic at Technical University Berlin (funded by the German excellence programme Emmy Noether) and the Free University Berlin (as an assistant professor). Before he spent two years as a post-doc at the University of Edinburgh with Prof. Chris Williams and Prof. Sethu Vijayakumar, and received his PhD in 2004 at the Ruhr-University Bochum. His recent focus of research is in Machine Learning methods, in particular Bayesian inference methods, and their application in the context of behaviour organization, sequential planning problems (Markov Decision Processes), optimal control and robotics. The methods he developed provide a new perspective on how motion planning and behaviour reasoning on structured representation (for instance distributed or relational) can be realized and inspires new approaches in particular in robotics. His work was awarded best paper at ICMLA 2007 and second best paper at RSS 2012 and UAI 2008. He gave a tutorial on “Machine Learning in Robotics” at RSS 2012, “Stochastic Optimal Control” at ICML 2008 and initiated the workshop series on “Robotics Challenges for Machine Learning” (NIPS 2007, IROS 2008). He is regularly area chair or program committee member at conferences like AIStats, ICRA, UAI, ICML, and IROS, as well as reviewer for the major journals on the field (Journal of AI Research JAIR, International Journal of Robotics Research IJRR, Journal of Machine Learning research JMLR, Autonomous Robots AURO, and many others).
Other Members: Ingo Lütkebohle, Andrea Baisero