Aim of the Workshop
The volume of researches and service and industrial applications using robot learning is continuously increasing. Machine learning is evolving really fast and applications of novel algorithms are often related to robotics. In a similar way, the capability of collecting information to guide the learning process is growing due to the presence of newer and more accurate sensors as well as a better integration between different devices. Nevertheless, there are specific difficulties in making robots learn autonomously new skills. The workshop will gather together different points of view in order to foster the discussion about recent uses of Robot Learning, and provide a view to future directions with an active interaction of researchers and practitioners.
Call for Papers
Papers are solicited in all areas of robot learning. In particular, we are looking for contributions applying robot learning to information coming from novel sensors, supporting the development and testing of new learning algorithms, or exploiting innovative features representation in the learning process. Submissions including experimental results involving actual applications for real robots are particularly welcome. Papers with descriptions of work in progress or preliminary results are also welcome to be submitted as shorter papers to be presented as posters.
Topics include but are not limited to:
- Robot Learning by Demonstration
- Imitation Learning
- Programming by Demonstration
- Reinforcement Learning in Robotics
- Deep Learning and Model Control for Autonomous Robots
- State Estimation, Mapping, and Computer Vision
- Multimodal Perception and Sensor Fusion
- Learning-based Human Robot Interaction
- Interfaces for Learning Natural Language Instructions
- Learning Applications in Manipulation and Field Robotics
- Bioinspired Learning and Control
The content of all submissions has to be original and must not be submitted for consideration in any other conference or journal. Papers must be written in English. Papers will be reviewed by at least two reviewers, who will give detailed comments to insure high quality of the contributions. Papers must be prepared according the following guidelines: Springer authors guidelines The page limit is 12 pages for full paper submissions. We will also accept shorter papers (up to 4 pages) to be presented as posters. All the accepted papers and posters will be included in a DVD with an ISBN number, which will be distributed to the workshop attendees.
Papers have to be submitted electronically in PDF format through the EasyChair Portal.
Special IssueAccepted papers will be invited to submit a revised version of their works to a Special Issue of MDPI Robotics
Paper submission:April 6, 2018 April 30, 2018 May 18, 2018
Author notification:May 6, 2018 May 15, 2018 May 28, 2018
Camera-ready: May 15, 2018May 31, 2018 June 05, 2018
Workshop date: June 11, 2018
All deadlines are 23:59 Pacific Standard Time
Dr. Serena Ivaldi
Title: Learning and control for human-robot collaboration
Abstract: To collaborate efficiently with humans, robots needs to be able to estimate the current activity of the human partner and his/her current goal. To anticipate the partner, robots need the ability to predict the future intended movement of the partner and use this information in their control loop. Robot controllers for interaction need to be “human-aware”, that is consider the human in their control. I will present our current approaches to achieve such goals based on probabilistic models for action recognition and prediction and multi-task constrained whole-body controllers combined with stochastic optimization. I will show our recent results with our humanoid robot iCub and humans equipped with a wearable motion capture system.
Bio: Dr. Serena Ivaldi is a permanent researcher at Inria (France). In 2014 she was Postdoc in the IAS group in TU Darmstadt (Germany); from 2011 to 2014 she was Postdoc at ISIR in UPMC Paris (France); from 2007 to 2011 she was PhD then Fellow in the RBCS group of the Italian Institute of Technology (IIT, Italy). She is currently PI of the European Project AnDy, and was PI of the European Project CoDyCo. She is currently serving as Associate Editor of IEEE Robotics and Automation Letters (RA-L), Intelligent Service Robotics (Springer), and Frontiers in Robotics and AI. Her research is focused on robot learning and control for improving human-robot interaction and collaboration.
Dr. Sylvain Calinon
Title: Robot learning from few demonstrations by exploiting the structure and geometry of data
Abstract: Many human-centered robot applications would benefit from the development of robots that could acquire new movements and skills from human demonstration, and that could reproduce these movements in new situations. From a machine learning perspective, the challenge is to acquire skills from only few interactions with strong generalization demands. It requires the development of intuitive active learning interfaces to acquire meaningful demonstrations, the development of models that can exploit the structure and geometry of the acquired data in an efficient way, and the development of adaptive controllers that can exploit the learned task variations and coordination patterns. The developed models need to serve several purposes (recognition, prediction, generation), and be compatible with different learning strategies (imitation, emulation, exploration).
I will present an approach combining model predictive control control and statistical learning on Riemannian manifolds to achieve such goal. I will illustrate the proposed approach with various applications, including robots that are close to us (human-robot collaboration, robot for dressing assistance), part of us (prosthetic hand control from tactile array data), or far from us (teleoperation of bimanual robot in deep water).
Bio: Dr. Sylvain Calinon is a permanent senior researcher at the Idiap Research Institute (http://idiap.ch). He is also a lecturer at the Ecole Polytechnique Federale de Lausanne (EPFL), and an external collaborator at the Department of Advanced Robotics (ADVR), Italian Institute of Technology (IIT). From 2009 to 2014, he was a Team Leader at ADVR, IIT. From 2007 to 2009, he was a Postdoc at the Learning Algorithms and Systems Laboratory, EPFL, where he obtained his PhD in 2007. He is the author of 100+ publications at the crossroad of robot learning, adaptive control and human-robot interaction, with recognition including Best Paper Awards in the journal of Intelligent Service Robotics (2017) and at IEEE Ro-Man’2007, as well as Best Paper Award Finalist at ICRA’2016, ICIRA’2015, IROS’2013 and Humanoids’2009. He currently serves as Associate Editor in IEEE Transactions on Robotics (T-RO), IEEE Robotics and Automation Letters (RA-L), Intelligent Service Robotics (Springer), and Frontiers in Robotics and AI.
Prof. Jun Miura
Title: Machine Learning Applications for Personal Service Robots
Abstract: There is an increasing demand for mobile service robots which support people in their everyday life. Example services are attending and monitoring, for which reliable person recognition and behavior planning are important. In order to provide personal services, recognizing a specific person among others and person-aware behavior generation are also necessary. In this talk, I will show some of our recent results on these topics to which various machine learning techniques are applied.
Bio: Dr. Jun Miura is a Professor at Department of Computer Science and Engineering, Toyohashi University of Technology (TUT), Japan. He received B.Eng. degree Mechanical Engineering in 1984 and Dr.Eng. in Information Engineering in 1989 from the University of Tokyo. From 1989 to 2007, he had been with Department of Mechanical Engineering, Osaka University, and joined TUT in 2007. From March 1994 to February 1995, he was a Research Scientist at Computer Science Department, Carnegie Mellon University. His research interests include intelligent robots, robot vision, and robotic applications in industry.
Dr. Jean-Baptiste Mouret
Title: Micro-data learning for animal-like adaptation in robots
Abstract: A large part of the impressive results achieved with modern machine learning (in particular, by deep learning) are made possible by the use of very large datasets. However, robots have to face the real world, in which trying something might take seconds, hours, or days. And seeing the consequence of this trial might take much more. In spite of these constraints, robots are expected to adapt like humans or animals, that is, in only a handful of trials: we refer to this challenge as "micro-data learning". In this talk, I will describe our ongoing efforts to design micro-data learning algorithms that allow robots to discover new behaviors by trial-and-error in a few minutes (a dozen of trials), and I will highlight how such algorithms allow robots to recover from unforeseen damage (e.g., a walking robot with a broken leg) without requiring a diagnosis.
Bio: Dr. Jean-Baptiste Mouret is a "research director" (senior scientist) at Inria, the French research institute dedicated to computer science and mathematics. He is currently the principal investigator of an ERC grant (ResiBots – Robots with animal-like resilience, 2015-2020). From 2009 to 2015, he was an assistant professor ("maître de conférences") at the Pierre and Marie Curie University (Paris, France). Overall, J.-B. Mouret conducts researches that intertwine machine learning and evolutionary computation to make robots that can adapt in a few minutes. His work was recently featured on the cover of Nature ("Robots that adapt like animals", Cully et al., 2015) and it received several national and international scientific awards.
Prof. Gerhard Neumann
Title:Information-Geometric Policy Search for Learning Versatile, Reusable Skills
Abstract:In the future, autonomous robots will be used for various applications such as autonomous farming, handling dangerous materials as for example decommissioning nuclear waste, health care or autonomous transportation. For such complex scenarios, it is inevitable that autonomous robots are equipped with sophisticated learning capabilities which enable it to learn from human teachers as well as from self-improvement.
In this talk, I will present our work on information-geometric policy search methods for learning complex motor skills. Our algorithms use information-geometric insights to exploit curvature and path information in order to perform efficient local search at the level of single elemental motions, also called movement primitives. Simultaneously to local search, the algorithms search on a global level by selecting between distinct solutions, allowing us to represent a versatile solution space with high quality solutions. Our algorithms can be used to efficiently learn motor skills, generalize these motions to different situations, learn reactive skills that can react to perturbations and select and learn when to switch between these motions. I will also briefly show how to extend our algorithms to learn from preference-based feedback instead of a numeric reward signal, enabling a human expert to guide the learning agent without the need for manual reward tuning. While I will use dynamic motor games, such as table tennis, as motivation throughout my talk, I will also shortly present how to apply similar methods for robot grasping and manipulation tasks.
Bio: Dr. Gerhard Neumann is a Professor of Robotics & Autonomous Systems in College of Science. Before coming to Lincoln, he has been an Assistant Professor at the TU Darmstadt from September 2014 to October 2016 and head of the Computational Learning for Autonomous Systems (CLAS) group. Before that, he was Post-Doc and Group Leader at the Intelligent Autonomous Systems Group (IAS) also in Darmstadt under the guidance of Prof. Jan Peters. Gerhard obtained his Ph.D. under the supervision of Prof. Wolfgang Mass at the Graz University of Technology. Gerhard already authored 50+ peer reviewed papers, many of them in top ranked machine learning and robotics journals or conferences such as NIPS, ICML, ICRA, IROS, JMLR, Machine Learning and AURO. He is principle investigator for the National Center for Nuclear Robotics (NCNR) in Lincoln which is an EPSRC RAI Hub and also leading 1 Innovate UK project on Tomato Picking. In Darmstadt, he is principle investigator of the EU H2020 project Romans and also already acquired DFG funding. He organized several workshops and is senior program committee for several conferences.
|10:00||Introduction to LAIAR-2018|
Dr. Francesco Nori
|Projects and Applications @ Google DeepMind|
Prof. Jun Miura
|Machine Learning Applications for Personal Service Robots|
S. Tortora, S. Michieletto, E. Menegatti
|Synergy-based Gaussian Mixture Model to anticipate reaching direction identification for robotic applications||bib|
Dr. Serena Ivaldi
|Learning and control for human-robot collaboration|
Dr. Sylvain Calinon
|Robot learning from few demonstrations by exploiting the structure and geometry of data|
P. Van Molle, T. Verbelen, E. De Coninck, C. De Boom, P. Simoens, B. Dhoedt
|Learning to Grasp from a Single Demonstration||bib|
F. Stival, M. Moro, E. Pagello
|A first approach to a taxonomy-based classification framework for hand grasps||bib|
Dr. Jean-Baptiste Mouret
|Micro-data learning for animal-like adaptation in robots|
Prof. Gerhard Neumann
|Information-Geometric Policy Search for Learning Versatile, Reusable Skills|
F. Stival, S. Michieletto, E. Pagello, H. Muller, M. Atzori
|Quantitative hierarchical representation and comparison of hand grasps from electromyography and kinematic data||bib|
M. Nowicki, J. Wietrzykowski, P. Skrzypczynski
|Adopting Learning-based Visual Localization Methods for Indoor Positioning with WiFi Fingerprints||bib|
|Unsupervised online motion prediction in industrial HRC applications, combining collision avoidance and forward-looking dynamic task-selection|
Waiting for the Workshop to start
Dr. Francesco Nori’s Introductory Talk on Google DeepMind research activities
Because of the too many people (more than 50 attendees)
we moved to a larger room
Prof. Jun Miura, Toyohashi University of Technology, Japan
Dr. Stefano Michieletto, IAS-Lab, University of Padua, Italy and Dr. Serena Ivaldi, INRIA Nancy Grand-Est, France
Dr. Sylvain Calinon, Idiap Research Institute, Martigny, Switzerland
Dr. Jean-Baptiste Mouret, INRIA Nancy Grand-Est, France
Prof. Gerhard Neumann, University of Lincoln, United Kingdom
Dr. Francesco Nori, Google DeepMind
Dr. Stefano Michieletto, University of Padova
Prof. Enrico Pagello, University of Padova
Phone: (+49) 7221 304 -0
Fax: (+49) 7221 304 -304
The Workshop has received an Official Sponsorship from:
|Department of Information Engineering (DEI) - University of Padua||Italian Association of Robotics and Automation (SIRI)”||H2020 EU project An.Dy, under Grant 731540|