The H2020 project "EXCELLABUST - Excelling LABUST in marine robotics", Centre of Excellence ACROSS CoE, Centre of Research Excellence DATACROSS and IEEE Croatia Section, Robotics and Automation Chapter will organize the lecture:
"Embodiment Approaches to Sensorimotor Robot Policy Training"
which will be held by Prof. Mårten Björkman, KTH Royal Institute of Technology, Stockholm, Sweden
The lecture will take place on Tuesday, 17th April 2018, starting at 09:30 in a seminar room at ZARI, building C, 9th floor.
You can find more about the lecturer and the seminar in the detailed news content.
For robots to become truly autonomous and be able to overcome changes in either the working environment or its own embodiment, it needs to an ability to self-learn tasks and gradually adapt to changes. In our work, we have tried to move the human designer as far as possible from the learning process and allow the robot to create its own model of the world and with minimal prior information learn to exploit its embodiment through exploration, often in collaboration with a human partner. We will present some work in this direction from simple reactive behaviors learned from the ground up to more complex predictive behaviors learned in stages that consist of both simulations and real robot experiments. Using combinations of Gaussian process models, deep neural networks, and reinforcement learning, the emphasis is placed on data efficiency and adaptivity, allowing the robot to learn in a data-driven manner with a minimal number of trials.
Mårten Björkman is an Associate Professor at the School of Computer Science and Communication at the Royal Institute of Technology, KTH. He received an MSc degree in computer science and engineering from Lund University in 1994 and a PhD degree in computer vision and robotics from KTH in 2002. He has been actively contributing to research to the EC funded projects CogVis, PACO-PLUS, eSMCs, and socSMCs. His research interests are a human-robot collaborative system, real-time object detection, and segmentation, and data driven mobile manipulation.