- Martin Riedmiller, Google DeepMind, UK
- Ruzena Bajcsy, University of California, Berkeley, USA
- Davide Scaramuzza, University of Zürich, Switzerland
Title: Intelligence scores goals - Machine Learning for autonomous robots
Abstract: In 1998, the Brainstormers entered the RoboCup Simulation League with the goal to investigate the usefulness of machine learning methods in intelligent soccer playing agents. I will review our approach of an intelligent control architecture that was applied successfully in four different RoboCup leagues. A particular focus will be put on neural network based reinforcement learning methods, which was our core technique from the very beginning. I will discuss more recent developments of Deep Reinforcement Learning in the context of moving towards general Artificial Intelligence.
Biography: Martin Riedmiller has been heading the Machine Learning Lab at the University of Freiburg since 2009. Recently, he has left the Machine Learning Lab and will now be taking a research position at Google Deepmind. His general research interest is applying machine learning techniques to interesting real world problems.
Presentation Slides: RoboCup_Symposium_2016_Presentation_Martin_Riedmiller.pdf
Title: Framework for Individualized Dynamical Modeling of Human Motion
Abstract: We are developing individualized musculoskeletal models to model movement dynamics based on non-invasive, multi-modal measurements (e.g., motion capture, accelerometry, EMG). The models are aimed to provide information on kinematic constraints and dynamic parameters relating to exerted forces and torques during a movement. Furthermore, we are developing automatic segmentation methods that partition continuous movement into motion primitives.
Biography: Ruzena Bajcsy is a Professor of Electrical Engineering and Computer Sciences and NEC chair holder at the University of California, Berkeley. She received the ACM/Association for the Advancement of Artificial Intelligence Allen Newell Award, the Benjamin Franklin Medal for Computer and Cognitive Sciences, and the IEEE Robotics and Automation Award for her contributions in the field of robotics and automation. She was named as one of the 50 most important women in science by the Discover Magazine.
Presentation Slides: RoboCup_Symposium_2016_Presentation_Ruzena_Bajcsy.pdf
Davide Scaramuzza, University of Zürich, Switzerland
Title: Towards Agile Flight of Vision-controlled Micro Flying Robots: from Active Perception to Event-based Vision
Abstract: Autonomous quadrotors will soon play a major role in search-and-rescue and remote-inspection missions, where a fast response is crucial. Quadrotors have the potential to navigate quickly through unstructured environments, enter and exit buildings through narrow gaps, and fly through collapsed buildings. However, their speed and maneuverability are still far from those of birds. Indeed, agile navigation through unknown, indoor environments poses a number of challenges for robotics research in terms of perception, state estimation, planning, and control. In this talk, I will give an overview of my research activities on visual inertial navigation and active perception with quadrotors, from slow navigation (using standard frame-based cameras) to agile flight (using event-based cameras).
Short Biography: Davide Scaramuzza (1980, Italian) is Professor of Robotics at the University of Zurich, where he does research at the intersection of robotics, computer vision, and neuroscience. He did his PhD in robotics and computer Vision at ETH Zurich and a postdoc at the University of Pennsylvania. From 2009 to 2012, he led the European project “sFly”, which introduced the world’s first autonomous navigation of micro drones in GPS-denied environments using vision as the main sensor modality.
For his research contributions, he was awarded the IEEE Robotics and Automation Early Career Award, a Google Faculty Research Award and SNSF-ERC Starting Grant. His research group is member of one of the three teams selected by DARPA for funding on vision-based agile flight (FLA program). He coauthored the book “Introduction to Autonomous Mobile Robots” (published by MIT Press) and more than 80 papers on robotics and perception. In 2015, he co-founded a startup dedicated to the commercialization of visual-inertial navigation solutions for mobile robots.
Presentation Slides: RoboCup_Symposium_2016_Presentation_Davide_Scaramuzza.pdf
Lab website: http://rpg.ifi.uzh.ch/people_scaramuzza.html