Photo by Simone Parisi
Eventually, our society will depend on robots as general-purpose helpers that will accompany us throughout our entire lives. This vision requires robots capable of learning motor commands by themselves to manipulate and interact with the world in a way that is useful for us. My work involves the investigation and proposition of learning control algorithms, particularly the representation and transferring of human motor skills into learned and optimized robot policies, and their implementation and validation using real robotic platforms.
News and activities
- March 2019. New paper. Koc, O., Maeda, G., Peters, J. “Optimizing the Execution of Dynamic Robot Movements with Learning Control” was accepted at IEEE Transactions on Robotics (TRO). Manuscript will be added soon.
- February 2019. I am acting as area chair for CoRL 2019.
- November 2018. I am giving an invited talk at Preferred Networks on machine learning perspectives in robot control.
- September 2018. I am serving as Associate Editor for ICRA 2019 (as such, I am not reviewing papers submitted to ICRA 2019).
- August 2018. New paper. “Reinforcement Learning of Phase Oscillators for Fast Adaptation to Moving Targets” was accepted to CoRL 2018! Check the paper and the video.
- May 2018. I am presenting our latest work at the Okinawa Institute of Science and Technology as part of the “AI and Brain Science” and “Adaptive Circuit Shift” Joint Area Meeting in May.
- May 2018. New Frontiers journal paper
- April 2018. New RAS journal paper
- I will be serving as Area Chair for CoRL 2018.
- I am Associate Editor for IROS 2018.
- Two papers accepted at ICRA 2018.
- Presenting our work on active learning at CoRL 2017.
- I am transitioning from TU Darmstadt to ATR.