Workshop on Embedded Optimization and Learning for Robotics and Mechatronics

Monday, October 10, 2022, 10:30 - Tuesday, October 11, 2022, 17:30

Room 1098, Kollegiengebäude I, Platz der Universität 3 , 79098 Freiburg, Germany

 

This workshop is part of the ELO-X Fall School and Workshop.

Invited external lecturers will present state-of-the-art techniques and applications in the field of Embedded Optimization and Learning.

Each talk will be followed by a Q&A session to ensure interactions between the speakers and the participants. 

Locations:

Invited speakers

  • Thomas Brox (University of Freiburg)
  • Taylor Howell (Stanford University, USA / Google Deepmind, UK)
  • Maximilian Manderla, Niels van Duijkeren, Ozan Demir, and Stefan Gering (Bosch Research, Renningen, Germany)
  • Silvia Mastellone (FHNW, Switzerland)
  • Toshiyuki Ohtsuka (Kyoto University, Japan)
  • Rien Quirynen (Mitsubishi Electric Research Laboratories, USA)
  • Davide Scaramuzza (University of Zurich, Switzerland)
  • Yuval Tassa (Google Deepmind, UK)
  • Abhinav Valada (University of Freiburg)
  • Melanie Zeilinger (ETHZ, Switzerland)

Programme

Abstracts of the talks (pdf)

Monday, October 10, 2022

10:30 Welcome Coffee
11:00 Opening of the Workshop
11:15 Real-Time Optimization Algorithms for Nonlinear Model Predictive Control of Nonsmooth Dynamical Systems - Toshiyuki Ohtsuka (Kyoto University, Japan) 
12:00 Lunch
14:00 Paths towards Open World Generalization in Computer Vision - Thomas Brox (University of Freiburg)
14:45 Interactive Learning for (Mobile) Manipulation - Abhinav Valada (University of Freiburg)
15:30 Break
16:00 Embedded Learning, Optimisation and Predictive Control for Autonomous Systems - Rien Quirynen (Mitsubishi Electric Research Laboratories, USA)
16:45 End of Talks
18:10 Transit from seminar location to Waldsee (joint walk or tram)
19:00 Welcome reception at Gaststätte Waldsee (Waldseestrasse 84, 79117 Freiburg)

Tuesday, October 11, 2022

09:00 The unreasonable effectiveness of Predictive Control and the tools to harness it - Yuval Tassa (Google Deepmind, UK)
09:45 Contact-Implicit Model-Predictive Control - Taylor Howell (Stanford University, USA / Google Deepmind, UK)
10:30 Break
11:00 Autonomous, Vision-based Agile Flight - Davide Scaramuzza (University of Zürich, Switzerland)
12:00 MPC@Bosch: from Control of Electric Drives to Robot Navigation (Part 1) - Maximilian Manderla, Niels van Duijkeren, Ozan Demir, and Stefan Gering (Bosch Research, Renningen, Germany)
12:30 Lunch break
14:00 MPC@Bosch: from Control of Electric Drives to Robot Navigation (Part 2)
15:30 Break
16:00 Learning for Predictive Control and Vice Versa - Melanie Zeilinger (ETHZ, Switzerland)
16:45 Optimal Adaptive Droop Design via a Modified Relaxation of the OPF - Silvia Mastellone (FHNW, Switzerland)
17:30 End of Workshop