Chemnitz University of Technology

The European Cross-Border University (ACROSS) is happy to announce the “International Summer School on Neurorobotics”, taking place at the Chemnitz University of Technology, from the 26th to the 30th of September 2022.  We will accept 25
students in situ. 15 students will be provided with a scholarship. The priority for scholarships will be for
students of the ACROSS universities.



This event is intended for all students interested in the emerging science at the intersection of computational neuroscience, artificial intelligence and robotics. In lectures and practical courses, we will cover all aspects relevant to brain-inspired robotics with a focus on latest simulation platforms. Lectures will be accompanied by talks from external speakers. In practical sessions we will program neurorobotics experiments to study the interaction between brain, body, and environment in closed perception-action loops. This summer school will be on-site and offers an immersive and highly interdisciplinary learning experience. All attendees will have the opportunity to present their work in a poster session.



EBRAINS is a new digital research infrastructure, created by the EU-funded Human Brain Project, that gathers an extensive range of data and tools for brain-related research. EBRAINS will capitalize on the work performed by the Human Brain Project teams in digital neuroscience, brain medicine, and brain-inspired technology and will take it to the next level. Read more and discover EBRAINS in our introduction document.



MiRo-E is the perfect robotic device for any level of education
MiRo-E’s friendly pet-like appearance and qualities stand out and appeal to everyone. Providing a sophisticated and technically advanced platform that can be utilised in multiple departments. Introduce new students into Robotics and CS, broaden research topics potential, and provide a brilliant outreach and public engagement tool.


Talk: Deep Spiking Reinforcement Learning

Using deep reinforcement learning policies that are trained in simulation on real robotic platforms requires fine-tuning due to discrepancies between simulated and real environments. Multiple methods like domain randomization and system identification have been suggested to overcome this problem. However, sim-to-real transfer remains an open problem in robotics and deep reinforcement learning. Here, we present a spiking neural network (SNN) alternative for dealing with the sim-to-real problem. In particular, we train SNNs with backpropagation using surrogate gradients and the (Deep Q-Network) DQN algorithm to solve two classical control reinforcement learning tasks. The performance of the trained DQNs degrades when evaluated on randomized versions of the environments used during training. To compensate for the drop in performance, we apply the biologically plausible reward-modulated spike timing dependent plasticity (r-STDP) learning rule. Our results show that r-STDP can be successfully utilized to restore the network's ability to solve the task. Furthermore, since r-STDP can be directly implemented on neuromorphic hardware, we believe it provides a promising neuromorphic solution to the sim-to-real problem.

Talk 2: Embodied Learning with the Neurorobotics Platform

The Neurorobotics Platform (NRP) is a cloud based simulation platform developed in the Human Brain Project that promotes research of Embodied (Artificial) Intelligence. The platform interconnects artificial and spiking neural networks with physics simulations of robots and musculoskeletal bodies. In this presentation the platform architecture and tools are introduced. Finally, example experiments are showcased including an embodied million neuron rodent brain deployed on a High Performance Computing cluster.


Here to Help Students Grow



Chemnitz University of Technology, Germany.

Florian Röhrbein studied computer science at TU Munich, then moved to LMU Munich and received his doctorate in 2005. A four-year industrial stay at the Honda Research Institute Europe was followed by a postdoctoral period at the Albert Einstein College of Medicine in New York. At University of Bremen he habilitated and went back Munich where he worked as managing director for neurorobotics in the Human Brain Project and became chief editor of "Frontiers in Neurorobotics". From 2018 to 2020 he again worked in industry and developed a company-wide AI strategy for a world market leader near Stuttgart. He founded several startups and became CTO of aivious GmbH in Munich and also of oxolo GmbH in Hamburg before he was appointed full professor at Chemnitz University of Technology.


Bialystok University of Technology, Poland

Michal Kuciej studied mechanical engineering at TU Bialystok, and in 2008 he defended his doctoral dissertation in the scientific discipline of thermomechanics. He habilitated in 2012, while in 2020 the President of the Republic of Poland awarded him the title of professor of engineering and technical sciences. He is the topic editor in "Advances in mechanical engineering" and vice editor in chief in "Acta mechanica et automatica". From 2020 he has been the director of the Doctoral School of TU Bialystok and the head of the Department of Mechanics and Applied Computer Science. He is also a coordinator (on behalf of TU Bialystok), of the "Hub of Talents 2" – project supporting the creation of startups in IT technologies and mechanical engineering. His main research interests are analytical and numerical modeling of frictional heating in brake systems.



Begin Today


 Participants will be notified in one week and registration will be confirmed immediately after. If we receive more registrations than the number of available slots, selection will be made on the basis of applicants’ CV and motivation letter.

Image by Maximilian Scheffler


send an email at with the following attachments:

  • Brief motivation letter (max. 1000 characters)

  • CV