Place: Technická 2, Praha 6 - lecture hall T2:D3-209 (2nd floor)
The lecture will be streamed online on Youtube for the general public.
Inspiration from the human brain translated into algorithms for self-improving robots. This could also be a summary of one of the topics addressed by the humanoid robotics group at the Department of Cybernetics of the Faculty of Electrical Engineering. Matěj Hoffmann is primarily concerned with the issue of a unique "electronic skin", with several thousand sensors picking up sensations from the surface of a humanoid robotic body. Algorithms inspired by the functioning of the human brain simulate the processes by which a small child learns about its surroundings and its own body. This allows the robot to experience its environment in a similar way to humans. The process of robotic learning is the same as for a child in its first year of life.
Thanks to new approaches, the adaptability of humanoid robots may become more human-like, with machines being able to react to changing conditions - for example, to notice someone entering their space or to be aware of touch. Such humanisation of robots may also result in them being able to move independently and safely in human-occupied environments. Another advantage of robotics studying such processes is the ability to simulate changes or injuries in humanoids that, for obvious reasons, cannot be practiced on human models.