This workgroup is about robotics with accelerated neuromorphic hardware. In particular, we will provide multiple BrainScaleS-2 systems that feature emulated neurons with dynamics 1000 times faster than those found in biology.
All systems are equipped with multiple digital and analog real-time interfaces for robotic applications. We can inject and extract spikes and generate current stimuli directly onto the physical neurons, corresponding networks can be expressed in PyNN or pytorch.
As a starting point, we will bring two results from previous years at CCNW: a controller for brushless electric motors and a microphone array for sound source localization. During the workgroup, we will try to find new applications for accelerated robotics and all participants have the possibility to connect the system to their sensors and actuators. We will specifically try to explore self-supervised learning strategies for robotic applications on neuromorphic hardware and combine all our expertise in that area to build a working demonstrator.
If you are interested, please consider joining the BrainScaleS-2 umbrella workgroup for an introduction to the platform.
Go to group wiki Go to wiki users Info
Day | Time | Location |
---|---|---|
Mon, 29.04.2024 | 16:30 - 17:00 | Disco |