Neuronal Computing for Competitive Engineered Systems and Autonomous Robotics

Understanding and applying the computational principles behind how brains turn perception into behavior is one of the most challenging research questions for the upcoming decades. My research group investigates theory, models, and applied technical implementations of distributed neuronal information processing, to (a) discover key principles by which large networks of neurons operate and to (b) implement those in engineered systems to enhance their real-world and real-time performance. In my presentation I will highlight the fundamentally different style of information processing seen in the brain from that seen in our computers, and show opportunities to bring the brain’s many computing advantages into today’s engineered systems. One line of work involves retinal-inspired event-based vision processing for high-speed closed-loop control systems or wearable devices. A second line addresses distributed large-scale neuromorphic networks for sensory perception, real-time decoding of brain recorded activity, and the generation of motor output, as needed e.g. to control neuro prostheses. I will demonstrate using brain-like approaches to information processing for technical and biomedical data to achieve levels of system performance that have previously been unattainable, whether in improved speed, improved power efficiency, or improved cognitive abilities.