Neuromorphic motor control

Responsible person: Elisabetta Chicca

A key feature of an intelligent artificial agent is the capability of acting upon the environment through actuators. Practical solutions to this need are already under investigation within the CITEC research line Motion Intelligence. Contrastingly, within the neuromorphic engineering research community, a lot of efforts have been devoted to sensing and computing, but only a few groups have been focusing on motor control. We believe neuromorphic systems have high potential for modeling how the nervous system produces motion. Furthermore, long term developments of neuromorphic approaches to motor control surely include the development low-power compact control modules for robotics, as well as better understanding of biological feedback and feed-forward control signals, with interesting applications to neuroprosthetics. With this project, we aim to realize efficient motor control by implementing low-power event-based local control modules which can be modulated by central control systems, i.e., local motor controllers using Pulse Frequency Modulation and DC motors.