Advancing robotics, a sweet spot for neuromorphic computing, the work at the Computational Brain (ComBra) Lab has recently received the attention of the media, including EngadgetHPCwireBit-TechCBRonlineRobot ReportRobotics Business Review and Unite AI. The news is spread in several countries including ChinaGermanyItalyBrazilUK and Korea, echoing the results to be presented at the 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2019), this November.

The paper describes the first end-to-end neuromorphic algorithm that is inspired by the brain's network to solve a foundational problem in real-time control of robots. The work hints that neuromorphic computing can scale to address real-world problems, by comparing to highly developed methods while offering unparalleled energy efficiency (here, two orders of magnitude less energy is needed, compared to a mainstream method for simultaneous localization and mapping of a mobile robot.)

The other two authors of the paper are:

1) Guangzhi Tang, a PhD student at ComBra Lab who has completed his MSc Thesis in the same Lab; and

2) Arpit Shah, a MSc student who conducted his Thesis at ComBra Lab. Aprit was awarded the Publications Award by the CS Department.