Neurobionics Lab

Gray Thomas, PhD

Post-doc

Mechanical Engineering

Website

About

Gray Thomas dreams of grandparents teaching soccer to children over the jet engine wails of their high power exoskeletons. With a self-designed 2012 B.S. in Engineering: Robotics from Olin College of Engineering fresh in his hand, he began his career at the Florida Institute of Human and Machine Cognition to work on an ostrich-inspired passive dynamic running robot. He joined their humanoid robotics team, and helped them win the 2013 DARPA Virtual Robotics Challenge. He worked in Java and added, among other features, a multi-contact controller modification that allowed the simulated robot to enter and exit cars. He got his PhD from the University of Texas at Austin in Mechanical Engineering, working with Professor Luis Sentis. His doctoral work delved deep into the theory of feedback control for uncertain series elastic actuators, which won him NASA Space Technology Research Fellowship NNX15AQ33H, “Controlling Robots with a Spring in Their Step,” and the opportunity to work with the NASA R5 Valkyrie humanoid. During this time, a collaborator (Prashant Rao) and he won the ICRA 2017 best paper award in manipulation for their work on the control limits imposed by series elasticity in robot hands. Yet he drifted away from humanoids when he discovered strong parallels between force control of series elastic actuators and strength amplification control for exoskeletons—and that exoskeletons are also multi-contact robots. Exoskeletons then led him to become a postdoc at the University of Michigan in October 2019, working with Professors Robert Gregg (Electrical and Computer Engineering) and Elliott Rouse (Mechanical Engineering). Gray’s expertise has grown to include frequency domain control theory, system identification for uncertain systems, and linear matrix inequalities for robust control.