Ramos’ humanoid robots push boundaries of controls
This semester, the MechSE department welcomed Assistant Professor João Ramos to campus. Ramos earned bachelor’s and master’s degrees in mechanical engineering at Pontifical Catholic University of Rio de Janeiro, and a PhD from MIT, where he continued in a postdoctoral position until his arrival at Illinois.
Ramos’ interest in robotics began during his undergraduate career, where he joined his university’s robotics team and worked on various robotics projects, including battle bots and small humanoid robots. This gave him a solid foundation for robust robotic design, as these applications pushed the robots’ hardware to their limits, causing failure modes not often seen under typical-use conditions for some of their components.
Ramos said he was drawn to Illinois because of its strong history in controls and robotics theory. He believes his hardware experience compliments many of the skillsets of faculty in The Grainger College of Engineering.
In his current research, Ramos hopes to develop high-performance humanoid and other mobile robots that he can push to their limits to perform physically demanding tasks, with special focus on the robot’s hardware. In part to avoid long-standing challenges on the controls side of robotics, Ramos uses human operators to control his robot. The first part of his research focuses on utilizing human motor control intelligence paired with whole-body haptic feedback from the robot to teach the operator how to control the physical interactions between the robot and its environment and understand what it’s experiencing.
“I think that if you can give a person the right way to communicate with the robot, then anyone can learn how to control it. It’s pretty much like learning how to play a new sport: in the beginning, it’s complicated because you have to deal with this new system but when you learn its dynamics you’re able to do something that we didn’t even imagine the robot could do,” explains Ramos.
Once Ramos has achieved the proper communication channel between operator and robot, he aims to challenge the robot’s locomotive capabilities and focus on what he calls dynamic mobile manipulation— which are tasks that involve whole-body coordination to manipulate large or heavy objects, such as lifting and throwing a payload.
“I want my robots to be able to move around and interact with the world in a robust and dynamic way, to complete tasks that you are not able to do if you move around in a slow and conservative way.”
Ramos hopes that someday his research can be used to create a remote avatar of a person that could be used in situations that are too dangerous or impractical to send people, such as fighting fires, handling hazardous materials, or responding to natural and man-made disasters.