MechSE assistant professor Joao Ramos has been awarded a CAREER grant from the National Science Foundation.
The title of his project is “Telecomotion of humanoid robots via bilateral feedback and intent prediction.”
Ramos’s work targets understanding the dynamics that arise in a novel bilateral teleoperation
framework to control the locomotion of a humanoid robot. Direct control of locomotion via teleoperation, called telelocomotion, is the fundamental step towards creating robot avatars with human-level motor skills.
“Bilateral telelocomotion is an understudied area that should prove to be a rich source of technical challenges,” Ramos said. “This problem is unique because, unlike conventional teleoperation of robot arms, the humanoid robot must dynamically regulate balance while performing any task.”
Teleoperated robots can transport the knowledge and skills of human experts to hazardous locations. However, workers still cannot perform meaningful physical tasks using existing telepresence technology. For instance, firefighters routinely realize demanding physical labor, such as manipulating an active hose while walking over unstructured terrain. This task can be performed via direct teleoperation if the operator can directly control the robot's entire body while experiencing the machine’s sense of balance and the interaction forces with the hose. But the task can only be performed reliably if the robot can predict and adapt to human motion intent.
In other words, the teleoperation of coordinated locomotion and manipulation requires bilateral motor and mind synergy between human and robot.
In this framework, the operator directly commands the robot gait and receives force feedback at a whole-body level, instead through a handheld joystick. The whole-body force feedback couples the locomotion dynamics of the operator and that of the robot. Navigation on uneven terrain is enabled with a new shared-autonomy framework in which the robot adapts to human locomotion intent on the fly.
Ramos joined the MechSE faculty in 2019 after serving a one-year postdoctoral appointment at MIT, where he had earned his PhD in mechanical engineering in 2018.