DARPA research uses human movement concepts to improve robotic control systems

1/19/2016 MechSE Communications

  Assistant Professor Amy LaViers was awarded a 2016 DARPA Young Faculty Award for her research

Written by MechSE Communications

 
Assistant Professor Amy LaViers was awarded a 2016 DARPA Young Faculty Award for her research project titled, “Choreography of Embodied, Platform-invariant Motion Primitives.” 
 
DARPA, the Defense Advanced Research Projects Agency, awards grants for research and technology that align with their mission of improving national security. The award identifies young researchers whose work shows promise for DARPA goals. In addition to funding, researchers are invited to attend various defense-related site visits to better understand how their work can impact national security over their career.
 
LaViers’ robotics research has a strong connection to dance, combining concepts from Laban/Bartenieff Movement Studies (LBMS) to contribute to the development of a supervisory control system with knobs that are aligned with the experience of human movement.
 
“The vision of robots as agents that can keep humans out of the way of dangerous work while still leveraging the decision-making power of human operators is limited by the operators’ ability to easily prescribe complex, high-level behavior to the sophisticated platforms that already exist,” said LaViers.
 
Examples of this include rescue workers commanding humanoids for disaster relief efforts, soldiers using drones to investigate a suspected danger, or a bomb squad attempting to dismantle an explosive device with a mobile manipulator arm. Results of the research could also have applications in advanced manufacturing, collaborative robots, and robotic prosthetics. 
 
Her work will use LBMS to develop a control system for the robot’s operator that offers a variety of movements. A spatial mesh, designed off the form of the human body, and other high-level movement ideas that can apply across robotic platforms, will provide spatial orientation and other movement instructions to the platform. 
 
“This approach to start with embodied, qualitative concepts is risky, but it has a lot of promise to give operators an unparalleled ease of control over platform behavior,” she said.
 
LaViers directs the Robotics, Automation, and Dance (RAD) Lab, where she develops algorithms for automation inspired by movement and dance theory. She has worked in the area of advanced manufacturing through an industry-university consortium and has forged interdisciplinary ties with university dance programs and the Laban/Bartenieff Institute for Movement Studies, where she is pursuing a Certification in Movement Analysis. 
 
Prior to Illinois, she was an assistant professor of systems and information engineering at the University of Virginia. She earned a PhD in 2013 in electrical and computer engineering at Georgia Tech.
 

Share this story

This story was published January 19, 2016.