LaViers argues for more human-like robotic motion
Assistant Professor Amy LaViers recently published an article in the prestigious multidisciplinary science journal Nature. Her article discusses changes necessary for robots to achieve more human-like motion.
Robots are capable of doing human tasks efficiently, but they lack the nuanced movement that comes from emotional thinking and complex muscular systems.
“Although the vision for most robots is that they will blend into our lives as caregivers or guides, they struggle to interact with humans or to operate in changing environments,” LaViers wrote. “Roboticists need to learn more about how humans move and the ways in which we physically outpace our artificial counterparts.”
People can move three-dimensionally at any given time. Our bodies can stretch and flex and experience varying ranges of motion all at once. In contrast, robots are stiff and limited, comprised of simple mechanics.
“Algorithms to control robots should include individual, varied or unexpected responses,” LaViers wrote. She believes that the latest generation of soft robots—those constructed from flexible or elastic materials—will allow for increased range of motion and utility.
LaViers first became interested in robotics as a student in the dance department while studying mechanical engineering at Princeton University. “I realized that the quantitative tools in my mechanical engineering courses, particularly those in dynamics and control, were analyzing the same thing as choreographers: the pattern of bodies moving through space and time,” she said.
In her research, LaViers works to identify, generate, and understand patterns of motion in the human body. Her group is currently generating styles of bipedal walking using an optimal control framework, as well as studying how simple artificial systems imitate human movement.
“These two projects highlight the intertwined roles that generating robotic motion and understanding human perception play in my group’s work,” LaViers said.
In order for robotic motion to continue to evolve, LaViers argues that roboticists need to actually stand up and try out the movements they are designing.
“I hope to see more involvement from artists, animators, and experts from our colleagues in the humanities,” LaViers said. “Robotics is an applied field that needs interdisciplinary researchers to solve the very real challenge of creating tools that serve humans well.”
LaViers joined the MechSE faculty in 2015. She was a recipient of the DARPA Young Faculty Award in 2015 and the DARPA Director’s Fellowship in 2017.
Photo at top: LaViers, a roboticist, using embodied practice to inform her understanding of the role of the spine in generating stability for human bodies. Photo credit Natalie Fiol.