Meet Associate Professor of Biomedical Engineering Hao Su, head of Tandon’s Biomechatronics and Intelligent Robotics Lab
Much of Hao Su’s work centers around mechatronics design, learning, and control of soft robotics for surgery and rehabilitation applications. He recently received widespread attention in major news outlets for a lightweight exoskeleton that allows wearers to expend 24% less metabolic energy when walking, as well as substantially less energy when running or climbing stairs. (The paper describing that research, “Experiment-free exoskeleton assistance via learning in simulation,” was published last year in Nature.) In a major innovation, Su and his team leveraged AI and computer simulations to train the device, eliminating the need for time-consuming human tests and handcrafted control laws. Typically, users must spend hours “training” an exoskeleton so that the technology knows how much force is needed and when to apply it, but Su’s framework allows the device to be used immediately.
Su has also developed upper-limb exoskeletons that can help with tasks requiring lifting heavy objects or holding your arms overhead for long periods, posing great potential for industrial use.
Su foresees his assistive devices being of use to both able-bodied people and those with mobility or manipulation impairments, and in addition to working in that realm, he has made great strides in developing surgical robots. Unlike the well-known Da Vinci systems, which perform laparoscopic procedures requiring incisions of about 8mm, his robotic “micro-surgical” system uses an exceptionally thin snake-like device that requires an opening of just 2mm, enabling surgeons to undertake complex neurosurgical and cardiac procedures with minimal invasiveness. In addition, the surgical robots developed in his lab are very compact thanks to his work on high-torque electric motors and bio-inspired robot mechanism design.
Su, who taught at North Carolina State University and University of North Carolina Chapel Hill before arriving in Brooklyn, explains that his work is unique because of his engagement in “full-stack” robotics, a way of working “from scratch,” from the design of high torque electric motors and soft wearable robots to control and machine learning for physical robotic systems (known as physical AI or embodied AI).
Su is a core faculty member of NYU Center for Robotics and Embodied AI, and he hopes to attract students at all levels, from undergraduates to Ph.D.s. “There are exciting opportunities to do hands-on work, potentially collaborate with clinicians, and make a real difference in people’s lives,” he says. “In everything we do, the main focus is on improving the quality of life for humans. I think of this as human-centered robotics, and all our basic research can be described as use-inspired. It’s wonderful to publish papers and present at conferences, but at the end of the day, we want our work to have real-world benefits for society.”
