Supported by a three-year, multimillion-dollar grant from the Defense Research Projects Agency (DARPA), a multidisciplinary team led by a Johns Hopkins engineer will investigate approaches to more quickly move self-operating systems from simulated environments to real-world use.

“As self-operating systems become more common, they are rapidly changing many aspects of our lives. These systems need to be tested in simulated environments to ensure what we learn in simulations can be applied to autonomous tech in a real-world setting, including autonomous systems used in humanitarian disaster relief or for navigation purposes,” said PI Vishal Patel, associate professor of electrical and computer engineering at the Whiting School of Engineering.

Patel is working on the project with Rama Chellappa, Bloomberg Distinguished Professor in electrical and computer engineering and biomedical engineering, Judy Hoffman at Georgia Institute of Technology, Wei-Chiu Ma at Cornell University, and Pavan Turaga at Arizona State University.

It is critical to develop theory and algorithms that bridge the gap between real and simulated environments, says Patel, noting that the testing can help create more successful autonomous navigation systems by overcoming challenges in these mixed environments.

“Receiving this DARPA grant is a significant milestone for both me and my colleagues,” says Patel. “It represents recognition of the innovative potential of our research and the trust that DARPA has placed in our team to tackle complex and impactful challenges in autonomy. This work could help advance the way we develop autonomous systems, and the results could have far-reaching impacts on how U.S. government organizations, including the Department of Defense, can conduct their work.”

The project aligns with the objectives of the Johns Hopkins Data Science and AI Institute, and the Johns Hopkins Institute for Assured Autonomy because the research team will be working with large amounts of data and autonomy transfer applications.