The average person can purchase a drone online or at a big-box retailer. This broadened accessibility has security implications for critical infrastructure such as power plants, sports stadiums and correctional facilities.
Traditional drone detection technologies, such as radar, have proven somewhat ineffective at identifying drones due to their small size and low flight altitude.
As drone technology continues to advance, researchers are developing new techniques for detecting them. Huaxia Wang, Ph.D., an assistant professor in the Henry M. Rowan College of Engineering, is leading a team of graduate and undergraduate students at the forefront of establishing new drone detection systems.
Blending his previous work in AI and wireless communication, Wang’s drone detection intervention includes an antenna to capture radio frequency signals transmitted between the drone and its operator, enabling the system to detect and estimate the locations of both.
Because there are dozens of different types of drones, Wang and his team must first collect each type’s signal “signature.” The second component of the process includes an artificial intelligence algorithm that can detect, classify and localize the drone in the sky.
“Every week, we are going to fly different types of drones on campus in order to collect their corresponding data,” Wang said.
That dataset will be widely available for other researchers, he said.
“We want to share the full dataset we have collected at the Rowan campus and let everyone utilize the data we have collected to boost this kind of drone defense research,” Wang said.
Wang is partnering with Ziang Gao, who leads the R&D team at New Jersey-based drone detection company AeroDefense, to advance and commercialize the technology. Together, the teams have collected data across real-world environments, including stadiums, urban areas, and office and university campuses, to train and validate the system under diverse operating conditions.
"We want to collect as much environmental data as we can to further enhance our model training because our model needs to see different scenarios like a rainy day, a sunny day, and also during the event, before the event and after the event,” Wang said. “We want our model to be aware of those environments and get sufficiently trained.”
