Successful communication between devices comes down to efficient computing, particularly for autonomous systems. For a group of autonomous devices working together to make coordinated decisions, the need for fast and accurate transmissions is essential.
Chinwendu Enyioha, an assistant professor in the UCF Department of Electrical and Computer Engineering, is working to develop efficient communication techniques for distributed computation. His efforts have earned him one of the most prestigious honors: the National Science Foundation Faculty Early Career Development Program (NSF CAREER) award.
“Cyber-physical systems including are becoming more commonplace,” Enyioha says. “For example, as more smart, intelligent devices are connected to data networks, the communication resources become constrained. Since communication bandwidth is a finite resource, I am fascinated by how we can effectively manage it, while making connected systems work at scale.”
His work, “Limited-Communication Control of Teams of Autonomous Systems,” investigates a novel framework and algorithms to more efficiently control cyber-physical systems. The research aims to improve communication between these systems, with the ultimate goal of enhancing performance to ensure that these connected devices can make accurate decisions.
“My research group studies how to design, control and make groups of autonomous cyber-physical systems work better and efficiently together,” he says. “Cyber-physical systems here include things like autonomous vehicles, energy and power distribution systems, remote sensors and traffic management systems amongst others.”
Limited bandwidth and restricted infrastructure, especially in remote or unfamiliar environments, can make accurate, timely communication between connected devices a challenging endeavor. The tools Enyioha and his team are developing are designed to operate even under adverse conditions, including limitations in communication bandwidth.
“Imagine a future world with aerial vehicles or remote aerial sensors on a mission in an environment with a ‘poor communication network,’” he says. “Outcomes from our work will enable teams of autonomous systems continue to carry out their mission with graceful degradation.”
His research will examine multiple pathways to achieve this goal, including an efficient data compression method that retains accuracy, even in low bandwidth, and a technique that will aggregate data in wireless channels.
“Even as machine learning and artificial intelligence becomes pervasive, computation does not always all happen where the data is,” he says. “Computation is often distributed amongst several servers. We are interested in making collections of such systems work efficiently.”
- Written by Bel Huston