Smart Car Researcher Wants to Eliminate Stoplights

Christos Cassandras has spent more than 40 years studying how computers and machines make decisions. Lately, his work has focused on a problem nearly every driver knows well: sitting at a red light with no cross traffic in sight, wondering why it takes so long for the light to turn green. Cassandras, an IEEE Life Fellow, is a professor at Boston University, where he headed its systems engineering division, a graduate program he helped create. Christos Cassandras Employer Boston University Title Professor Member grade Life Fellow Alma maters Yale, Stanford, and Harvard He is the recipient of this year’s IEEE Intelligent Transportation Systems Society Outstanding Research Award for his work on autonomous vehicles and the systems that help them drive—work that could one day get rid of traffic lights altogether. That idea—proving with mathematical models and intelligent algorithms using cars’ relative speeds, mass, and distance as inputs that self-driving cars will behave safely before they are ever let loose on the streets—has become the foundation of Cassandras’s career and the reason his name keeps coming up in conversations about transportation. In the 1990s he theorized that a machine such as a car could be understood as two things at once: an object obeying the laws of physics and a computer processing information and making decisions. This led to what we now call cyber physical systems. That framework is still the standard way engineers describe a car’s mechanical behavior. “Ultimately, the key word is safety,” Cassandras says. “You would not buy a self-driving car unless the manufacturer could guarantee that it’s safe for you.” Growing up in Greece during a dictatorship Born in Athens, Cassandras was a teenager during Greece’s military dictatorship, a period he says scarred him and many others but also taught him some hard lessons. His father was a partner in an insulation manufacturing company, and his mother was a homemaker. His friends pulled Cassandras toward engineering. As a teenager, he ran with a group of classmates who read about philosophy, history, and science. They were especially struck by the 1969 Apollo 11 moon landing and the questions it raised about the technology that made it possible. That same year also marked the first time a piece of data was sent from one computer to another—an early step toward the creation of the Internet. Several inspiring teachers pushed his curiosity about technology even further. He was awarded a scholarship to study in the United States, and he enrolled at Yale, where he first studied physics and philosophy before switching to engineering. He earned his bachelor’s degree in engineering and applied science in 1977. At the time, Yale didn’t yet offer separate degrees in fields such as electrical or mechanical engineering. From there, Cassandras went to Stanford, earning a master’s degree in electrical engineering in 1978. There, he became interested in game theory, the mathematical study of decision-making and strategy. A mentor pointed him to Harvard, home to a leading game theory researcher at the time. His new mentor, Professor Yu-Chi “Larry” Ho, now an IEEE Life Fellow, advised him to abandon game theory, which Ho considered a scientific dead end at the time. Ho suggested a newer, less-explored area of research, that of emerging dynamic systems in modern technology whose behavior could be understood and managed through occurrences of discrete events. Cassandras took the advice. He earned a second master’s degree from Harvard the following year and completed his Ph.D. in applied mathematics there in 1982. He and Ho remain friends and colleagues. Fiat assembly line births of a new kind of engineering Ask people in intelligent transportation systems circles why Cassandras’s name carries so much weight, and the answer traces back to a theoretical shift he helped pioneer starting in the 1980s. Cassandras recognized that human-made systems—computers, factory machines and, eventually, vehicles—operate on a different kind of logic than the one that governs the natural world. The physical world runs on time-driven physics, the same math Isaac Newton used centuries ago. But machines built by people, he realized, are better understood as event-driven: They mostly sit still until something happens, like a button click or a part arriving on a conveyor belt. That insight didn’t come from a textbook. It took shape during Cassandras’s early graduate work, when he was handed a real-world problem. He was assigned to figure out how to manage the buffers—the temporary holding areas for car parts—on an assembly line for Fiat, the Italian carmaker. Solving that concrete, practical puzzle helped him see how event-driven thinking could describe an entire class of systems that classical physics-based math couldn’t handle well. This came to be known as discrete event systems. By the early 1990s, that insight had matured into a formal framework: hybrid systems theory, now k

Extras afișat conform termenilor RSS ai publicației · 153 cuvinte · textul integral și figurile sunt pe site-ul publicației