“The movie The Matrix shows a scary virtual world formed by digital data streams, but isn't it cool to have a data stream reflecting all the details of our traffic world that we can use to save lives, time, and energy?
That is what I do with my research on roadside LiDAR networks.
When I was pursuing my bachelor's degree and master's degree in electrical engineering in China, my initial career plan was to be a software engineer and get nice paychecks. My master’s thesis on the Global Positioning System and the Geographic Information System opened a new gate to applications of data and technologies in transportation systems. "Well,” I thought in preparing my thesis defense, “I may be a good one-among-millions of software engineers, but I can do something unique by integrating my electronic engineering capabilities and transportation systems."
I came to the United States for my civil engineering (transportation) Ph.D. at Texas Tech University in 2007. Roadway design, signal timing planning, traffic performance evaluation, and traffic simulation exposed me to the amazing transportation engineering world and prepared me for my dream career.
Like many dream stories, the process was not smooth, and sometimes it was boring, frustrating, confusing, and disappointing. I watched freeway CCTV videos daily to report any traffic incidents as a part-time traffic engineering student; collected traffic information by cameras and manually extracted the required information; and stood at an intersection to document the traffic queue length change. There were a bunch of traffic data, but the quality was a long way from answering our questions. Engineers and researchers either manually reviewed videos and extracted very limited information or just gave up and left the questions for the future. I worked on various transportation engineering projects, met different data challenges, but I could not find a good solution to the data problems, knowing that data is the foundation of all traffic systems.
Before finding my uniqueness, my hard work paid off in 2013 by bringing me to an assistant professor position at the University of ÁùºÏ±¦µä, Reno in the department of Civil and Environmental Engineering. It was a big and exciting change for a researcher, but my questions about traffic data challenges were still not being answered.
At the University, we began to have biweekly meetings — initiated by Dr. Mridul Gautam in early 2016 — to chat and brainstorm ideas for a multidisciplinary team of researchers. Magic happened in the summer of 2016 at a brainstorm meeting when a colleague mentioned a LiDAR sensor's price had been reduced to $8,000. I had heard about LiDAR sensors, a major component of autonomous vehicles for high-accuracy spatial point measurement data, but it had been too expensive —more like tools for rich companies or programs. A price of only $8,000 was the threshold price for my research program and an acceptable price for roadside traffic sensing systems. It was the moment that I have prepared myself for many years and it suddenly happened.
Yes, the data is beautiful, unique, accurate, and full of details. I was happy — even a bit crazy with glee — when I opened the data file from the first roadside test. I guess it's like when the protagonist of The Matrix saw the raw data flow of his virtual world for the first time. At that moment, I knew that the sensor data would change transportation engineering a lot! Moving LiDAR sensors from vehicles to roadside infrastructure is the silver bullet for traffic data and information. It generates the data stream of our real world with fantastic detail and accuracy. Since then, I have focused all of my research efforts on developing the hardware system, data processing algorithms, software, communication to connected and autonomous vehicles, and automatic extraction of meaningful traffic measurements.
In 2017, we set up the world’s first LiDAR-equipped, smart-and-connected intersection in Reno, ÁùºÏ±¦µä. In 2019, we implemented the world’s first LiDAR-controlled pedestrian crossing signal, the first real-time traffic signal system controlled by cloud point sensing data. In 2021, the University and Velodyne LiDAR published the first-in-class white paper that demonstrates the ability of LiDAR sensors to make transportation infrastructure more efficient, sustainable, and safe. Powered by the LiDAR network data, new traffic analysis and systems have been developed and employed by cities and states.
I found my uniqueness in integrating technologies and transportation, and now I continue to apply the 'silver bullet' to solve traffic problems one by one."
Learn more about Hao Xu.