Findings Staff Report | April 14, 2019
Last summer, five cars drove in a straight line down a Virginia highway, one following closely behind the next—without drivers.
It was a major step forward in the development of connected and automated vehicles and an experiment co-designed by University of Cincinnati professor Jiaqi Ma and his team at the Next Mobility Lab.
In truth, the last 18 months have been quite busy for Ma, whose lab has been featured in news articles by WCPO and WVXU. He's expecting his first automated vehicle any day now and was just granted a $70,000 OoR Strategic Team award with partners Ming Tang and Julian Wang to begin getting the vehicle “in the loop.” OoR was excited to feature his work in Research + Innovation Week.
“R&I Week focuses on remarkable faculty, such as Dr. Jiaqi Ma, working in future-forward fields that help both Cincinnati and UC attract world-class research talent who want to collaborate to solve real-world problems and top-notch student talent who are excited to learn from them and make a difference" said Assistant Vice President for Integrated Research Jennifer Krivickas.
We sat down with Ma to talk about his work and a new initiative called the Greater Cincinnati Advanced Transportation Collaborative.
Q: First, tell me briefly about yourself. Where are you from? How did you get to UC?
A: I got my Ph.D. at the University of Virginia in Advanced Transportation Systems focused on connected vehicles. We are trying to connect them together, so that they communicate in real time. It could have great benefits for safety and the mobility of our transportation system. Then, I worked for the Federal Highway Administration in the Saxton Transportation Operations lab of the Turner-Fairbank Highway Research Center, for almost 5 years in connected and automated vehicles, forming my main research focus: Cooperative Automation. I was hired by UC in 2017 with the goal to boost the advanced transportation program at UC.
Q: I’ve been hearing about the Greater Cincinnati Advanced Transportation Collaborative. What is it?
A: It’s a partnership housed in the Office of Research and involves key researchers from different colleges and departments who have a variety of expertise related to emerging transportation technologies and believe that advanced transportation can create a safer, healthier tomorrow. We just launched the website during Research + Innovation Week. There are three aspects of our work. One, we simulate how collaborative or autonomous vehicles might work. Two, we study how the human driver responds and reacts while taking a simulated collaborative or autonomous test drive. And finally, we just acquired an automated vehicle, so now we can test different algorithms on real roads and collect data and share it with the research community. Partners include the regional Ohio-Kentucky-Indiana Council of Governments, Ohio Department of Transportation, DriveOhio (the state’s smart mobility initiative), Ohio Turnpike and Infrastructure Commission, Cincinnati/Northern Kentucky International Airport and City of Cincinnati.
Q: You were recently involved in a test of several automated vehicles on a highway in Virginia, doing something called ‘platooning.’ What was that all about?
A: These are connected automated vehicles in real time in a complex environment. We want them to talk to each other so they can drive in a small tight platoon, with small gaps between them. Platooning can be much more stable than our current driving environment. It could stabilize our traffic flows and, in our estimation, increase highway capacity by 50 percent. Last summer, I was a Co-PI on that project that was done by the U.S. Department of Transportation. We used a five-vehicle fleet to do platooning experiments—one of the first times on a public road. It was a team effort, we (Ma’s Next Mobility Lab) created the algorithm, designed the experiment and analyzed the data. Our long-time partner, Leidos, did the software and experimental design.
Q: How did the experiments go?
A: It was one of the first real experiments done in traffic, so it was expensive and resource and time intensive, but it proves the concept. We proved that platooning and collaborative merge works. A 0.6 second time gap— it’s possible and the platoon performs well. And now we have data to continue studying highway capacity and to calibrate our simulation models to create a better modeling tool for the future research.
Q: When do you suspect the common person could see a line of trucks or cars platooning on a local highway?
A: There is a distinction between a fully autonomous vehicle – Level 5 – and a longitudinally controlled vehicle, like the ones in the platooning experiment—Level 1. The Level 5 vehicles are absolutely not here yet and probably not for another 10-20 years. But platooning would just require some roadside infrastructure and the proper devices in a vehicle, so from my perspective that can come faster, and we could start seeing the early benefits that automated vehicles could have on our transportation system.
Q: What else are you up to these days?
A: Another aspect of my main research is Urban Analytics. These days, vehicles are equipped with many advanced sensors, such as lidar, radar, and cameras. There are all sorts of devices along the roadway that could be connected by BlueTooth or other wireless technologies. How do we use all of these data to work smarter, more efficiently and safely? We often don’t observe our systems that are already in place for these opportunities. We are also working very closely with other government agencies, using our expertise to help solve a problem or create new technologies to enhance their state of the practice. On a recent project with the Ohio Department of Transportation, we teamed up with UC DAAP to create simulated snow and ice plow driver training. This is traditionally behind-the-wheel training, but when the winter comes, frankly you don’t have enough time to do the training. People have to get out on the road and clean things up. We are building a simulation of 200 miles of real Ohio roads—the most difficult sections of road according to ODOT data—and creating high-resolution simulated scenarios in which drivers can be trained to make decisions on driving under adverse weather and applying chemicals. Potentially, this can reduce crashes by 30 percent, and our data is showing possibly more. The simulation will be in a trailer that can be moved around the state and we expect to start training drivers sometime in 2020.
Q: You were recently awarded a $70,000 Strategic Grant from the Office of Research. How will it be used?
A: The Office of Research has been very supportive of my work and contributed to purchasing my lab’s automated vehicle, fitted with many different sensors. The money from this grant will allow us to create hardware- and software-in the loop capabilities, which means testing our algorithms and collecting data related to automated driving.
Featured image at top: UC professor Jiaqi Ma stands next to a driverless vehicle outfitted with sensors. Photo/Bryan Brown/UC