AUSTIN, Texas – A sea of brake lights can be seen across Austin highways during daily rush hour. Austin commuters are all too familiar with the headache of the city’s traffic congestion problems.
The University of Texas is working on a new artificial intelligence coupled with a supercomputer to innovate Austin’s traffic flow. Researchers with the Texas Advanced Computing Center are utilizing raw traffic footage to analyze the flow of transportation.
"It's an object identification algorithm where it's training the software to be able to identify different things," said Jen Duthie, a consulting engineer at the Austin Transportation Department.
The artificial intelligence (AI) will characterize how people, cars, buses, bicycles, and motorcyclists move through traffic lights and interact with each other. The information will be logged for traffic engineers to track the problems that arise.
The UT researchers said by understanding traffic volume over-time they will be able evaluate the areas with the most dangers. For example, the algorithm will be able to log how many wrong-way drivers a particular street has.
"Having all of this data will help us know where potential problems are, which is really exciting so we don’t have to wait until that crash occurs," Duthie said. "We can see if there are a lot of behaviors that look like they could lead to collisions, so that we can then focus on that intersection in a more proactive manner."
The automation will observe safety issues as well. The system will look at places where pedestrians are illegally crossing busy streets and how far they are willing to walk to use a crosswalk.
City planners will be able to use the AI to target the areas of Austin that need the most improvements without having to comb through countless hours of raw footage.
“The highly anticipated introduction of self-driving and connected cars may lead to significant changes in the behavior of vehicles and pedestrians and on the performance of roadways,” said Natalia Ruiz Juri, a research associate and director of the Network Modeling Center. “Video data will play a key role in understanding such changes, and artificial intelligence may be central to enabling comprehensive large-scale studies that truly capture the impact of the new technologies.”
The technology reinforces the City's committment to Vision Zero. The initiative seeks to reduce traffic fatalities to zero through engineering, education, and enforcement.
"As traffic changes so rapidly in Austin we need to be able to keep an continuous eye on what's happening and how things are changing, so that we’re designing appropriately," Duthie said.