How artificial intelligence is being used to help reduce crashes


The AI studies cars, pedestrians, and bicyclists with their activity, including locations, lane, speed and acceleration. — Photo by Nabeel Syed on Unsplash

Some roadways in Florida are becoming a testing ground for artificial intelligence to help find solutions to prevent crashes.

In a leap forward toward safety, Broward County is using the University of Florida’s artificial intelligence (AI)-based algorithms to track cars, bicyclists and pedestrians by camera. As part of a pilot program, researchers are analysing how to avoid collisions before they even happen – and using AI to help surface some solutions.

Broward is the latest community in the state to use this new approach. Pilot programs also have taken place in Gainesville and Seminole County.

Before, researchers had to see when a “crash happened and afterwards we have to look at information to determine the cause, what kind of safety solution may be helpful,” said Min-Tang Li, spokesman for Broward’s Traffic Engineering Division. But this new approach “will allow us to be more proactive – to identify potential crashes – and also gives us a better picture how things happen.”

Six intersections along Stirling Road in Broward are being examined for the number of times crashes nearly happened there, “because traffic intersections are prone to crashes leading to injury or fatality,” according to a research report.

An analysis

The technology studies video data from traffic intersections to understand where “near misses” – but not actual accidents – are happening.

That can include what time of day most near-accidents happen, and whether there are more issues with drivers making left turns or right turns, said University of Florida professor Sanjay Ranka. “The overall goal is use of these near-misses” to have fewer accidents “and obviously less fatalities,” he said.

The AI studies cars, pedestrians, and bicyclists with their activity, including locations, lane, speed and acceleration, Min-Tang Li said.

Relying on witnesses in the past only gave a limited sense of what happened. That kind of insight can be misguided because there are factors that influence facts, such as where the witness was standing, Li said.

“We can’t get the whole picture, more objective information is difficult to obtain,” Li said.

Ranka said one outcome from the Gainesville study was realising that pedestrians needed more time to cross one particular intersection near a high school. But doing that would slow down car traffic. The final suggestion was to increase the pedestrian crossing time, but only in the afternoons during peak pedestrian hours. “There was no point in making changes throughout the day,” he said.

Thanks to the program in Broward, some changes already have occurred, such as new signs reminding drivers to stop for pedestrians. The final analysis won’t be ready until at least June. The program is expected to be expanded elsewhere in the county, officials said. – South Florida Sun-Sentinel/Tribune News Service

Follow us on our official WhatsApp channel for breaking news alerts and key updates!

Next In Tech News

'Just the Browser' strips AI and other features from your browser
How do I reduce my child's screen time?
Anthropic buys Super Bowl ads to slap OpenAI for selling ads in ChatGPT
Chatbot Chucky: Parents told to keep kids away from talking AI dolls
South Korean crypto firm accidentally sends $44 billion in bitcoins to users
Opinion: Chinese AI videos used to look fake. Now they look like money
Anthropic mocks ChatGPT ads in Super Bowl spot, vows Claude will stay ad-free
Tesla 2.0: What customers think of Model S demise, Optimus robot rise
Vista Equity Partners and Intel to lead investment in AI chip startup SambaNova, sources say
Apple plans to allow external voice-controlled AI chatbots in CarPlay, Bloomberg News reports

Others Also Read