Brain4Cars analyzes driving behavior with AI to predict and prevent road accidents

By Chloe Olewitz Published October 14, 2015

As cars gets smarter, tech developments in our dashboards aim to help humans keep up, or at the very least to keep human drivers alert behind the wheel. To make this happen, researchers have developed a deep learning neural network called Brain4Cars that combines the many approaches of top-of-the-line smart car technology, and augments them with artificial intelligence.

Cars obviously aren’t the only things getting smarter, and as our gadget collections pile up we’re prone to becoming ever more distracted. Even with existing road safety regulations and legislation to keep drivers safe, 33,000 people die in road accidents every year. What’s more, over 90 percent of road accidents in the U.S. can be attributed to driver error, according to the National Highway Traffic Safety Authority.

Recommended Videos

Many car manufacturers have already taken to installing safety systems in new models that alert drivers to dangerous behaviors and external environmental risks. Brain4Cars basically takes all these technologies and rolls them into one, combining safety measures like internal and external cameras with GPS, vehicle dynamics reporting, and an extensive database of driver behavior recordings.

What sets Brain4Cars apart is what it does with this comprehensive, real-time data collection. By feeding information from the data sources into a proprietary artificial intelligence, the system is able to reliably predict individual driver behavior in real-time, and anticipate errors up to 3.5 seconds in advance with an 80% accuracy rate.

Researchers from Stanford and Cornell modeled the Brain4Cars system on Recurrent Neural Networks with Long Short-Term Memory. The artificial intelligence analyzes in-car visual cues like the driver’s facial expressions and sightlines, while external camera feeds provide the system with a contextual understanding of the safety and appropriateness of driver choices in relation to the situation unfolding on the road.

In theory, a deep learning architecture like this could come to learn the habits of each individual driver it assists. Over time, Brain4Cars would be able to predict driver accuracy with greater precision and recall as data from the specific driver becomes part of the network’s knowledge base.

This background contextual information provided to the neural network was amassed over the course of 1,180 miles driven collectively by ten different drivers in California. Road areas included freeway and city driving to expand the AI’s contextual knowledge for commonly encountered road situations and external car environments.

Even driverless cars get their fair share of heat when it comes to on-road errors, but prediction about the future of autonomous cars suggest that accidents and collisions will be mostly due to human error — not robotic drivers. But instead of replacing human drivers entirely, Brain4Cars combines the best of driver safety tech to bring a bit of robotic precision support to humans behind the wheel.

Topics

Former Digital Trends Contributor

Chloe is a writer from New York with a passion for technology, travel, and playing devil's advocate. You can find out more…

Cars

Rivian set to unlock unmapped roads for Gen2 vehicles

rivian unmapped roads gen2 r1t gallery image 0

Rivian fans rejoice! Just a few weeks ago, Rivian rolled out automated, hands-off driving for its second-gen R1 vehicles with a game-changing software update. Yet, the new feature, which is only operational on mapped highways, had left many fans craving for more.
Now the company, which prides itself on listening to - and delivering on - what its customers want, didn’t wait long to signal a ‘map-free’ upgrade will be available later this year.
“One feedback we’ve heard loud and clear is that customers love [Highway Assist] but they want to use it in more places,” James Philbin, Rivian VP of autonomy, said on the podcast RivianTrackr Hangouts. “So that’s something kind of exciting we’re working on, we’re calling it internally ‘Map Free’, that we’re targeting for later this year.”
The lag between the release of Highway Assist (HWA) and Map Free automated driving gives time for the fleet of Rivian vehicles to gather ‘unique events’. These events are used to train Rivian’s offline model in the cloud before data is distilled back to individual vehicles.
As Rivian founder and CEO RJ Scaringe explained in early March, HWA marked the very beginning of an expanding automated-driving feature set, “going from highways to surface roads, to turn-by-turn.”
For now, HWA still requires drivers to keep their eyes on the road. The system will send alerts if you drift too long without paying attention. But stay tuned—eyes-off driving is set for 2026.
It’s also part of what Rivian calls its “Giving you your time back” philosophy, the first of three pillars supporting Rivian’s vision over the next three to five years. Philbin says that philosophy is focused on “meeting drivers where they are”, as opposed to chasing full automation in the way other automakers, such as Tesla’s robotaxi, might be doing.
“We recognize a lot of people buy Rivians to go on these adventures, to have these amazing trips. They want to drive, and we want to let them drive,” Philbin says. “But there’s a lot of other driving that’s very monotonous, very boring, like on the highway. There, giving you your time back is how we can give the best experience.”
This will also eventually lead to the third pillar of Rivian’s vision, which is delivering Level 4, or high-automation vehicles: Those will offer features such as auto park or auto valet, where you can get out of your Rivian at the office, or at the airport, and it goes off and parks itself.
While not promising anything, Philbin says he believes the current Gen 2 hardware and platforms should be able to support these upcoming features.
The second pillar for Rivian is its focus on active safety features, as the EV-maker rewrote its entire autonomous vehicle (AV) system for its Gen2 models. This focus allowed Rivian’s R1T to be the only large truck in North America to get a Top Safety Pick+ from the Insurance Institute for Highway Safety.
“I believe there’s a lot of innovation in the active safety space, in terms of making those features more capable and preventing more accidents,” Philbin says. “Really the goal, the north star goal, would be to have Rivian be one of the safest vehicles on the road, not only for the occupants but also for other road users.”

Cars

Jaguar Land Rover, Nissan hit the brake on shipments to U.S. over tariffs

Range Rover Sport P400e

Jaguar Land Rover (JLR) has announced it will pause shipments of its UK-made cars to the United States this month, while it figures out how to respond to President Donald Trump's 25% tariff on imported cars.

"As we work to address the new trading terms with our business partners, we are taking some short-term actions, including a shipment pause in April, as we develop our mid- to longer-term plans," JLR said in a statement sent to various media.

Computing

DeepSeek readies the next AI disruption with self-improving models

DeepSeek AI chatbot running on an iPhone.

Barely a few months ago, Wall Street’s big bet on generative AI had a moment of reckoning when DeepSeek arrived on the scene. Despite its heavily censored nature, the open source DeepSeek proved that a frontier reasoning AI model doesn’t necessarily require billions of dollars and can be pulled off on modest resources.

It quickly found commercial adoption by giants such as Huawei, Oppo, and Vivo, while the likes of Microsoft, Alibaba, and Tencent quickly gave it a spot on their platforms. Now, the buzzy Chinese company’s next target is self-improving AI models that use a looping judge-reward approach to improve themselves.