Using AI To Help Us Drive Safely

Researchers at the University of Cambridge and Jaguar Land Rover teamed up to create a smart algorithm that can make driving safer. This algorithm predicts the best times for drivers to use in-car systems or get messages without causing distractions. They figured this out by testing it on real roads and using fancy computer learning.

The algorithm keeps an eye on how much ‘work’ the driver is doing, like if they’re in a new place or just doing their usual drive. It’s super flexible and can quickly adjust to changes in the driver’s behavior, road conditions, or the type of road.

This clever tech can then be used in car systems, like the screen that shows directions or the one with music and stuff. It helps make sure that when your attention is really needed on the road, you won’t get any distracting messages or alerts. It’s like having a smart assistant in your car, making sure everything is safe and easy for you while you drive.

Safer motoring

“More and more data is made available to drivers all the time. However, with increasing levels of driver demand, this can be a major risk factor for road safety,” the researchers explain. “There is a lot of information that a vehicle can make available to the driver, but it’s not safe or practical to do so unless you know the status of the driver.”

The researchers wanted to find a better way to measure how much attention drivers need without using fancy gadgets like eye trackers or heart rate monitors. Instead, they focused on everyday car data like how you steer, accelerate, and brake. They also wanted their method to handle different data speeds and types, like from biometric sensors.

To figure out how much mental effort drivers were putting in, the researchers made a simple test. They put a phone with a navigation app in the car, and a little light blinked now and then. Drivers had to press a button when the light turned red and they felt things were easy.

Driving aids

By watching videos of this test and looking at button data, the researchers figured out when driving was tough – like at busy intersections or when another car was acting weird. They then used this real-life data to create a smart computer system.

This system can understand a driver’s average workload and quickly estimate how hard they’re working at any given moment, using info like steering and braking. It’s like having a smart helper that can tell when driving is a breeze or a bit tricky.

“For most machine learning applications like this, you would have to train it on a particular driver, but we’ve been able to adapt the models on the go using simple Bayesian filtering techniques,” the researchers conclude. “It can easily adapt to different road types and conditions, or different drivers using the same car.”

Facebooktwitterredditpinterestlinkedinmail