Articles, Machine Learning, Sensing

This AI Startup is On a Mission to Eliminate Car Accidents

The hype for connectivity has never been so high. Between automobiles, smart devices and the Internet of Things, virtually every tech company is attempting to connect their electronics in some shape or form.

But while many of these ideas have already fizzled, motor vehicles stand to gain the most from connectivity. Forget email, text and entertainment – the future of in-car connectivity could reduce the number of deadly accidents. And for Derq, an ambitious Dubai-based startup, the future is almost here.

“We develop artificial intelligence-driven V2X applications,” said Georges Aoude, co-founder and CEO of Derq. “These applications are for smart cities, connected cars and autonomous vehicles.”

Derq was spun out of MIT with one lofty mission: to improve road safety by eliminating car accidents. That may sounds like an impossible task, but if Derq could merely reduce the number of accidents by a few percentage points, the positive impact would be enormous.

“We use AI, machine learning, to predict dangerous situations/dangerous driving, and send alerts to surrounding vehicles about these dangers with enough time for them to react and avoid accidents,” said Aoude. “Why is this important? Today, already, you have different types of users – cars, trucks, pedestrians, cyclists. It’s a really complex environment and those users crash often.”

With the addition of new vehicle types – including partial and fully autonomous – roadways are likely to get more complex in the years to come. While self-driving cars are being designed to simplify mobility, their arrival will not bring about any major changes overnight. But they could add a new layer of safety concerns as human drivers learn to share the road with robots. Thus, there needs to be a way to improve safety for all drivers, regardless of how much technology is under the hood.

“We make the car smarter about the environment, and we make the environment smarter about the vehicles in it,” Aoude added. “By doing that, we are actually improving the safety of the cities and of the users. This will be important not just for today but for the future where autonomous cars will be added to the mix.”

Alert Delivery

Most cars are not yet connected to each other or the Internet, but Derq is working around this challenge. The company’s current plan is to use external technology (such as cameras and radar already in place within a city) to detect the speed at which a car is moving. Derq will then use that information to determine (based on data collected over time) the likelihood that a car will stop at an intersection if the light changes.

With the proper infrastructure in place, this info can be retrieved from any vehicle, even those that aren’t connected. When applicable, the data can then be relayed to vehicles that are connected via 5G or DSRC.

“These alerts could be visual or audible,” Aoude explained, adding that an alert could be as simple as a beep. “It gives you enough time to react to the alert and take actions that will help you avoid accidents.”

Derq may experiment with other options, such as in-car vibrations, depending on how end users (OEMs and ultimately consumers) respond to these ideas.

More Than Drivers

Road safety is not limited to those sitting behind the wheel. Pedestrians may also be hurt when accidents occur, so Derq is looking at ways to improve their safety as well.

“You can track pedestrians externally with sensors,” said Aoude, comparing this method to the way automobiles can be monitored. “In the future, if you have more and more humans using connected devices, whether a phone or some other device, we can tap into that as well.”

About the author:

Louis Bedigian is an experienced journalist and contributor to various automotive trade publications. He is a dynamic writer, editor and communications specialist with expertise in the areas of journalism, promotional copy, PR, research and social networking.

Leave a Reply

Your email address will not be published. Required fields are marked *