Analysis Featured

Computer Vision Can Reduce Auto Fatalities & Save the Planet

collision avoidance

Collision avoidance systems can make transportation better for the entire planet. Uber wants to replace its human fleet with robotic cars. Google is already testing autonomous vehicles with great success. The word is that autonomous driving is safer than human drivers, but we all know how hard it is to give up the reins. Chris Carson, CEO of Caruma Tech, says it will be 10 years, at least. Goldman Sachs predicts almost 60% autonomous driving by 2030, making a distinction between limited self-driving and full autonomy. Meanwhile, advanced driver assisted systems (ADAS) can make driving safer with drivers behind the wheel and make planet Earth a healthier place to live.

“The average car is parked 95 percent of the time,” Carson said, quoting the Goldman Sachs study. “If autonomous driving vehicle were running 100 percent of the time, that would suggest we might need only five autonomous vehicles for every 100 cars driven by a human.”

He envisions drivers sitting behind the wheel working on their laptops as their vehicles move them to work and back home again. With cloud-based predictive analytics empowering a network of autonomous vehicles, it’s feasible that an Uber-like fleet of vehicles could deliver workers to and from jobs each day and vehicle ownership could be a thing of the past.

“We could reduce roads from four lanes to just two,” he said. “And then we could plant more trees.”

If Carson’s vision is correct, we could reduce our dependence on fossil fuels, reduce dangerous emissions, and make driving safer all at the same time.

Cars Kill More People Than Guns

Nearly 1.3 million people die in road crashes each year around the world, according to the Association for Safe International Travel. Another 20 to 50 million are injured or disabled. Autonomous vehicles promise to take care of that, but Carson says we’re a decade away from that reality, at least. Carson has a more immediate solution.

Using computer vision and machine learning technologies, Carson started Caruma Tech in 2014 to make cars safer and more secure by providing dual cameras that use object detection sensors to alert drivers of possible collision dangers. These may include but are not limited to potholes, pedestrians, other vehicles, emergency vehicles, road signs, and buildings. If a vehicle does get too close to an obstacle that could pose a danger, Caruma Cam will send an alert to the driver to react to the danger. The system can also detect whether the driver is drowsy or distracted.

When a vehicle is parked, Caruma Tech’s Caruma Cam will act as a security system, monitoring the vehicle for potential break-ins and vandalism. If it detects a threat, it will send an alert to the vehicle owner’s smartphone where he’ll able to see inside the vehicle in real time.

The Benefits of Caruma Cam

“Caruma Cam deploys two HD cameras,” Carson said. “There’s an outward-facing cam and an inward-facing cam, giving the driver a total view of 340-degrees.” Each camera, mounted in an above-head console where a radar detector is typically placed, has a field of vision of 170-degrees.

Caruma Cam relies on ADAS to deliver benefits to three primary markets – fleet managers, insurance companies, and automotive companies. With 90 million new cars hitting the road every year, Carson sees a lot of potential for these markets, but he’s just getting started.

“Any time there’s a new innovation,” he said, “people don’t know how to take it. When I started talking about this two years ago, nobody understood what I was talking about. Now they’re starting to get it.”

Caruma Tech targets insurance companies hoping they’ll install Caruma Cam in their customers’ vehicles, which should reduce insurance rates for good drivers. Fleet managers can better protect their assets while monitoring driver behavior. Automobile manufacturers can add a new attraction to vehicles to make them more competitive and sell more vehicles while protecting drivers on the road.

“More than half of all auto collisions are due to distracted driving,” Carson said. “There hasn’t been a real immediate solution for that, until now.”

More important than the benefits of saving lives, Carson believes, is the potential to collect information on driver habits.

“We can start collecting that data now and industrialize the platform,” he said.

Using a combination of device storage and cloud-based storage, Caruma Cam is capable of collecting information regarding driver behavior, which insurance companies can use to set rates by, fleet managers can use to manage their employees by, and automobile manufacturers can use to build safer vehicles. Carson sees a future where every vehicle equipped with object detection capabilities and driver behavior monitoring, which can keep everyone on the network safer and automobile transportation more efficient.

The basic cam product uses a 64-bit quad-core CPU at 2.4 GHz and an 18-core GPU at 750 MHz. The system memory is 4 GB. To connect to the cloud, Caruma Cam uses 4G LTE, although Carson sees the day when 5G will be ubiquitous.

“4G is not instant,” he said. “5G has no latency.”

If all cars on the road were equipped with Caruma Cam, a vehicle two blocks over could alert the cloud of a hazard and every vehicle on the network would know about it.

“Autonomous vehicles have line-of-sight vision,” he said, “but they don’t have the ability to tell you what objects lie around the corner. If a vehicle with an object detection system two miles away can alert you to a road hazard now, you can plan your route around that hazard and avoid a delay or a potential collision.” That’s where he sees this technology moving.

Caruma Tech Closes Round A with $7 Million

While Caruma Tech does have one competitor, that startup hasn’t gone to market yet. In fact, Carson said the only competition is indirect.

“There’s one product that includes certain features we offer, but they’re not as comprehensive as we are. We’re going to be first to market, and that’s a good position to be in.” That should be some time next year, he said.

Caruma Tech started with $1 million in seed money, mostly from private investors. The company is now getting close to ending a Series A funding round for $7 million, which will help drive its marketing power closer to the finish line.

Carson got the idea for Caruma Cam after earning his Ph.D. in computer vision in Waseda University in Japan. He wrote his dissertation on robotic navigation for security robots, but since he was a car enthusiast (he drives a Porsche 911), he decided to put his mind to work on how he could use his expertise to improve vehicles. Now, after two years, he’s starting to see traction in each of the markets he’s identified.

The most challenging of the three markets, however, is the automotive market. Since the sales cycle is a three-to-five year cycle, it will be some time before he sees Caruma Cam in vehicles coming off the assembly line. But he has built a relationship with a tier 1 company while establishing direct communications with at least one auto manufacturer. To date, Caruma Tech has pre-sold 30,000 units and hasn’t officially entered beta yet. Carson expects to enter the retail market within three to four years.

When more people die of auto collisions every year than all wars combined, it’s time to think of a solution that will save peoples lives, prevent loss due to destruction of property, and create a safer environment for all.

“We’ve stopped calling them collisions and started calling them accidents,” Carson said. “They’re not accidents. Ninety percent are due to driver mistakes. They can be avoided.”

They most certainly can be, and while Google and Uber may bank on autonomous vehicles in the long run, in the short run, entrepreneurs like Carson are paving the path for cloud-based automotive technology designed to make vehicles safer, more secure, and equipped with predictive analytics sure to make the future brighter than laser high beams.


Allen Taylor

About the author

Allen Taylor

An award-winning journalist and former newspaper editor, I currently work as a freelance writer/editor through Taylored Content. In addition to editing VisionAR, I edit the daily news digest at Lending-Times. I also serve the FinTech and AR/AI industries with authoritative content in the form of white papers, case studies, blog posts, and other content designed to position innovators as experts in their niches. Also, a published poet and fiction writer.

Add Comment

Click here to post a comment

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