IoT Blog
IoT Blog

How Sierra Wireless Helps NetraDyne’s AI Make Roads Safer for Truck Drivers

by Benoit Tournier, Director of Marketing, Transportation

The fusion of artificial intelligence and the Internet of Things has moved beyond theoretical discussions and demonstrations, and innovative companies are introducing some amazing solutions that allow connected machines to vary the outcome of a situation or behavior based on knowledge or observation. NetraDyne, for example, is using video-based analytics to capture the most complete view of the driving environment and make roads safer.

You might not realize this, but there are over 3.5 million truck drivers in America alone. It’s one of the most important job sectors for a large percentage of American workers, and they work hard. The average truck driver puts 2000-3000 miles under his wheels every week—and every single one of those miles brings with it the risk of an accident. A car could merge too quickly, causing the truck to jackknife. A driver with thousands of miles behind him could drift off or start driving too fast, pushing to meet a deadline. And for fleet managers and owner-operators, these risks bring with them enormous costs. 

In 2015 alone, the USDOT (United States Department of Transportation) and NHSTA (National Highway Traffic Safety Association) estimated that automotive accidents cost $242 billion, including $57 billion of workplace productivity—and each accident involving a semi can cost shipping companies up to $74,000. Astronomical figures like these only underline how important shipping is to the US economy. Trucks carried over thirteen billion tons of freight last year, and that number is expected to rise by another ten billion over the next two decades. All of this means that we will need more trucks on the road—in an industry that is already facing a shortage of nearly 50,000 drivers. 

This raises a problem: how to ensure safety and prevent accidents from happening.

Many fleet managers and owners are turning toward innovative technology. OSHA (Occupational Safety and Health Administration) recently recommended that fleets employ a telematics system to report information on driver behavior, including speed and safety. But some systems are more advanced than others. For example, NetraDyne Driveri™ system’s enhanced video capabilities, which uses game-changing artificial intelligence to record nearly unlimited interactions, analyzing issues and instantly reporting back to the cloud—and to waiting fleet managers. These capabilities, underpinned by Sierra Wireless’ cellular modules, are helping to minimize the long-standing risks of the trucking industry. 

Case Study - Blog CTA - NetraDyne

Instant Video Analysis Through Artificial Intelligence
Human intelligence can understand the dynamics of the road, of course, but it has its limits. Previously, the only tools fleet managers had to prevent accidents were safety policy and training. Managers obviously couldn’t be everywhere at once, and if an accident happened, they had to rely on the driver’s report, plus eyewitness or police statements if any existed. And there was no way to tell if drivers were narrowly avoiding accidents. Behavior could only be judged after something went wrong. There were no preventative measures.

More recently, shipping companies have begun using video equipment that turns on when brakes are slammed or the steering wheel is jerked to record the circumstances surrounding an incident. But this system relies on basic machine intelligence: it only snaps to life when something goes wrong, which means that it lacks context and doesn’t help with prevention. You’re able to see video of the truck swerving off the road, but the camera might not start until the car that cut the truck off has already gone off-screen. In other words, you’re still not getting the complete picture. 

That’s where NetraDyne’s Driveri system comes in. It doesn’t just record video. It is the first system to use video sensors paired with artificial intelligence to actually see what is happening on the road and report these findings to a central cloud, where they can be analyzed, reviewed, and acted upon within minutes, giving feedback to the driver to avoid an accident. 

Hypothetical One: the Swerving Car
Suppose that a truck driver, rounding an S-Curve on a highway, runs off the road. His manager assumes he is at fault, but the Driveri system shows that a car cut in front of the rig, veering from lane to lane before trying to squeeze in ahead of it. The system judges that the truck driver was traveling an appropriate speed and behaved as he should—avoiding other cars in the vicinity and preventing a worse accident. By analyzing everything on the road, the fleet managers has a clear understanding of what happened and can avoid penalizing the driver.

Hypothetical Two: Going too Fast

Or suppose a driver is slightly behind schedule, having hit unexpected traffic near Kansas City. In order to make up for lost time, he speeds up, changing lanes with more frequency and trying to pass in conditions that aren’t conducive. Analyzing the road conditions, the amount of traffic, and even the weather (it had started to rain), Driveri sends an alert of potentially unsafe driving back to the cloud, which performs AI-enabled analysis and relays a message to the driver on how to handle his increasingly unsafe driving conditions. This all happens within minutes, heading off a potentially critical situation.

The Impact of Instant Analysis
This kind of instant analysis brings benefits for managers and drivers alike, including:
  • Real-time alerts on driver behavior
  • Real-time alerts on road and weather conditions
  • Smart, proactive guidance to drivers in dangerous conditions
  • A way to reward good behavior and not punish faultless accidents
  • A way to keep drivers accountable for accidents that are their fault

Of course, doing all this isn’t easy. The need to upload both video and data requires a constant, stable, and powerful connection between every truck in your fleet, the cloud, and your fleet management team. 

How Sierra Wireless Powers Driveri
In order for the Driveri system to work, there would need to be a video uplink in every truck, capable of instantly transmitting large video files and AI-generated analysis to an accessible cloud from anywhere in the country. This would require a 4G LTE network—10 times faster than 3G—since any delays would make the system less effective. The situation was further complicated by the need for constant communication between mobile and stationary objects.

That’s why NetraDyne turned to Sierra Wireless AirPrime WP Series modules, which provide continuous, high-bandwidth connectivity. These modules accommodated the transfer of analyzed video footage from device to cloud, then from cloud to driver in the form of actionable alerts. Wideband services allowed for more audio and video files to transfer, which was key for surveillance, and 4G LTE kept data considerably more secure—no small matter to companies looking to protect proprietary information. 

Because Driveri uses a Dedicated Artificial Intelligence Deep Learning TeraFLOP processor that needs vast computing capabilities, it required a 4G LTE system to produce instant, actionable intelligence. Sierra Wireless technology enables it to pull on the full power of LTE to upload videos for real-time analysis. As NetraDyne President Sandeep Pandya puts it, “We couldn’t function without the video uplink provided by the Sierra Wireless LTE modules.”

This technology will prevent accidents. It will minimize driver error. It will make up for driver shortfalls by turning new drivers into safe drivers. It will save money caused by productivity harming accidents, helping keep America’s trucking industry strong. It’s just one of the ways Sierra Wireless is teaming up with leaders in shipping and transportation industries to keep the roads safer for everyone. Start with Sierra to learn more about how our innovative wireless solutions can help your company prepare for the IoT revolution.