Connected Cars Will Save Drivers Time, Money and Strife

Connected cars, vehicles connected to the internet, other vehicles, and infrastructure, hold the promise of reducing traffic and pedestrian fatalities in addition to speeding up response times of emergency personnel like police and fire trucks.

Dozens of companies are working on the different technological and regulatory hurdles, but all towards the same goal. As with hands-off driver assist technology, government regulations and court rulings are struggling to keep up with the pace of evolving technology, though one government body took a big step forward in April.

The Federal Communications Commission (FCC) granted a joint waiver request in April (ET Docket No. 19-138) that allows automakers like Ford, Jaguar Land Rover and Audi, departments of transportation in Utah and Virginia, and private companies to deploy cellular Vehicle to Everything (C-V2X) technology.

"The FCC waiver is a tipping point in the connected vehicle transportation industry. From our perspective in the business community, it's huge because it allows us to invest confidently in C-V2X technologies knowing that there's going to be spectrum and regulatory certainty that we can get this technology deployed," Bryan Mulligan, president of Applied Information Inc. told Newsweek.

C-V2X encompasses the technologies that connect cars with the world around them including traffic lights, construction zones, emergency vehicles and each other. Audi is testing such technologies now in school zones Georgia, as is Stellantis, formerly Fiat Chrysler Automobiles. When something arises like an ambulance or school bus, a notation comes up on the infotainment screen, telling drivers to slow down or be aware.

Mulligan explained that there is not one silver bullet that is going to magically solve everything like the industry thought in the '80s and '90s. Now there are a lot of corresponding technology layers, starting with local sensors on the cars, which act as short-range protection.

"Because safety is in layers, lidar, for example, is in the group of short-range sensors. There are radar, LiDAR, and vision sensing. And all of those have a role," Mulligan said.

"The long range is the cellular network (Cellular-V2X) that is transferring information so that you can reroute people around roadworks, and you can do all sorts of safety things by making a longer-range enunciation of danger. All those complementary technologies are needed," he said.

Each layer has benefits and drawbacks. Short-range sensors like cameras and radar do not work when they are obscured by snow, ice, or dirt, and need to be cleaned regularly. Dirty radar sensors will either pick up an object in the road late or not at all. The medium-range DSRC is for car-to-car and some car-to-infrastructure applications, but those need direct line of sight to work best. It also places most of the design/installation/maintenance infrastructure costs on the sanctioning body, usually a DoT or state/municipal government.

Cellular V2X is the longest range as it works on the cellular signal that also facilitates phone calls. The drawback there is when a driver is out of cellular range.

Applied Information's Glance cellular-V2X software can connect school-zone signs, traffic signals, school buses and emergency vehicles. The signs and signals can be adjusted to show lower speed limits or late bus information when necessary due to weather or other problems.

During real-world experimentation in its Alpharetta, Georgia test city, bus drivers save time by getting green lights along with a 10 percent savings in diesel costs. And the connected public is more informed of where and when school zones are active.

"You only need one in five cars in the line to travel at the right speed and you've lowered the speed in school zones, and you've made a safer environment for the kids. The crashes that do occur are much more survivable," Mulligan said.

Not every car needs to be connected

Wejo, a Manchester, England-based company, has been connecting cars, collecting, and analyzing randomized car data since 2014. Founder and CEO (Chief Executive Officer) Richard Barlow explained that in addition to traffic new use cases for connected vehicles are coming out all the time. He agrees that a company doesn't need to connect every car on the road.

"You don't need all the data all the time. If you think about safety, when we're getting data from 5 percent of all vehicles in New York to monitor traffic flows, you don't need more than that. You don't need any more cars to be connected," Barlow told Newsweek.

Those connected vehicles provide data that will lead to smarter vehicles, infrastructure, and cities. Research from the U.S. Department of Transportation has found that the technology could reduce unimpaired vehicle crashes by 80 percent, while also reducing the 4.8 billion hours that Americans spend in traffic annually.

American drivers also spend an average of 17 hours per year looking for parking, resulting in a cost of $73 billion in time, fuel and emissions. Connected cars can make that easier by using mapping data. Israel-based Otonomo, like Wejo, does the aggregating, cleansing, reshaping, and enriching connected car data so customers can find insights on parking and dozens of other parameters. Its real-time mapping data is helping the smart parking market, which is predicted to be valued at more than $16 million by 2028.

Connected vehicles are also more than Google Maps or Waze. Those usually get used, Barlow tells Newsweek, only when navigating to an unfamiliar destination, and on average less than 10 miles per month. It also uses machine learned outcomes to estimate where congestion is. Connected cars produce hard data. Some of that is collected and analyzed by automakers, others use companies like Wejo to summarize the information.

"Most manufacturers have some form of cloud that receives the data in a raw environment. There's no industry standard so most manufacturers just leverage the mobile carrier's network, 5G or LTE. The vehicles are preprogrammed when they leave the factory to send unstructured data," said Barlow.

"We usually get the data like that, and then we apply our data models, where we translate all our OEMs (original equipment manufacturers) different profiles of data. We work with 28 OEMs and Tier 1 partners, and we translate all that data to look the same. The outcome, whether it's for data or insights, isn't standardized for these use cases."

To translate that to safety for pedestrians and cyclists, Wejo uses lane indexing. It doesn't just get location from its vehicles; it can get steering angle change or suspension or height change. It can get parking sensor and antilock brake system info. It knows when a vehicle is going around the corner, or when a vehicle isn't changing lanes. That's another thing that can't be done with mobile phone data because the location for lane indexing needs to be within inches, not feet.

"We can be accurate because even if the GPS is inferior quality, we know the vehicle is still turning because we're getting the steering wheel data. So having that lane index, we're working with smart cities to help them understand the flows of traffic. We're helping them understand pedestrian flows too with a product called Meta Twin. It's a 3D environment where we can simulate cyclists and pedestrians crossing the road and how vehicles would behave in that in that situation," Barlow said.

"And that's because we've received so much data, billions of miles, more than Tesla, more than Waze. We've seen vehicles making emergency stops at the traffic lights. We've seen vehicles coming to pedestrian crossroads. We've seen a vehicle swerve and then imply that they're avoiding a cyclist. We've learned so much from 21 trillion data points. Now we can work with smart city planners to help them understand how to make our roads safer for pedestrians and cyclists."

One company, that Wejo chooses not to name, says it was 30 percent more efficient getting live data of traffic flows compared with its old system. That's a major use case for emergency responders that want to avoid congestion and get to a scene quicker.

"So that's one use case. And then the other use case is what emergency responders should be looking for to know whether they should be responding to an incident in the first place. If it is a low-speed fender bender, then you don't need the police to go and close the highway or to slow down the traffic flow. You frankly need a recovery vehicle, and you need to tell the driver to move the vehicle off the highway," said Barlow.

Applied Information has another test site in Marietta, Georgia, with connected fire and emergency vehicles that can talk to the traffic lights when responding to a call. But those V2X connections need to take place far away from the signal, giving time for pedestrians and traffic to clear.

"The fire truck sends a signal to the traffic intersection to say I'm coming, and I've got my lights on and I need a green light. A light might typically take 30 seconds to safely turn because you've got to get all the pedestrians out of the way. That's where the long-range technologies come in. Marietta for example, hasn't had a crash between a member of the public and an emergency vehicle in the last six years since doing this," said Mulligan.

"Then your Audi or your Travel Safely app or Waze app picks up and says there is a fire truck approaching from behind. And that is a layer of safety that is brought to bear by connected vehicles."

The downside to all this connectivity is security, privacy, and storage. Hackers have already shown they can get into vehicles' systems and control almost everything including the steering, throttle and brakes. Automakers can usually fix these software issues as soon as they pop up through over-the-air updates, but it is a growing concern.

Upstream, a provider of data-management and cybersecurity solutions, just published its fifth annual study looking at more than 1,173 cybersecurity occurrences over the past 13 years. Charging stations only made up 4 percent of the occurrences while application programming interfaces (APIs), which allow communications between software components, were responsible for 12 percent, a 380-percent year-over-year rise.

Another problem is the sheer amount of data being sent. Between the standard radar, LiDar, camera and ultrasonic sensors, not to mention the motion sensors for the steering wheel, brakes, suspension, an average car can log up to 19 terabytes of data per hour. A 4K camera alone can record 5.4 terabytes per hour. Over a year, a vehicle could send more than 5,000 terabytes of data (a terabyte is 1,000 gigabytes).

That data must then be stored. In 2021, total global data storage capacity was 8 zettabytes (ZB) and is set to double to 16 ZB by 2025. Vehicles collecting the maximum amount of data would fill that first 8 ZBs up in two years.

Still, according to the connected vehicle companies, there's reason to be optimistic.

"This is just the tip of the iceberg," said Barlow. "We are eager to see how telecom operators and car manufacturers can unleash the power of further crossing mobility, infrastructure, networks, and city data together. Unlike V2V, which depends upon enough vehicles talking to each other, V2X creates benefits from day one."