Sevana: Key questions for the connected car ecosystem

Connected cars are already driving the machine-to-machine space, research firm Infonetics has concluded. Gartner predicts there will be a quarter-billion connected vehicles on the road by 2020, with new vehicles dramatically increasing the proportions of connected cars — and added that“the proliferation of vehicle connectivity will have implications across the major functional areas of telematics, automated driving, infotainment and mobility services.”

But there are still unanswered questions that must be addressed through research and development, regulatory agencies, and by automakers and consumers alike, for the connected car to be successful. Some of them include:

–How can the connected car be secured? This question is being grappled with across the industry, especially and is part of larger issues of “Internet of Things” security as wireless connectivity comes to different verticals. Graham Trickey, head of connected living for the GSMA, said that “the problem for the connected car is the different ways that attacks could be taken on — there are a lot of different systems in the car, and a … large surface area where they can be attacked.”

“As soon as you put vehicles on the network — or any device on the network — it becomes susceptible to security risks,” said Jeffrey Miller, IEEE senior member and associate professor of engineering practice at the University of Southern California.

Thomas Schultz, responsible for business development for Spirent Communications’ automotive unit, said he typically sees automotive OEMs take at least two different approaches when tackling security, often involving consultants checking architectures and simulating zero-day attacks, as well as manual hacking teams that double-check to see security measures are actually working. The ecosystem also needs to address consumer education on security, he added – people simply aren’t used to having to think about the need for a firewall for a vehicle.

“I think the challenge is also in making sure that the driver, and users of [connected car]services, are aware that there is no 100% guarantee for security,” Schultz said.

The most important thing, Miller said, is to make sure that vehicle control systems can’t be taken over, even if information for the car is able to be compromised. Hackers have demonstrated the ability to do just that, hacking a Jeep wirelessly earlier this year and prompting both a recall, a massive software update effort and a network-level fix by Sprint.

–What will be the role of the driver? When it comes to driver assistance and autonomous vehicles, Miller said he expects to see driverless vehicles on the road within the next four years — Google, for example, already uses them. But, he said, he doesn’t expect to see the removal of brakes and steering wheels any time soon, and does anticipate that having a licensed driver in the car will be required for the foreseeable future. A recent IEEE survey of members and on social media showed that even among a very tech-savvy audience, most felt comfortable with being in a driverless car themselves — but wouldn’t want their children to be transported by one. So drivers may be alerted by connected car technology and have to respond, or they may simply be along for the ride as autonomous features become more widespread.

The U.S. Department of Transportation outlines a number of scenarios in which drivers could be alerted to upcoming construction, workers on the roadside, and traffic conditions or if a nearby driver loses control of his or her vehicle — and then drivers can respond accordingly. Some features already exist, such as blind-spot and parking sensors. And autonomous driving features are hitting the road, if in limited deployment: Tesla’s most recent software update for its Model S, includes a number of autonomous driving or driver assistance features collectively called AutoPilot. They include AutoSteer, which engages Tesla’s Traffic-Aware Cruise Control; Auto Lane Change; side collision warnings; and Auto Park.

–What wireless technologies and networking techniques will be used for different use cases? Bluetooth, Wi-Fi, 2G, 3G, LTE, radar, near-field communications and GPS are among the technologies being integrated into the connected car, for a range of uses from radar’s use in vehicle avoidance and driver assistance features, to navigation and infotainment. Emergency connectivity is increasingly being mandated around the globe, and vehicle-to-infrastructure and vehicle-to-vehicle communications, such as 802.11p or Dedicated Short Range Communications, are still emerging and being standardized.

Trickey of the GSMA also noted that depending on latency requirements, some use cases may rely on cellular technologies and require routing back to the core, and others may need to be handled in a mesh-network fashion.

–How will liability and insurance be handled? Although driver-assistance features are being introduced and eventually, fully autonomous vehicles are expected to be able to reduce up to 80% of accidents that are caused by driver error, there will inevitably be situations where such vehicles are in accidents — and the ethics behind the programming that will decide a vehicle’s path is already being hotly debated.

In the IEEE survey, most respondents felt that either the car manufacturer or the developer of the vehicle’s software would be held responsible in accidents. Miller, for one, said that the detailed data available from connected cars is likely to help paint a clearer picture of accidents than human recollection and witness accounts do today, and that much like current liability arrangements, the circumstances of each accident are likely to determine fault.

“It’s not cut and dried,” he said. “It’s going to be based on the circumstances around the accident.”

–Where is the ROI? Brian Greaves, director of product development for IoT solutions at AT&T, said that services will have a significant role for mobile carriers in the connected car ecosystem. For example, split billing is something that AT&T has enabled — the ability for, say, drivers to be charged accordingly for streaming radio data, and auto manufacturers to be billed for telematics data that they do not wish vehicle owners to have to pay for. Trickey also said that split billing has been a particular area of interest among GSMA members. Miller said that along with infotainment options, direct marketing via connected car may have a huge impact: the ability to advertise to drivers based on their location.

For an in-depth examination of connected car trends, testing and technical challenges, read RCR’s new special report on the connected car and watch the webinar.

Image copyright: aimage / 123RF Stock Photo

The post Key questions for the connected car ecosystem appeared first on RCR Wireless News.