Data and the Connected Car: Part 2

The Car Is a Sensor

Christina Rux, product manager TSDP at WirelessCar,  says she and her team have been looking into how the connected car can be used to gather data about the environment. “We see that the car is a sensor, a moving sensor,” she explains. “It can act as a device for data collection.”

This is particularly useful in countries where harsh winters can cause extensive damage to road surfaces. “So we asked ourselves, 'How can you improve road conditions by using sensors for road data collection?'”

There is a tremendous opportunity in assisting road authorities, who today often simply drive around to examine the roads for damage. “Using the cars of drivers to gather that data is a much more efficient solution,” Rux says. “The cars would provide such data as vehicle speed, weather conditions, potholes in the road and chassis vibration.”

This solution could also provide real-time information on road conditions to drivers on the road, according to says David Green, product management director at Volvo Car Corporation, “This is the next development in connectivity, icy road detection,” he says. “You pass over an icy road in your car, and it then tells all other connected cars in the vicinity about it.”

This type of solution is imminent, says Jörg Lützner, Head of Services & Commercial Vehicles Division, Continental Automotive GmbH. “Vehicles will be providing such data as their position, weather characteristics, road characteristics, traffic characteristics and make it all available to the community in real time. The data will be gathered by the vehicle’s sensors and cameras and then aggregated with back-end data.”

Commercial vehicles are especially useful in this regard, he says. “Trucks are incredibly good input providers. They’re on the road 10 hours a day, much more than passenger vehicles.”

This will enable a vehicle to “look” significantly beyond the line of sight and provide drivers with a so-called eHorizon, which recognizes road events before the driver can see them.

With real-time data being communicated within a community of drivers, we are not far from system in which a connected car is able to communicate to all other cars in the vicinity where it is and what it is going to do, a prerequisite for the Holy Grail of connectivity, the autonomous car. But for that to happen, Volvo’s Green says, “the cars have to be on the same system, the same wavelength, so to speak. That will require industry standardization.”

But, here again, uncertainty reigns. As Green says, “We don’t know what is needed at the moment, and we really need to decide what the standards are.”

This will also demand some sacrifice on the part of car-makers, he says, because “it involves giving up some differentiation.”

 

Autonomous Driving

In the meantime, cutting-edge OEMs such as Volvo and BMW are refining the technology that will make the autonomous car a reality. The reason for building a car that drives itself is simple: “If you take people out of the driving loop, it will be safer,” Green says. “There will be fewer accidents.”

For its part, traditionally safety-conscious Volvo has pledged that there will be no deaths or serious injuries in a Volvo car by the year 2020. “And by 2017, we will have a fully autonomous car driving real people – not technicians – around Gothenburg,” Green says.

However, the road to actually having an autonomous car on the market may be longer than expected. “Legislation is going to dictate the journey,” he explains. “And it’s going to determine what level of autonomous driving is permitted on what roads.”

Ulrich Fastenrath, head of Trafffic & Routing at German car-maker BMW, agrees that the journey will be a gradual one. “The autonomous car is a long-term goal,” he says. “The transition from assisted driving to autonomous driving must be smooth.”

The stages along the way are categorized as “partially automated,” “highly automated” and “autonomous,” Fastenrath explains. He agrees with Green that highly refined mechanical systems make for better drivers than humans do. “Machines don’t get tired,” he says. “They don’t drink alcohol.”

Currently, while most connected vehicles can be described as partially automated,  some – the BMW Series 5, for example – have the capability to be highly automated at low speeds, as when they are caught in traffic congestion.

“In this case, the car takes more control of the car, keeping it in the lane and braking automatically to avoid a collision,” Fastenrath says. “However, according to the Vienna Convention of 1968, today’s drivers must be in control of their vehicles permanently. They have to be able to override actions of assistant systems any time.”

That caveat may be a long-term condition of the autonomous car since, as Fastenrath admits, “It will have a very low failure rate, perhaps 10^-8, but it will never be zero.”

This means that, ultimately, the relationship between man and machine will be turned topsy-turvy. Currently, with the highly advanced ADAS available in many luxury automobiles, the sensors, radars and cameras that assist the many safety features are what Volvo’s David Green calls “the last resorts.”

However, with the autonomous car, it will be the driver (or any capable passenger) who will be forced to intervene in case of a malfunction. The question then will be if, after relying on automation for so long, anyone will still have the necessary ability to drive a vehicle.

 

Using Data for Urban Routing

Autonomous driving is also about traffic management, Fastenrath says. “We are trying to develop mathematical models to manage huge fleets of cars. To make it work, we have to manage the infrastructure.”

So, BMW is also at the forefront of anticipating how data will be used in the future when most of the world’s population will be living in urban conglomerates and the task of getting from point A to point B will be far more complex than it is today, autonomous car or no.

One solution will be the routing of traffic based on accurate real-time traffic information (RTTI) sent to the vehicle and used for on-board route calculation.

“If you are given estimated time of arrival at your destination, you expect it to be correct, and not wrong by a factor of two,” Fastenrath says. “We have to push that to 100%. To get to their appointment or flight on time, customers want and need to be able to rely on their system 100%. To get that quality of service, you need to be able to constantly update predictive data.”

In the future, real-time traffic information will be pooled with parking availability probabilities and public transport schedules to create so-called intermodal routing systems to cope with the inevitable urban congestion, Fastenrath says. “So that you can seamlessly reach your destination with whatever means of transport available.”

BMW already has an intermodal system on the market in its i3 electric vehicle which, for example, combines data on available parking and data with available means of transport.

Strategic routing will be based, in part, on ”floating car data,” which enables systems to analyze the trajectories of vehicles and when they change from driving to other modes, such as parking,  Fastenrath explains. This requires data from a variety of sources, such as phones and other cars.

“A good service needs a certain minimum density of data,” Fastenrath says. “We will probably have to aggregate BMW data with other sources. Our data sources are increasing, but they may not be sufficient.”

Ultimately, Fastenrath says, urban road networks may be run by so-called cooperative (or load-balancing) routing, in which decisions are made based on the best way to get drivers to their destinations while keeping traffic flowing throughout the network.

Take the example of 100 drivers on the same road and heading for the same destination, and there are two reasonable routes to get there. According to Fastenrath, analysis of data from the traffic network will be used “to distribute traffic in an optimal way through the network.” This means, that of those 100 drivers heading for Point B, 80 may be directed via route 1 and 20 via route 2, to keep the traffic flowing.

 

The Thorny Issue of Data Privacy

Real-time dealer management, remote diagnostics, autonomous driving, cooperative routing and all the other connected-car services and solutions that are available today or will be tomorrow are all dependent on huge amounts of data, much of which will be personal. The accessibility of personal data is a continuous concern in both the United States and Europe, which has drawn up laws concerning its use and is currently looking to make those regulations  even more stringent.

The sensitivity of this issue was highlighted just recently when Apple announced, on Sept. 17, that data in its new software system, iOS 8, was so deeply protected that it would be impossible for it to comply with government warrants demanding personal customer information stored on Apple mobile devices. However, data stored in Apple’s cloud service did not benefit from the same level of protection.

Rux said that WirelessCar was “already following proposed EU data collection regulations” in which privacy demands are set at a high level by default. The EU defines personal data as “any information relating to an individual, whether it relates to his or her private, professional or public life. It can be anything from a name, a photo, an email address, bank details, posts on social networking websites, medical information, or a computer’s IP address."

The newly proposed EU privacy regulations also require that “valid consent must be explicit for data collected and purposes data used. Data controllers must be able to prove ‘consent’ (opt-in) and consent may be withdrawn.”

According to Rux, “OEMs are going to have to become more aware of data privacy and make it more transparent to the end user.”

Rux says that customers may be convinced to give up data if they are given some value or other financial inducement. She also notes that, while the new EU data protection law would apply to all EU member states, different cultures within the Union will have different approaches to releasing their personal data.

“For example, my mother is from East Berlin, and she would never give up her personal data,” she says. “She had bad experiences with that in the past.”

In addition, some of the customers using WirelessCar services are well-known and “they never give up their data,” Rux says.

There are also differences in the approach of generations to the protection of data. Telenor Connexion’s Andrea Sroczynski says, “The iPhone generation doesn’t mind being tracked.”

However, Apple’s recent move to make its device data much harder to access suggests that the “iPhone generation” just may be growing wary of having its data taken without permission and without getting anything in return.