IoT Smart Roads: Paving the way

Forget for a moment the headline-grabbing testing of autonomous vehicles by Google and other high-tech companies and think instead about a more down-to-earth innovation: a smart highway that demonstrates the near-term potential to improve road transportation – IoT Smart roads.

In 2013, the Austrian, German, and Dutch transport ministries agreed to work cooperatively to create a long highway corridor spanning the three nations – from Rotterdam to Vienna – that would test and implement cutting-edge safety and traffic management technologies.

The first results came in during 2016 and 2017, showing promise in the smoothing of traffic flows and the reduction of holdups, among other things. Best of all, the technologies provide travel benefits whether the driver is a sophisticated computer system or a human being.

The Cooperative Intelligent Transport Systems (C-ITS) Corridor, Joint Deployment has the potential to increase transport efficiency, improve road safety, and provide environmentally friendly mobility. Its supporters also believe it will create additional services and new business models. The system uses a set of international specifications that provide the technical basis for the deployment to bring benefits for any vehicle using the road, Vital systems The real revolution in road transportation will come through not just the specially equipped vehicles used to demonstrate more advanced features of the scheme.

The C-ITS project shows autonomous, self-driving vehicles may have garnered most of the attention but the real revolution in road transportation may come from making the underlying road infrastructure and the wider environment smarter through applied IoT principles. For example, roadways that are aware of issues, such as traffic conditions, icy surfaces, or the presence of wildlife, can help individual vehicles to travel more safely and smoothly while providing system-wide management to ensure optimal routing and traffic efficiency for all.

Vital systems: The real revolution in road transportation will come through the application of IoT principles.

Infrastructure must be part of the conversation

Autonomous vehicles benefit, too, not only because of the additional situational awareness smart roads can provide but also because self-driving vehicles with no connection to local traffic information can’t contribute to the improved efficiency that most roads badly need.

“With autonomous vehicle development, much of the attention has been given to how clever and independent the vehicle can be,” notes Ian Hughes, senior analyst for the Internet of Things at 451 Research in the UK. However, each vehicle is part of a wider network of links involving both vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications. “If an autonomous driving system was being built ‘greenfield,’ without any reference to existing systems, everything would be designed to interact for efficiency and safety. As we transition to autonomous [vehicles] certain hubs of infrastructure will upgrade, while others lag and require vehicles to be independent,” says Hughes.

Just as toll roads offer a better service for a price, we may see some of this infrastructure appear for premium use. For example, a motorway might accept autonomous vehicles using specific protocols to keep them all informed and traveling safely at speed until they break away to their destination, he explains.

“Smart cars will become more integrated with national intelligent transport infrastructures and systems such as satnavs in cars, traffic signal control systems; parking information, weather reports, bridge de-icing, container management systems; variable message signs; automatic number plate recognition; or speed cameras to monitor applications, such as security CCTV systems and similar,” predicts Kevin Curran, a senior member of the IEEE and professor of cybersecurity at Ulster University. He explains that ultimately the road beneath us would become more communicative with smart cars via embedded road sensors which could be turned on during preventive road construction maintenance or in emergency conditions, alerting every vehicle to the need to reduce speed or halt. “With such a technological advantage, we can expect the end of roadside cones demarking construction and traffic flow,” he says.

Other glimpses of future intelligent infrastructures beyond the C-ITS example can be seen in initiatives such as one in Birmingham in the UK, where IBM is helping with the use of Big Data analytics to understand parking patterns in order to manage congestion better. In the Birmingham system, IBM has deployed ultra-low power wireless sensors in roads and provided an accompanying app for drivers to get real-time availability and prices for parking. Common input sources for managing traffic include road sensors, video cameras, and GPS updates from public transport. Huawei recently demonstrated smart technologies employing coordinated vehicle grouping, vehicle-intersection management, and lane-change regulation. Curran warns that the Internet of Vehicles (IoV) poses stricter latency and reliability requirements for the era of 5G networks.

“On motorways a latency of one millisecond equals an approximate driving distance of three centimeters and only fast networks, such as 5G networks, can support millisecond level latencies, thus ensuring safe driving,” he argues. The general approach of automotive OEMs so far has been to march ahead with technology enabling vehicles to talk to other vehicles (V2V) but without incorporating connectivity to the cities and their infrastructure (V2I), notes Mark Zannoni, research director for smart cities and transportation at IDC Government Insights. “You don’t see carmakers coordinating with governments; they are just going along on their own,” he says.

The autonomous vehicle outlook

If you have a city-based smart infrastructure more of the benefits of autonomous vehicle development swill become evident, such as the improvements in safety.

Right now, autonomous vehicles can sense other cars around them and when a car ahead applies its brakes yours will, too, Zannoni says. That improves safety because it is estimated that 95% of vehicle accidents result from driver error. However, thecae of a child darting into the road unexpectedly may result in an autonomous vehicle applying its brakes quicker than a human would but it may still hit the child. In other words, it is not a big change from the statuesquo.

“If you had city-based and road-based smart infrastructure, not simply technology in the car, it would sense the presence of the child, calculate the child’s possible path, and communicate that to the vehicle, helping to avoid an accident,” Zannoni adds – and that kind of development would have a profound safety benefit.

Along more rural roads data relating to parameters such as atmospheric temperature or the temperature of a bridge deck would also improve safety and some of that could come from V2V communication.

“If a vehicle hits a patch of ice it can share that data through edge computing or the cloud to warn other cars or send a message to maintenance crews,” says Zannoni.

Of all the key technologies needed to aid the progress of autonomy, getting accurate vehicle location information is perhaps the most important. GPS in cities isn’t accurate enough, according to Zannoni, due to “urban canyoning,” the blocking of GPS signals in built-up areas. To try to overcome this, peer-to-peer sensors were set up in New York’s Manhattan district, with funding from the US Department of Transportation, and the result was improved locational accuracy to a tolerance of two centimeters; much better than with traditional GPS.

Zannoni says that, while autonomous technology is already a reality, it will be a while before we see a society of autonomous cars. “A car can operate by itself but, when you put it around other cars or involve traffic police or pedestrians, its performance has problems,” he says.

The challenge therefore is to get it to work successfully in society. Zannoni explains: “That’s why I think it’s vital to have infrastructure that can communicate with autonomous vehicles and deal with congestion.”

Road smarts Traffic infrastructures will eventually need to be in constant communication with autonomous vehicle.

The smart money is on infrastructure

The infrastructure will become a key element in the operation of autonomous vehicles, agrees Walter Sullivan, who is head of the innovation lab at Elektrobit, a maker of embedded and connected technology for the automotive sector, based in Erlangen, Germany. The potential for smart infrastructure is tremendous, he believes – from efficient distribution of traffic through the road network, to optimized, automatic notification of breakdown service vehicles, to improved response times from first responder services. With the help of smart infrastructure, for example, paths and roadways can be cleared and roadways can be coordinated with public transit.

“The scenarios are bounded only by imagination and ability to invest,” says Sullivan, adding that making the transition to more capable mobility services in urban planning means the infrastructure we currently put aside for parking of vehicles can be reconsidered and perhaps reduced. A key issue will be cost-effectiveness. “Infrastructure which is meant to cooperate, or interact, with autonomous vehicles will naturally be built out where autonomous vehicles are concentrated – and that is unlikely to be in rural areas,” he says. “It really comes down to utilization and cost-benefit.”

Street cab named desire could be the key

Given the role of smart infrastructure in powering the future of road transportation, what is the realistic outlook for autonomy?

“We probably have a similar view to commercialization as many others do. Autonomy has the opportunity to be introduced into targeted vehicle fleets operating in geographically restricted areas at first,” Sullivan replies. The kinds of fleets he believes will be targets for autonomy would typically be operated by mobility, cargo, or transportation companies, and have relatively high utilization rates. That, in turn, provides more justification for the costs needed to implement an autonomous system, which will take us into the first half of the 2020s and will include L4 and L5 levels of autonomy, he predicts.

“In parallel [with targeted vehicle fleets], we’ll see increasing L2 and L3 implementations reaching commercialization in traditional passenger vehicles,” Sullivan says.

As we progress through the decade more people will have access to fully autonomous vehicles through different mobility services that they might use, such as rental vehicles or even taxi services like Uber or Lyft. Along the way, consumers will gradually develop confidence and comfort in the vehicles’ capabilities, he says.

In the second half of the 2020s and early 2030s, Sullivan predicts we will see L4 autonomy introduced into cars which are more in the mainstream of those vehicles consumers can buy. “There may be early versions of L4 systems introduced into high-end vehicles in the latter half of the 2020s but they are not likely affordable for the typical car buyer. The industry still has some tough challenges to overcome, however at this time the focus and progress of the industry are rather promising,” he says.

Deployment of self-driving vehicles does have risks. “We are still working on perfecting some of the foundational technology, from the sensors to be used, to the algorithms, and AI that will act on the data our sensors give us,” explains Sullivan. As an automotive industry software company helping carmakers to scale up from traditional driver assistance to L4 and L5 autonomy, Elektrobit is, in general, optimistic about the future.

“I believe that from a technology perspective, we will solve those challenges and the risks therefore will be more related to acceptance, regulation, security, and demonstrated safety improvements,” predicts Sullivan.

Safety first for self-driving breakthrough

Is the public really interested in autonomous vehicles or do they just want safer ones? Sullivan reckons the answer probably depends on how the products are presented. If people can have point-to-point transportation which is safer, cheaper, and more convenient than driving their own car, then the value proposition would be compelling, he believes.

The experience for most people in the first years of autonomy will be through ride-hailing and sharing. Sullivan’s view for those attracted to a ride hailing service is that it will be a safer and probably cheaper alternative offering the same or a better quality of service.

Ken Philmus, senior director for global business development at Conduent Transportation, agrees. “I’m confident that autonomous vehicles are inevitable but full integration is clearly not in the near future as there is a tremendous amount of work to be done to gain consumer acceptance,” he says. There are over 40 US companies testing different prototypes of autonomous vehicles yet standards to allow them to work with one another are just beginning to be developed at the federal level. According to Philmus, “as integration moves forward there may initially be separate lanes for autonomous vehicles or even a focus on autonomous freight movement.”

Smoke without fire is the main obstacle
Full autonomous vehicle implementation can’t be achieved without the ability to obtain information about road and traffic conditions. Along the way, traffic management as we know it today will be very different. For example, traffic management centers will become communication centers that not only share information with each other but also with vehicles in “absolute real time,” according to Philmus. This will require massive infrastructure changes.

Although the technical challenges can be mastered, drivers will be the greatest hurdle. “People who simply enjoy driving (…) will see autonomous vehicles as taking away their personal freedom,” he says.

Drawing an analogy between driving and cigarette smoking, Philmus adds: “Despite all the clear evidence that not smoking will prevent illness and death, millions still do and seemingly will always continue to smoke. I’m confident that we will see the same types of resistance at the various implementation stages of autonomous vehicles.”

READ MORE ARTICLES