It’s almost impossible for today’s drivers to imagine, but it’s likely that in 20 years only a few road users will be sitting behind the wheel and steering their own cars. Instead, vehicles will autonomously head for their destinations thanks to a real-time combination of on-board computing and infrastructure and traffic data supplied by a traffic management system. Rules of the road, traffic signals, and road signs will be digitally transmitted to each vehicle. This is one of the scenarios that transport experts at Siemens are currently developing. Of course, no one can accurately predict when or to what extent it will be realized. “But one thing is certain,” says Claus Beringer, who is responsible for ‘highly-automated and networked driving’ at Siemens Mobility – Intelligent Traffic Systems, “transportation will be influenced by self-driving vehicles and will ultimately be characterized by autonomous driving.”
The Invisible Chauffeur
Self-driving vehicles are on the horizon. Autonomous taxis and minibuses could cover the last mile between homes and rail stations, and private cars might become superfluous in cities. It could all add up to vastly improved urban environments. Siemens is working with municipalities, public transit authorities, and research institutes to develop these transport concepts.
When Siemens installed the first traffic light at Potsdamer Platz in Berlin 92 years ago, no one imagined that it would replace the policeman directing traffic at the corner. Today intersections are interlinked and smart traffic management systems guide rivers of vehicles. In short, it took a long time for the current traffic picture to evolve. But according to Beringer, things will develop much more quickly as the technologies underlying self-driving vehicles mature. “Many players are currently developing technologies that can’t even be seen yet or can be seen only in research projects,” he points out; “but a lot will change very fast – and will have a disruptive effect – in just a few years.”
Five Stages to Autonomous Driving
Established automakers consider advanced driver assistance systems (ADAS) such as automatic braking and lane-keeping systems to be the first steps toward automated driving. In contrast, new players such as Google are counting on completely self-driving vehicles that will orient themselves without data transfers from the outside whenever possible.
In point of fact, according to the Society of Automotive Engineers (SAE International), there are five stages leading to fully autonomous driving. The first two are driving while using assistance systems, and partially-automated driving, where drivers usually steer themselves and perceive and assess the environment themselves. Starting with stage three, onboard systems handle most of the perception and recommend actions to the driver, such as braking if an obstacle is in the vehicle’s path. In the fourth stage, the system responds to events in the environment without requiring intervention by the driver. For example, it accelerates the vehicle when the road is clear of traffic. Finally, in the fifth stage, fully autonomous operation is achieved when everyone in the car is a passenger.
Toward an Information Infrastructure
The automotive industry is currently somewhere between stages two and three. Modern high-end to luxury class cars are already equipped with sophisticated assistance systems that can even brake independently when obstacles suddenly appear. But many current systems use cameras or radar, which experts don’t consider safe enough for fully self-driving cars. “What must be done now is for traffic management centers to transform information relevant to safety, such as road and traffic signs, which the driver – or a camera – currently sees, directly into data and feed this data into the vehicle’s management system,” says Markus Schlitt, head of the Intelligent Traffic Systems Business Unit at Siemens Mobility.
With this in mind, Schlitt and his team are developing systems that will allow vehicles to receive traffic and safety-related data, analyze it, and respond to it. As a first step, they have developed a communications platform (Central Management System) that can be used by Road Side Units (RSU) to exchange data bidirectionally between vehicles and the infrastructure. “RSUs can be connected to a traffic control center via the Central Management System (CMS),” explains Schlitt. “That would create a WLAN radio network to inform drivers in fractions of a second about traffic accidents, construction, road ice, or an approaching emergency vehicle.” The German automotive industry, research institutes, and Siemens are working on this program, which is called “Car2x-Netz.”
But who will compile and manage the data and ultimately control the traffic flows? “Those roles will primarily be played by cities and highway departments, because they often govern what are known as public-sector statutory functions,” says Schlitt. “Public transit agencies and rail operators will also participate to keep their mobility services attractive and competitive.” That’s why Siemens is already working with several partners on potential pilot projects, for example with the Singapore city administration and Deutsche Bahn (German railways) to develop implementation scenarios. Deutsche Bahn wants to get its passengers to its trains as conveniently as possible. In contrast, cities want to determine whether self-driving cars will soon be competing with public transit and taxis. “Imagine if your child could be picked up by a self-driving car and taken to school every day,” is the scenario Beringer describes.
All of this means that cities will be faced with new challenges. Since the most expensive part of a chauffeur service – the driver – will no longer be necessary, many people will be able to afford this new luxury, and they won’t be taking the bus or a subway. One result could be many more cars on the road, which would threaten gridlock. That’s one of the reasons why public authorities in both Singapore and Berlin are interested in using self-driving vehicles for public transit as well.
For example, local public transit could be made more attractive and cost-effective, thus minimizing the increase in the number of self-driving cars used for personal transportation. “Self-driving minibuses are the best way to close gaps in the ‘last mile’ between homes and the subway,” explains Schlitt. Public rail transport is already well ahead of car traffic when it comes to automated – if not autonomous – travel. Self-driving buses would be an ideal complement to driverless subway trains.
If a scenario of this kind comes true, public administrations will have a lot of new tasks to perform. Entire fleets of self-driving buses and mini-buses would have to be coordinated, while fixed routes would be eliminated in some cases, giving way to a more demand-based system. Charging stations would also have to be provided, because, as most experts agree, tomorrow’s traffic will be electric.
Taken together, such changes could pay immense dividends for cities and their inhabitants. Traffic jams and associated pollution would give way to cleaner air and quieter, wider streets since parked cars would be a thing of the past. Traffic-related injuries and fatalities would diminish as autonomous vehicles gradually replaced vehicles driven by humans. And everything would be more relaxed. It could all add up to a lovely utopia – one that has a good chance of coming true.