A vehicle that moves completely without a human driver: This is the vision behind the concept of autonomous driving. But what exactly does this much-discussed technology trend mean? In which direction is the development moving, and which possible applications and challenges face autonomous vehicles?

The diversity of autonomous vehicles

Although public debate regarding autonomous driving usually refers to autonomous passenger cars and HGVs, there are many more options for autonomous driving which is, in principle, possible for all means of transport. Even the Airbus aviation group is already considering small autonomous aircraft that can be used as air taxis in cities. This is not completely unrealistic as the autonomous autopilot has been a natural feature of aircraft for decades. Autonomous container vessels could already be used in maritime transport in just a few years from now. Land transport options include autonomous trains, and autonomous tractors on farms or autonomous drills in mining are also conceivable. The use of autonomous vehicles in all areas is expected to increase safety because many accidents are still caused by human error.

Autonomous vehicles in road transport

A similar scenario applies to autonomous road vehicles for transporting passengers, such as passenger cars or buses. They will make travel not only safer, but also more convenient, faster and efficient, for instance, by using travel time for activities other than steering the vehicle and by improving traffic flows in general.

Autonomous passenger transport can fundamentally change our entire view of mobility because it eliminates traditional borders between transport systems: An autonomous car can, in principle, be a private car, a taxi, a bus or a shared car.

The development of autonomous road traffic vehicles is taking place in several steps. The industry generally distinguishes between five stages representing the different levels of autonomy:

Stage 0: Driver only, the driver alone controls the vehicle.
Stage 1: Assistive systems support the driver (such as distance and cruise control).
Stage 2: Semi-automation of the vehicle (such as parking and lane-keeping assistants)
Stage 3: High level of automation, the vehicle changes lanes and performs similar functions autonomously, but the driver must be potentially able to take control and is prompted to do so when necessary.
Stage 4: Full automation, the vehicle steers automatically, the driver must intervene in an emergency only.
Stage 5: The only human action is to start the vehicle and to select the destination – the vehicle does the rest automatically.

Up to and including stage 2, much of the above is already reality today and available in new car models. Stage 3 could already be reached in a few years, albeit initially only on motorways and not in cities or on main roads. Vehicles of stage 4 and higher, which can in fact be truly called fully autonomous, are still in the initial phase and are currently tested on several test tracks in Germany.

Challenges for autonomous driving in road traffic

However, it will still take several years until autonomous vehicles will be used nation-wide because several technical and legal challenges are yet to be mastered.

Technical issues mainly concern three areas, i.e. detection of surroundings, orientation as well as driving, braking and steering. Steering is the easiest of these problems because automatic steering as well as automatic aligning and braking systems are already in use at least in stage 2 vehicles. Detecting surroundings in the open air, however, is much more difficult because the functioning of sensors must be ensured in road traffic in the sunlight, when it rains or at night. This is why several systems are used, such as 3D cameras, radar sensors or laser scanners.

Technical obstacles must also be overcome when it comes to orientation. This requires high-definition maps which are much more detailed than the GPS systems currently available. Furthermore, the vehicles must be able to communicate with each other (Car2Car communication), for instance, in order to warn against dangerous situations. Another area is Car2X, i.e. communication with the environment, such as traffic lights or level crossings.

The legal issues that arise from autonomous driving are another challenge. The Vienna Convention on Road Traffic that was adopted in 1968, for instance, provides that drivers must at all times be able to control their vehicles. A revised version of the Convention came into effect in 2016 and opened the door for systems that are capable of independently controlling a car on condition that they can be deactivated or overridden by the driver at any time.

Another legally relevant question concerns liability in the event of an accident. Can drivers be held responsible even though they did not steer the vehicle? Or is the manufacturer of the car responsible? Another topic of intense debate concerns moral issues, so-called dilemma situations. If a collision cannot be avoided, should the vehicle drive into a group of pedestrians or divert into oncoming traffic, for example?

Autonomous systems in production and logistics

Whilst it will take years or even decades for autonomous vehicles to become a reality in many areas, the use of independently moving robots or machines has a longer tradition in parts of the economy.

The beginning was marked by so-called driverless transport systems which were used for the first time in the 1950s and 1960s in the US. Driverless transport systems, for instance, in the form of automated fork-lift trucks, are today used to transport materials at logistics hubs, such as ports, distribution centres and production facilities. Special sensors and software enable these mobile transport robots to automatically find their way. The advantage: These machines are permanently ready for use and can therefore boost productivity significantly. However, these robots are used primarily in enclosed and confined spaces and, for safety reasons, in areas where there is at present no or only a limited risk of coming into contact with humans.

Examples of these applications are the transport system of the company Bär Automation at the Audi R8 factory in Neckarsulm or the automated Drytower warehouse system from the company Totech GmbH which are both presented under the ‘Solutions’ topic on this website.