This fact check offers a brief overview of the status of existing regulations, national and international activities in the field of autonomous systems with a focus on road traffic and gives an insight into the role of VDE and DKE in the (further) development of norms and standards.
What does autonomous driving mean?
“Autonomous” refers to the final stage of the development of automated vehicles and means the fully automated and targeted driving of a vehicle or autonomous system without human intervention. The person is then only a passenger. Autonomous vehicles that are networked with their environment can contribute to road safety and environmental protection by reducing the risk of accidents and saving CO2 because artificial intelligence (AI) optimizes driving style and route selection. Traffic searches for a parking space could soon be a thing of the past, lost time in traffic jams could be avoided and traffic volumes could be reduced. At the same time, comfort is increased, as the traffic is no longer constantly monitored and dangerous situations are recognized early on by the networked vehicle. Automation and networking therefore have the potential to make driving or transport in autonomous systems more efficient, safer, and more environmentally friendly.
How does autonomous driving work?
The technical prerequisite for automated driving is the interaction of various driver assistance systems in the vehicle that are built into one another. These driver assistance systems include, for example, braking, lane, turning or parking assistance systems. Complemented by sensor systems for environment detection such as laser, ultrasonic, radar and lidar sensors, moving objects and people can be identified and processed by the on-board computer. The electronics in the vehicle then ensure that the engine, brakes, and steering are electrically controlled after the data has been processed and that all active and passive (security-) systems in the vehicle are networked with one another.
Only through the increase in computing power and the further development of computer processors during digitization can the huge amounts of data in the vehicle and the cloud be processed by artificial intelligence (AI). While vehicles collect and save high-resolution maps (HD maps) in real operation, this data is transferred "over the air" to the cloud and converted into simulations by engineers. The technology is being further improved and enhanced through constant updates and virtual trainings.
The field of application of autonomous systems and vehicles is not limited to road vehicles such as robotic taxis and shuttles. Even autonomous trains, subways, ships and tractors in agriculture, as well as autonomous industrial trucks in warehouses are already in use today, but mostly with a special / exemption permit on public roads or in defined operating areas. In the future, unmanned electrical flight systems, so-called drones and autonomous flying taxis will also be added and have the potential to revolutionize urban and regional transport.
The goal of the German federal government is to make Germany a pioneer in the field of autonomous driving. Because in the future, too, the automotive added value should remain at home in Germany. Starting next year, with innovative legislative initiatives, Germany will offer the possibility of regular operation of autonomous vehicles, from which the German automotive industry and suppliers should benefit. But local public transport providers are also hoping for many opportunities as a result, e.g., as a supplement to classic regular services or feeder in the outskirts and rural areas.
Why is the VDE dealing with the topic?
The VDE (Association for Electrical, Electronic & Information Technologies) advocates the interests of German industry and science on a national, European and international level. The DKE (German Commission for Electrical, Electronic and Information Technologies of DIN and VDE) focuses primarily on electrical safety aspects and consumer protection. In the standardization committees, the experts discuss and deliberate on the issues of tomorrow and jointly discuss how Germany can assert itself as a business location in international competition. The complexity of autonomous systems lies in the large number of individual systems and the smooth interaction across various interfaces under the most adverse environmental and EMC conditions.
As an expert in the field of autonomous vehicles, Simon Burton speaks with DKE speaker Holger Lange, Senior Specialist at VDEIDKE, how artificial intelligence can proactively avoid critical situations in autonomous driving and what role functional safety plays in this. To the interview.
The question of the standardization and certification of artificial intelligence (AI) systems remains open. For this purpose, DKE and DIN (German Institute for Standardization) have published the “Standardization Roadmap Artificial Intelligence” (as of November 2020) on behalf of the Federal Ministry for Economic Affairs and Energy (BMWi). Particularly in the field of the certification of AI in autonomous systems, additional challenges arise due to continuous and almost uncontrollable updates. The experts of the committee DKE / AK 801.0.8 "Specification and Design of Autonomous / Cognitive Systems" deal with what such an AI seal of approval could look like and for this purpose have the "Specification and design of autonomous / cognitive systems" in the form of the first AI Standards developed in VDE-AR-E 2842-61-1.
When will autonomous driving become a reality?
Autonomous driving offers numerous opportunities and advantages in terms of climate protection, safety, and economic growth. Chancellor Angela Merkel also commented on this in 2017 with her view on the subject of autonomous driving: “In 20 years we will only be able to drive independently with a special permit. We are the greatest risk” aiming at the increased safety aspects in autonomous and networked vehicles. The former Google boss Eric Schmidt said: “The sooner we can have cars driven for us, the more lives we can save. Self-driving cars should become the predominant mode of transport in our lives”.
The technology organization VDE is also addressing this prevailing question in its new meta-study “Logistics, Energy and Mobility 2030”. The experts state the time until 2045, until fully autonomous driving, i.e., level 5, will be a reality in Germany and mention further potential in areas of application in logistics such as autonomous delivery systems.
Politics and business are working together on the nationwide introduction of autonomous vehicles in Germany. On the part of the economy, we are working on a continuous further development of the technologies and existing products with guaranteed reliable functionality and smooth interaction. Politicians are committed to harmonizing the international legal framework at the international level. However, to convince dying people of autonomous systems, uniform norms and standards are needed, among other things. This fact check has also shown this and identified further fields of application in which autonomous systems are already in use - such as industrial trucks, trains, ships, and drones DKE and VDE support these and other projects by offering a neutral platform for experts to bring the latest scientific findings and technological progress into the further development of standards and norms.