Autonomous driving is one of the major developments of our time. It is about much more than technical progress. Safety and responsibility are key issues in the major transformation of the automotive industry. We discover more about this from Prof. Dr. Johannes Betz from the Technical University of Munich in an expert interview.
Prof. Dr. Johannes Betz holds a position as Professor of Autonomous Vehicle Systems at the Technical University of Munich and focuses primarily on the development of new algorithms that enable dynamic trajectory and behavior planning, adaptive control and continuous system learning. The aim is to develop the next generation of intelligent, autonomous vehicles whose functions enable safety and trustworthiness.
The integration of ethical and responsible behavior is also very important to Johannes in this context. In an expert interview, Johannes therefore not only showed us the technical challenges of this major transformation in the automotive industry, but also the ethical responsibility that autonomous vehicles will have in road traffic.
Prof. Dr. Johannes Betz
Alexa Lohrer: Hello Johannes, thank you for taking us along on the subject of autonomous driving and introducing us to the context! Autonomous driving is being talked about as the big transformation in the automotive sector. Where do we stand on this? What are your current challenges?
Prof. Dr. Johannes Betz: We are currently in an exciting phase in the development of autonomous vehicles. Technologically, we have made significant advances in sensor technology, data processing and software, enabling vehicles to accurately sense their surroundings. However, one of the biggest challenges is integrating these technologies into everyday driving scenarios, where the predictability of human behavior and unexpected road conditions play a major role. We are also working on scaling the technology to make it safe and reliable for the mass market. There are already some manufacturers that enable autonomous vehicle functions without human intervention, such as Mercedes Benz with DrivePilot or driverless cabs from Waymo in San Francisco.
Are the challenges in companies different from those in research? How far have car manufacturers come in this area?
Johannes: In the industry, the challenges of autonomous driving focus primarily on market launch, cost efficiency and compliance with regulatory requirements. Car manufacturers are working hard to optimize their autonomous driving systems for the mass market, with ensuring the reliability of these systems under various conditions playing a key role. They are faced with the task of developing technological solutions that can not only be produced efficiently and cost-effectively but are also capable of ensuring safety and performance in all driving environments.
What role does the development of AI play in this context? Which AI systems are used to really capture the entire road traffic, process information and react accordingly?
Johannes: Artificial intelligence is the backbone of autonomous vehicle systems. We primarily use deep neural networks, which make it possible to process large amounts of sensor data in real time. These systems are not only able to recognize objects such as pedestrians and other vehicles, but also to interpret complex scenarios and make predictive decisions. The challenge is to train these systems so that they react safely even in unexpected or rare situations and offer a high level of reliability.
Where are we heading? In your opinion, what are the next milestones that can be achieved through AI and technical advances? And what are the potential uses for autonomous driving in the end? Is it more in the area of services, i.e. cabs or passenger transportation? Is its use in the transportation of goods more realistic?
Johannes: The future of autonomous vehicles looks promising and the next milestones in this area will be shaped by advances in artificial intelligence. A key milestone will be the full integration of autonomous vehicles into urban traffic, ensuring seamless interaction with human drivers and pedestrians. The networking of vehicles (V2X) with each other also plays a major role here, enabling more efficient traffic flow control and helping to reduce the number of accidents.
In terms of possible uses, I see great potential both in the service sector and in the transportation of goods. In the service sector, autonomous cabs and shuttle services in urban areas could reduce the need for private vehicle ownership and reduce traffic density and environmental pollution. Their use in freight transportation also offers considerable advantages. Various manufacturers are currently developing autonomous trucks that can then drive autonomously on the highways. Autonomous driving could enable logistics companies to increase their efficiency, for example by reducing operating costs and shortening delivery times.
In the context of autonomous driving, the ethical responsibility lies in the protection of human life. How can you be sure that an autonomous vehicle protects human lives? Are there any other aspects that are relevant to the ethical and responsible implementation of AI systems in autonomous vehicles?
Johannes: Safety always comes first. Autonomous vehicles are designed to choose the safest course of action in every situation. This includes the vehicle adopting a conservative driving style when in doubt. We are constantly working on testing our algorithms against all possible situations and developing them further. Ethics in AI is also an ongoing topic of discussion in development, particularly when it comes to decision-making under uncertainty and handling user data.
Are AI systems from the field of autonomous driving also used in other areas? Or vice versa, are AI systems from other areas also used in autonomous driving? Are there mutual learnings?
Johannes: In fact, there is a significant overlap and mutual enrichment between the AI systems used in autonomous vehicles and those used in other areas. AI technologies originally developed for autonomous driving, such as machine vision and decision-making algorithms, are increasingly being used in areas such as robotics for industrial automation, surveillance technology and even in the healthcare industry, where they are used to analyze medical images, for example.
Conversely, the development of autonomous vehicles is also benefiting from advances in other areas of AI. For example, improvements in natural language processing and interactive systems have made user interactions in vehicles more intuitive (e.g. ChatGPT). In addition, the application of reinforcement learning, originally used in games and simulations, leads to better and adaptive algorithms for driving behavior under variable conditions. These synergies not only increase the performance of the systems, but also accelerate the innovation cycle in the automotive industry.
Thank you, Johannes, for this exciting insight into this complex topic and your view of the future. It sounds like things will develop at a rapid pace over the next few years and we are excited to see what the future holds. AI is not only advancing the automotive industry, as Johannes has already mentioned. There is also enormous potential in other areas and sectors.
Du you want to be able to recognize the potential of AI?
Discover our AI orientation Workshop now, where we take you on a guided journey
weitere Beiträge
The current query has no posts. Please make sure you have published items matching your query.