Generative methods for
perception, prediction and planning
Discover the project and its
for automated driving.
Generative methods for
perception, prediction
and planning
Discover the project and its approaches for automated driving.


Automated driving functions are very limited in their scope of use. The reasons lie in the system architecture and the discriminative machine learning methods applied today. Based on generative methods, NXT GEN AI METHODS – nxtAIM introduces the bidirectional flow of information as a new paradigm into the chain of effects and enables massive improvements for the development of autonomous driving functions. Foundation models for driving data will emerge as an outstanding result for industry implementation.


nxtAIM will utilize the massive potential of generative methods to develop new approaches for better scalability, better transferability, and better traceability. The focus is on the development of generative methods that are complementary to the established discriminative methods of artificial intelligence and thus initiate a paradigm shift towards a bidirectional flow of information in the value chain.


The mobility of the future is autonomous: people and machines share the traffic space. They interact and cooperate. However, the necessary level of cognitive skills of autonomous road users can only be achieved through a comprehensive use of machine learning methods.

With regard to the scalability, transferability and traceability of the developed data-based driving functionality, hurdles remain on the way to higher degrees of autonomy, nevertheless. These limitations stem from the current system architecture and the specific discriminative machine learning methods employed.

Generative AI offers an alternative approach and has impressively demonstrated its capabilities and technological maturity in recent applications such as large language models or text-to-image generators. By using generative methods and the resulting development of foundation models, nxtAIM tries to initiate nothing less than a paradigm shift in system architecture and AI methodology.

Today’s system architectures are based on linear, unidirectional information processing along the so-called chain of effects consisting of perception, environment modeling, prediction & planning and subsequent implementation. Generative methods are used to create a feedback channel and information processing is expanded bidirectionally.

Project Consortium

Facts & Figures


€ 43,5 M

Project Coordination

Jörg Reichardt


Ulrich Kreßel



21 Partners

OEM, Suppliers, Technology providers, Research facilities, Universities, External partners


€ 27,0 M


36 Months

1 January 2024 – 31 December 2026
Scroll to Top