Bosch and CARIAD Collaborate on AI and Automated Driving
The automotive industry is undergoing a significant transformation, driven by advancements in artificial intelligence (AI) that are reshaping the way we drive.
In a bold move to make driving safer and more convenient, Bosch and CARIAD, part of the Volkswagen Group, have teamed up to develop an AI-powered software stack that will revolutionise automated driving.
By leveraging AI, this collaboration aims to bring us closer to fully autonomous vehicles, making human-like driver assistance systems a reality.
A New Era in Automated Driving
Bosch and CARIAD’s collaboration focuses on developing the software stack for Level 2 and Level 3 autonomous driving. These levels of automation allow the vehicle to handle many aspects of driving, including acceleration, braking, and steering, under specific conditions. By integrating AI into every step of the development process, the two companies aim to create driving functions that act as naturally as a human driver. The ultimate goal? To make automated driving functions available to millions of consumers, from mainstream to premium car owners.
Mathias Pillin, CTO of Bosch Mobility, underscores the role of data and AI: “When it comes to getting automated driving systems onto the roads reliably and at scale, data and AI are the key.” Bosch and CARIAD are leveraging cutting-edge AI techniques to handle the complex tasks involved in perception, interpretation, decision-making, and action. This AI-powered software will make driving systems more responsive, intuitive, and safer for everyday use.
Building a Robust AI Framework
At the heart of this collaboration is the development of an AI stack that covers all critical driving functions. The AI is used in the full range of system components, from object recognition to decision-making and control over the powertrain, steering, and braking. AI’s ability to process vast amounts of data from various sensors such as cameras and radars ensures that the vehicle can respond quickly and effectively to a dynamic road environment.
By using AI-driven perception models, the system anticipates not only the current behaviour of road users but also predicts their future actions, improving the vehicle’s ability to react to complex traffic scenarios. Bosch and CARIAD have embraced AI methods found in generative AI applications, like language models, to analyse traffic and ensure a more fluid driving experience.
The software stack will evolve with AI powering all the modules, ensuring that the system remains adaptable, scalable, and future-proof. As the companies continue to expand their capabilities, AI will enable more intelligent decision-making and smoother vehicle control, enhancing safety and convenience for drivers.
Data-Driven Development
In automated driving, data is invaluable. Bosch and CARIAD have made data-driven development a cornerstone of their strategy. The two companies are working with a large fleet of test vehicles on public roads across Europe, Japan, and the United States. These test vehicles are equipped with advanced sensors that capture real-time data, which is then used to train and fine-tune the AI system.
The vehicles in the test fleet include high-profile models such as the ID.Buzz and Audi Q8. These vehicles are not just gathering data; they are part of a continuous loop of feedback and improvement. As the vehicles accumulate data, the software undergoes updates multiple times a day to improve performance. This iterative process ensures that the AI systems are trained and refined with a constant influx of new data, enabling the system to handle rare and complex driving situations, or “corner cases.”
By using these real-world scenarios, Bosch and CARIAD are ensuring that the software stack can adapt to unpredictable environments and enhance system reliability. The fleet’s data helps the team identify gaps in the AI’s decision-making, leading to further refinement of the algorithms to make the system more robust.
Collaboration for Global Impact
Bosch and CARIAD’s collaboration is not only about developing technology for the Volkswagen Group’s vehicles but also about creating solutions that can be deployed globally. Bosch plans to make the software stack available to other car manufacturers, enabling a wider reach and adoption of automated driving technology.
Peter Bosch, CEO of CARIAD, emphasised the European angle: “We’re demonstrating that the German automotive industry has mastered the key technologies of artificial intelligence and automated driving. Our goal in the Alliance is to make the convenience and safety of automated driving systems accessible to as many people as possible so that they can gain valuable time when they’re in their car.” This statement reflects the European commitment to digital sovereignty and global competitiveness in AI-powered automotive technology.
The new architecture for software-defined vehicles will allow Bosch and CARIAD to integrate their automated driving functions seamlessly into next-generation vehicles. This architecture is designed to be scalable, which means it can be applied to a wide variety of vehicles—from mass-market to high-end models. This approach could accelerate the mass adoption of automated driving systems worldwide, reducing road accidents and improving efficiency.
AI at the Core of the Technology
The key to this innovation is the AI-driven, end-to-end architecture. Bosch and CARIAD are developing the entire software stack using their own intellectual property and source code, which gives them full control over the technical aspects. This complete technical control allows the partners to maintain high standards in data protection, security, and driving safety, while ensuring that innovations can be introduced rapidly.
The stack is designed to be transparent, safe, and explainable. This is crucial, as it allows for confidence in the AI’s decisions. The system’s actions are traceable, and its reasoning can be understood and verified, ensuring that automated driving functions are both safe and reliable.
Additionally, Bosch and CARIAD are exploring multimodal AI approaches. These approaches combine visual and linguistic data to enhance the system’s understanding of complex scenarios. For example, by using vision-language-action (VLA) methods, the system can better identify hidden risks and respond to them appropriately. These approaches promise to take the AI’s performance to a new level, making the driving experience even more intuitive and secure.
Looking to the Future
The path to fully automated driving is still being paved, but Bosch and CARIAD’s efforts represent a significant step forward. By 2026, the AI-powered software stack will be ready for use in production projects. The ongoing test fleet, the use of AI throughout the development process, and the ability to integrate these solutions into various vehicle models will bring us closer to the vision of safer, more convenient automated driving systems.
Thanks to this partnership, the future of driving looks increasingly autonomous, where AI ensures that vehicles are safer, smarter, and more efficient. By focusing on scalable solutions and driving innovation with data, Bosch and CARIAD are setting a new standard for the automotive industry.