CO-LOCATED EVENTS

Blog

Keynote: Current challenges of AI and machine learning applications at the BMW Group

As the automotive industry accelerates towards autonomous, connected, electrified and shared mobility, the integration of AI and ML technologies is essential for achieving added value for virtually all customer facing products and services. At BMW, we are at the forefront of this innovation, but the journey is fraught with complex challenges that must be overcome.Read more

Tech Take by Scaleout Systems: Federated learning, fleet learning, autonomous driving, self-supervised learning, semantic segmentation

In this demo, we will showcase: How federated learning addresses the challenge of large data volume and data privacy by keeping data on-vehicle, enabling fleet learning.  How self-supervised learning overcomes the need for labeled data in computer vision tasks, making it ideal for on-vehicle deployment where manual data annotation is impractical In our demo, we leverageRead more

Solution Study: Generation of logical simulation scenarios using Generative AI

Generative AI is being quickly adapted in the automotive industry. The talk emphasizes on how generative AI could be leveraged in context of AD/ADAS, especially in simulation and validation, with a clear focus on the data driven development tool chain. In this session. you will learn more about: The impact of Generative AI on theRead more

Solution Study: Learning from best-practice annotation setups for autonomous systems

As autonomous systems continue to evolve, the diversity and complexity of required annotations have grown exponentially. Challenges are ranging from multi-sensor fusion data and complex scene understanding to early fusion approaches and voxel representation support. Modern annotation processes demand the right expertise, tooling, workflows and workforce to deliver high-quality datasets at scale. In this session,Read more

B | AI Infrastructure Stream | Solution Study: Eliminate costly hardware and drive tangible benefits from your SDV stack – How to replace headlight & tire pressure sensors with AI

COMPREDICT’s AI-based Virtual Sensors bring measurable benefits to your SDV stack by providing an intelligent and cost-efficient alternative to expensive and failure prone hardware sensors. This session will focus on how essential components such as Headlight Adjustment and Tire Pressure sensors can be effectively replaced with AI-based Virtual Sensors. These solutions not only maintain the highest levels of accuracy,Read more

A | Perception & Sensor Technology Stream | Solution Study: From Lab to road – Mastering MLOps for autonomous driving

The rise of autonomous driving technology promises to revolutionize transportation, offering safer, more efficient travel. However, the journey from concept to reality is fraught with challenges, particularly in the realm of Machine Learning Operations. This presentation will explore the five biggest MLOps challenges in autonomous driving: data management and quality, model training and validation, real-timeRead more

Case Study: Object-centric open-vocabulary image retrieval with aggregated features

The task of open-vocabulary object-centric image retrieval involves the retrieval of images containing a specified object of interest, delineated by an open-set text query. As working on large image datasets becomes standard, solving this task efficiently has gained significant practical importance. Applications include targeted performance analysis of retrieved images using ad-hoc queries and hard exampleRead more

Case Study: Utilizing intelligent data pipelines for autonomous data collection

Intelligent data pipelines are key for enhancing the performance, cost reduction, safety, and reliability of AVs. By leveraging advanced data processing techniques and ML algorithms, intelligent data pipelines can use the raw sensor inputs and transform them into usable data for AI models, and perform the best selection of data for annotation. Addi’s team hasRead more

Case Study: Towards zero collisions – faster – by harnessing the power of deep learning

As software-defined vehicles promise to make vehicles safer, the critical role of AI-powered perception and planning cannot be overstated. The session discusses the underlying methodologies and technologies for a modern AD/ADAS stack, highlighting the pivotal role of data and deep learning. Zenseact’s software stack, which is being release in the Polestar 3 and Volvo EX90,Read more

Case Study: AUTOSAR – Towards SDV

This presentation explores the challenges of SDVs and the evolving ecosystem surrounding their development. Reichart discussed AUTOSAR’s opening strategy and its role in advancing SDV technology. Additionally, he outlines AUTOSAR’s future evolution to meet the demands of SDVs, focusing on enhanced communication protocols and software architectures. Learn how to navigate the path towards a saferRead more