Blog

Access to real-world data at scale is the key to solving the toughest challenges in autonomous vehicle development. By harnessing crowdsourced driving data, we can accelerate AV safety, testing, and validation, bringing us closer to mass-market adoption. This session explores how crowdsourced data enhances sensor fusion, improves HD mapping, and provides real-time insights that helpRead more

After another insightful & productive day full of presentations & discussions from the industry, our event chair concludes the event with a short summary of the most important insights from the sessions.Read more

This presentation focuses on using AI to generate test cases, test data, and automated test suites for validating vehicle-related APIs. It addresses the integration of AI-generated tests into CI/CD pipelines and explores methods for validating AI-driven outputs. The session also discusses the maintenance of AI-generated test assets throughout the software lifecycle. Attendees will gain practicalRead more

This session explores the use of vision-language models (VLMs) to extract data relevant to operational design domains (ODDs) for urban automated driving. It covers insights on VLM performance against diverse ODD requirements, highlights the value of large-scale camera data from vehicle fleets, and introduces new datasets scheduled for release in 2025. Attendees will gain aRead more

As generative AI expands beyond text, its applications in autonomous vehicles are redefining how machines interact with the physical world. This session explores the use of generative models in handling 3D models, sensor streams, and domain-specific data within the context of autonomous systems. Attendees will gain insights into the adaptation of generative AI for non-languageRead more

As SAE Level 3 and 4 automation becomes increasingly viable for commercial vehicles, the design of HMI in long-haul trucks and urban buses presents new and complex challenges. These large vehicles operate under fatigue-prone conditions, with dynamic environments and unique cabin ergonomics, demanding careful consideration of how drivers and automation interact. This presentation explores theRead more

This presentation explores how data fuels the development of AI-powered ADAS in autonomous vehicles. It addresses key challenges in data privacy, secure sharing, and maintaining data quality across the AI lifecycle. We’ll also examine how closed-loop testing and monitoring ensure system reliability and safety in real-world conditions. In this session, you will learn more about:Read more

Autonomous vehicles must operate within clearly defined ODDs, but real-world uncertainties like rain, fog, and low visibility challenge the boundaries of these domains. This presentation explores how AV systems detect when they are leaving their intended ODD, how we assess the risk of false negatives, and how we can verify system behavior under uncertain conditions.Read more

1 start-up – 5 min – multiple opportunities: The most innovative start-ups showcase their services, products in a dedicated session during the conference. Don’t miss the chance to meet our selected start-ups at the end of the break in the plenary room and network with them at any time during the conference to learn moreRead more

As autonomous vehicles demand higher performance and safety, traditional E/E architectures are giving way to zonal designs. This presentation explores how TSN and fail-silent domain controllers can be architected to meet ASIL-D requirements for timing, fault isolation, and availability. It presents a reference architecture leveraging TSN standards (e.g., 802.1Qbv, Qci) and lock-step computing with safetyRead more