As a premier knowledge exchange platform in North America bringing together 250+ stakeholders who are playing an active role in the vehicle automation scene, this year's topics and focus lies in testing & validation, sensor fusion, deep driving, cafety-critical systems, computer vision, software architectures and much more.
Join Tech.AD and discuss your roadmap to vehicle automation now!
What is the state of the most state of the art algorithms for computer vision and machine learning? What role will Deep Neural Networks (DNN) have in future self-driving cars and what will the associated challenges of this be? What are the implications of machine learning for functional safety? What level of traceability is possible with neural networks? How much real-time data and decision logs can we access? Can we really implement end-to-end learning or is it better use hybrid approaches? What are the key challenges in AI and perception technologies for autonomous vehicles?
To what extent do edge cases define the development of autonomous vehicles? Which test cases can we cover with simulation and which not? Will virtual kilometers (simulations, HIL/SIL) be accepted as an industry validation method? What can we learn from the latest research in advanced, large-scale testing of autonomous vehicles and validation methods? What are the challenges behind using synthetic data for validation?
What are the key challenges in the Cybersecurity of Autonomous Vehicle Platooning? What are the different control algorithms? How do different wireless technologies perform in high-density truck platooning achieved with vehicle-to-vehicle (V2V) communications? What results and challenges do we see from the early deployment of autonomous shuttles and autonomous platoons? What data is already being collected, what data is missing and what are future research requirements?
In what areas are ISO 26262 insufficient? Does SOTIF fully close the gap where ISO 26262 is insufficient? How to achieve functional safety for autonomous vehicles? How can we ensure functional safety with AI and machine learning? What do we still need to do to achieve functional safety for deep learning algorithms?
Which kind of sensors provides the most value? How to push hard for cost optimization without comprising safety? Do we need deep learning for sensor fusion? Can radar sensors replace ultra-sonic sensors completely? Which sensors are most suitable for data fusion in close range? Which combinations of sensors will be required for robust machine perception in HAD? To what extent are high-resolution maps required for offering context information for scene understanding?
How far can vehicle architectures solve the functional safety need for autonomous vehicles? What are the challenges to step from fail-safe to fail-operational architectures as a foundation of machine learning systems in autonomous vehicles? What is the case for a flexible fusion architecture? How to understand the system architecture for automated driving and scale-up system architectures? What are the challenges in the development and testing of fail-operational automated driving architectures? What are the newest concepts, challenges, use cases & game changers to the new AUTOSAR Adaptive Platform for Connected and Autonomous Vehicles?
How can we leverage V2V and V2Cloud communications to automate trucks? What can we learn from this for autonomous passenger vehicles? How will 5G drive and accelerate the development of autonomous vehicles? What new opportunities will 5G create for autonomous driving? What current challenges on the road to level 5 could 5G resolve? What information could now be transferred through V2V with the support of 5G and how can it be used in the processing of data collection from machine learning, sensors etc?