International VDI Conference “AI in Vehicles & Software Engineering”

• Large Language Models enable unprecedented capabilities for AV systems and in-cabin AI assistants
• While first LLMs are already deployed in-car, Multi-modal models like VLMs as well as agentic AI are making their way into vehicles
• This session will explain which benefits these models demonstrate, how they can be deployed in-car, and how to build applications on top of them 
  Felix Friedmann, Engineering Lead AI for DriveOS, NVIDIA, Germany

• What is Gen AI and what will it be used for in embedded devices and in automotive?
• Embedded hardware that meets the requirements to run Gen AI
• Trade-offs between model size, accuracy, latency, and energy efficiency
  MSc, Peter Kristiansen, Head of Business Development, co-author: PhD, Andreas Ask, both: Embedl, Sweden

• Small Language Models (SLM) for minimum latency, privacy, reliability and cost efficiency
• SLM integration into series vehicles with Software-Defined Vehicle (SDV) architecture
• Implementation for reconfigurable dashboard feature
  Alexander Kugler, Lead Solution Manager Infotainment & Connected Mobility, FEV.io GmbH, Germany

• Influence of driving style in the context of trailers and maintenance
• Behavior analysis using telemetry data
• Machine Learning Techniques for behavior analysis
• Integration with digital twin solutions
  Ing. Laura Gregorc, MSc, Automotive Data Science / Researcher, Virtual Vehicle Research GmbH, Austria

• Fully automated vehicles allow for new vehicle interiors and cooperative human-machine interactions
• We present an adaptive vehicle interior concept with AI assistant for mobile work and leisure activities
• AI assistant gradually learns user preferences for a situation-dependent vehicle interior
  Frank Beutenmüller, Managing  Director, co-authors: Patrick Vaudrevange & Raghav Nath, all: TWT GmbH Science & Innovation, Germany

• Connect and optimize all your AI data
• How to optimize and improve performance for large scale AI workloads
• Learn how organizations today are retooling their IT operations to embrace artificial intelligence (AI), machine learning (ML), and advanced analytics
  Frank Kraemer, Systems Architect, IBM, Germany

• Identifying the optimal use cases for AI in testing
• The good, the bad, and the ugly of AI in testing
• Success stories of AI in automotive testing
  Robert Fey, Product Management, Synopsys, Germany

• Specific requirements for data and models in engineering
• Data- & knowledge-engineering
• Use cases
  Dr.-Ing. Michael Dlugosch, Head of Business Unit Automotive, Fraunhofer-Institute for High-Speed Dynamics, EMI, Germany

• Engineering data analysis is difficult because of data silos (different data formats, poor data catalogues, many subsystems / signals)
• We show how LLM-based assistants can be used to contextualize data, query data and visualize results
• Concrete applications from testing and fleet data analysis
• Best practices how to integrate AI assistants into existing platforms & tools
  Dr. Stefan Suwelack, CEO and Co-Founder Renumics, Germany

• AI ADAS workload scaling.
• Memory bottleneck
• HW-aware Optimizations
• HW/SW codesign and future trends. 
  Dr. Ing.Taha Soliman, Research Project Lead and Expert, Robert Bosch Gmbh, Germany

   At the end of the first conference day, we kindly invite you to use the relaxed and informal atmosphere at our conference dinner for indepth conversations with other participants and speakers.

• RUST
• Real-time perception
• Safe AI
  Dr. Jürgen Hess, VP Engineering, Spleenlab GmbH, Germany

• Supporting requirements engineering with AI
• AI enhanced systems and software architecture management
• Applying AI for specifying system behavior
• AI and testing
  Andreas Graf, Head of AI and Dataspaces, co-authors: Sruthi Radhakrishnan & Ario Giancarlo Cecchettini, all: itemis AG, Germany

• Generative Al CSMS-Framework for monitoring and defending against cyber attacks
• Malicious Generative AI to generate new unknown attack vectors
• Up to date and one step ahead against cyber attacks with GenAI
  Dr.-Ing. Ugur Akcakoca, Head of Department ES² - Embedded Safety & Security, EDAG Engineering GmbH, Germany

• Challenges in software testing: High effort in creating and maintaining test cases and integrating them into existing workflows
• How AI can help: Leveraging natural language and generative AI to automate and streamline testing processes
• ZenTestAI in action: A solution combining intuitive language inputs with automated test execution to make quality assurance more user-friendly and efficient
  Dragan Sunjka, Lead IT Consultant, msg for automotive gmbh, Germany

Nick Westphal, Digital Transformation Owner - Head of AI, EDAG Engineering GmbH 
Florian Sauerbeck, Sr. Software Engineer, driveblocks GmbH 
Robert Fey, Product Management, Synopsys

• Heterogeneous regulatory landscape
• Relevance of global standards
• Lifecycle compliance
  Christian Pahlke, Head of Strategic Initiatives AI and Cybersecurity, TÜV Süd, Germany

• Identifying current relevant AI regulations
• Application of AI regulations to the automotive sector
• Cybersecurity and new AI regulation
  Javier Mendoza, Team Lead Security Analyses and Concepts, AVL Software and Functions GmbH, Germany