7th International VDI Conference - Autonomous Driving

Registration & Welcome Coffee

Chair’s Welcome & Opening Address

I. Strategy

Level 4 Automated Driving – From Concept to Product
Alexander Schaab, Vice President Autonomous Driving, Mercedes-Benz Research & Development North America, USA

II. Perception, Prediction & Planning

Autonomous Planning & Control for Intelligent Vehicles in Traffic

• Addressing the trajectory planning problem for AVs in traffic
• „Dynamic cell“ to dynamically modify the state of traffic according to different variables
• Achieving desired driving styles of AVs via reinforcement learning

Dr. Jianbo Lu, Technical Expert & Group Leader, Ford Motor Company, USA

Networking & Coffee Break

Safe but not Cautious Path Planning for AD

• The importance of assume/guarantees
• Understanding what risks the vehicle gets exposed to
• Driving safely without perfect perception

Dr. Mattias Brännström, Technical Expert in Safe Decision & Control for AD, Zenuity AB, Sweden

Building Intelligent Autonomous Driving Systems out of Silicon Blocks

• Where does the intelligence reside in an autonomous system?
• Requirements & constraints on the hardware
• Expectations for the software enablement
• How does safety and security get factored in?

Ali Osman Ors, Director, Automotive AI Strategy & Partnerships Automotive, NXP Semiconductors, Canada

Start-up Session

A speed-networking session bringing together creative talents and experienced experts – meet some of the most exciting vehicle automation Start-ups

Lunch

Object Prediction in Urban Environments

• Gathering & interpreting context information to understand the AV‘s current situation
• Generating reasonable predictions of each traffic participants‘ behavior
• Overview of how these challenges are addressed at BMW, focusing on urban environments

Dr. Michael Karg, Expert Autonomous Driving, BMW Group, Germany

Generating a HD Vehicle Horizon for Automated Driving

• Benefits of ADAS & HD maps
• ADASIS specification for vehicle-wide map distribution
• First-to-market ADASIS v3 implementation
• From R&D to series production

Dr. Michael Reichel, Director, Head of Product Management Automated Driving, Elektrobit Automotive GmbH, Germany

Interactive World Café Sessions – Exemplary topics:

• Can we use the same Radar Sensor Systems for all Levels of Vehicle Automation?
  Dr. Jürgen Dickmann, Head of Radar Sensors/Radar Perception, Mercedes-Benz AG, Germany
• How to Develop Robust, Safe, Secure & Cost-Efficient Architectures for AVs?
  David Sanchez, Assisted and Automated Driving Senior Technical Specialist, Jaguar Land Rover, UK
• Current Challenges for and Approaches to Automated Vehicle Validation.
  Prof. Dr. Christof Ebert, Managing Director, Vector Consulting Services, Germany

Networking & Coffee Break

III. Processes & Productization

Mass Deployment of Safe AI-based Systems for Autonomous Vehicles

Eric Zhang, Director, Functional Safety, Automotive AI Processor, Horizon Robotics, China

Autoware: An Open Source Autonomous Driving Software Ecosystem

• Motivation for open source Autonomous Driving (AD) software stacks
• Benefits and challenges of using open source software for AD
• Introduction of Autoware: An open source AD platform
• Building & maintaining a community of users
• Autoware as a platform for AD education & research

Dr. Simon Thompson, Lead Engineer, Tier IV, Japan

Continuous Development of ADAS & Autonomous Driving Functions

• A scalable hardware & software platform as key enabler
• Use of AI & Machine Learning for environment perception
• Demonstrating the potential of Visteon‘s cloud-based workbench for the continuous development process
• Roadmap for ADAS and autonomous driving & current status at Visteon

Matthias Schulze, VP Technology Management, Head of ADAS, Visteon Electronics, Germany

End of Conference Day One

Get-Together

At the end of the first conference day we kindly invite you to use the relaxed and informal atmosphere for in-depth conversations with other participants and speakers.

IV. Safety

Safety & SOTIF Processes Deployment for Autonomous Driving

• Safety & SOTIF issues of driving automation
• Need of a quantitative approach for SOTIF objectives definition
• Importance of a shared scenario database for validation
• Proposal of a global methodology based on simulation & driving

Dr. Vincent Abadie, Senior Expert ADAS & AD, Nicolas Becker, Senior Expert Safety, Dr. Stephane Géronimi, Expert Safety ADAS & AD, all: Groupe PSA, France

Responsibility Sensitive Safety: Towards an Open, Transparent Industry-Driven Approach to AV Safety

• Providing a check on AV decision-making by digitizing implicit rules of the road
• Introduction to RSS principles & the RSS Open Library
• Integration with AV software stacks
• Towards a common safety metric, safety benchmarks

Dr. Ignacio Alvarez, Senior Research Scientist, Intel Corporation, USA

Ensuring the Safety of Machine Learning-Based Autonomous Systems

• Need for safety assurance of learning-based autonomous systems
• Bridging the gap between machine learning & formal verification
• Synthesising runtime monitors for ML units

Dr.Ing. Vahid Hashemi, Technical Project Manager, AI Safety Engineering, Audi AG, Germany

Real-word Driving Scenario Identification for AV Functional Safety

• Determination of safety critical scenarios to validate control laws with driving simulators
• Statistical analysis of parameters for each scenario
• Generating validated, annotated & analyzed data for AV commercialization

Gilles Le Calvez, Vehicle Program Director, Vedecom, France

Networking & Coffee Break

V. Validation & Testing

VVM – Pegasus Project Family: Release Methods for Highly Automated Vehicles – First Advances

• Overview main targets VVM - Verification & Validation Methods for L4 & 5 Automated Vehicles
• Condensation of test space by causal analysis of critical phenomena
• Integral approaches of safety argumentation

Roland Galbas, Technical Expert & Project Lead, co-authors: VVM-Team, all: Robert Bosch GmbH, Germany

Generation of Simulation Data Using Police Recorded Accident Data for the Development of Active & Passive Safety Systems

• Creation of realistic scenarios from police recorded accidents
• Assumptions to simulate the trajectory of each participant
• Variation of accident parameters like initial speed in simulation
• Speed profiles from GIDAS or/and AIMATS

Martin Urban, Research Associate, co-author: Dr. Christian T. Erbsmehl, Group Leader, both: Fraunhofer Institut for Transportation and Infrastructure IVI, Germany, Jorge Lorente Mallada, Design Engineer, Toyota Motor Europe, Belgium

Real Data Based Validation of Highly Automated Driving Functions Using Simulation Methods

• Approaches to machine learning methods for scenario identification
• Scenario extraction from recorded data
• Automated transfer to simulation environments

Raphael Pfeffer, Product Manager Test Systems, IPG Automotive GmbH, Germany

Lunch

First Results from L3Pilot – Driving Automation

• Testing the viability of automated driving on public roads: Focus on large-scale piloting of SAE L3 functions, some L4 functions
• Wide range of driving situations, including parking, overtaking on highways & driving through urban intersections
• Evaluating technical aspects, user acceptance, driving & travel behavior, impact on traffic & safety

David Sanchez, Assisted & Automated Driving Senior Technical Specialist, Jaguar Land Rover, UK

VI. Homologation & Certification

Current State of Future UN Homologation & Approval Regulations for Autonomous Vehicles

• Framework for automated vehicles regulatory activities
• Focus on work currently done on Level 3 technologies (ALKS)
• Review of existing regulation regarding Level 2 technologies (ACSF)
• Data Storage Systems for Automated Driving activities & Event Data Recorder

François E. Guichard, Secretary to the Working Party on Automated/Autonomous & Connected Vehicles, UNECE, Switzerland

Applicability of Traffic Laws to Automated Driving

• Need for conformity with both (supranational) vehicle regulations and (national) traffic laws
• Current debate at the UNECE Global Forum for Road Traffic Safety (WP.1)
• National activities for amending traffic laws to facilitate safe operation of AVs
• Regulatory issues for L4 automated driving in conjunction with traffic laws

Bogdan Bereczki, Product Safety and Integrity, Autonomous Intelligent Driving GmbH, Germany

Closing Remarks by the Conference Chair

End of Conference