W03 New Platforms for Future Cars: Current and Emerging Trends

Printer-friendly versionPDF version
Location / Room: 
Konferenz 3


Selma Saidi, Hamburg University of Technology, DE (Contact Selma Saidi)
Muhammad Shafique, TU Wien, AT (Contact Muhammad Shafique)

The automotive industry is rapidly moving towards the adoption of new hardware solutions, such as multi- and many-core processors, FPGAs and neural networks in order to comply with the increasingly number of ECUs and the high computational power required by future autonomous cars and advanced vision processing. Moreover, with the emerging trends in autonomous driving such as V2X communication and Internet of Cars, security is also arising as a serious concern for many automotive players. Therefore, new architecture solutions pose numerous challenges such as migrating legacy code, data fusion and processing of a multitude of sensors piling into autonomous cars, mapping affecting processing power demands in new systems architectures, employment of deep learning and other advanced AI techniques, the verification and validation of timing safety and providing secure interfaces to connect the cars to each other and to the outside world.

The goal of this workshop is to bring together researchers as well as practitioners from academia and industry to discuss and investigate cutting-edge methods and tools that will be driving technology innovation in the field of autonomous systems in order to devise highly efficient systems and systems-of-systems for future self-driving cars.

The program of the workshop includes invited talks from industry and academia experts. The workshop also involves short presentations based on selected submitted contributions in addition to a poster and demos session.

Invited Speakers:

  • Michael Kiehn, Ibeo Automotive Systems GmbH, DE
  • Jens Benndorf, Dream Chip Technologies, DE
  • Radu Grosu, Vienna University of Technology, AT
  • Dirk Ziegenbein, Robert Bosch GmbH, DE
  • Jürgen Becker, KIT - Karlsruher Institut für Technolgie, DE
  • Wilfried Steiner, TTTech Computertechnik AG, AT
  • Rolf Ernst, Technische Universität Braunschweig, DE
  • Bernd Elend, NXP Semiconductors, DE
  • Albrecht Mayer, Infineon Technologies, DE


Infineon Technologies, Dream Chip Technologies, fortiss GmbH, TU Braunschweig, University of Valencienne.


08:15W03.1Welcome and Opening

Selma Saidi, Hamburg University of Technology, DE
Muhammad Shafique, Vienna University of Technology (TU Wien), AT

08:30W03.2Session 1 : From Advanced Data Sensing to Smart Processing in Autonomous Cars
08:30W03.2.1Processing needs for next generation 3D LIDAR
Michael Kiehn, Ibeo Automotive Systems GmbH, DE

09:00W03.2.2Automotive Multiprocessor ADAS Chip Design in 22nm FDSOI Technology
Jens Benndorf, Dream Chip Technologies, DE

09:30W03.2.3Parking with a Worms Brain
Radu Grosu, TU Wien, AT

10:00W03.3Coffee Break
10:30W03.4Session 2 : Current Practices in Hardware and Software Automotive Design
10:30W03.4.1APP4MC: An Open Source Platform for Automotive Multi-Core Systems
Dirk Ziegenbein, Robert Bosch GmbH, DE

11:00W03.4.2Reliable Heterogenous Multi-Core Platforms - Cyber-Physical Systems in Embed- ded: Automotive, Industry 4.x , Avionics -
Juergen Becker, KIT - Karlsruher Institut für Technolgie, DE

11:30W03.4.3Short Presentations

11h30: Driving Against the Memory Wall: The Role of Memory for Autonomous Driving
Matthias Jung and Norbert Wehn, TU Kaiserslautern, DE
11h40: Towards a Predictable Execution Model for Heterogeneous Systems-on-a-Chip"
Andrea Marongiu, University of Bologna, IT
11h50: Architecture Exploration for Safety-Critical Systems
Alexander Diewald, Simon Barner and Sebastian Voss, fortiss GmbH, DE

12:00W03.5Lunch Break
13:00W03.6Session 3 : Communication and Software Integration in Automotive Systems
13:00W03.6.1Highly-Integrated E/E Architectures by means of Vehicle-Wide Virtualization
Wilfried Steiner, TTTech Computertechnik AG, AT

13:30W03.6.2System Level Logical Execution Time (LET)
Rolf Ernst, TU Braunschweig, IDA, DE

14:00W03.6.3Short Presentations

14h00: An Efficient and Streamlined Open Source Implementation of the LET Paradigm
Matthias Beckert and Rolf Ernst, TU Braunschweig, DE
14h10: Big Data Analytics for Smart Cities: The H2020 CLASS Project
Eduardo Quinones (1), Marko Bertogna(2), Erez Hadad(3), Ana Juan Ferrer(4), Luca Chiantore(5) and   Alfredo Reboa(6)(1)Barcelona Supercomputing Center, ES,(2)Univ. of Modena, IT, (3)IBM, IL, (4)ATOS,   ES, (5)City of Modena, IT (6)Maserati, IT
14h20: WCET Analysis of Automotive Buses using WCC
Dominic Oehlert and Heiko Falk, TU Hamburg, DE

14:30W03.7Coffee Break
15:00W03.8Session 4: Security as a Pilar for Automotive Functional Safety
15:00W03.8.1Secure CAN Communication for Improved Functional Safety - a New Approach Without the Challenges of Using Cryptography
Bernd Elend, NXP Semiconductors, DE

15:30W03.8.2Cyber Security Paranoia: a Threat for Functional Safety?
Albrecht Mayer, Infineon Technologies, DE

16:00W03.9Short Presentations: New Platforms for Deep Learning

16h00: Embedded GPUs in Future Automated Cars
Joerg Fickenscher(1), Frank Hannig(1), Mohamed Essayed Bouzouraa(2) and Jürgen Teich(1)
(1)Friedrich-Alexander-Universität Erlangen-Nürnberg,DE (2) AUDI AG, DE
16h10: Convolutional Neural Networks on Embedded Automotive Platforms: a Qualitative Comparison
Paolo Burgio, Gianluca Brilli, Antonio Marra and Marko Bertogna, University of Modena, IT
16h20: Road Extraction and Object Recognition for Autonomous Cars
Smail Niar(1), Mohamed Yazid Laachachi(1), Mohamed Ayoub Neggaz(1), Ihsen Alouani(1), Hasan Yantir(2), Abdelmalik Taleb-Ahmed(1), Fadi Kurdahi(2) and Ahmed Eltawil(2)
(1)University of Valencienne, FR, (2)University of California, Irvine, US

16:30W03.10Interactive Discussion: Posters and Demos
17:30W03.11Closing Session