W03 DATE Workshop on Autonomous Systems Design (ASD2019)

Printer-friendly versionPDF version
Date: 
2019-03-29
Time: 
08:30-17:30
Location / Room: 
Room 1

Organisers

Rolf Ernst, TU Braunschweig, DE (Contact Rolf Ernst)
Selma Saidi, Hamburg University of Technology, DE (Contact Selma Saidi)
Dirk Ziegenbein, Robert Bosch GmbH, DE (Contact Dirk Ziegenbein)

Publicity Chair

Sebastian Steinhorst, TUM, DE (Contact Sebastian Steinhorst)

ASD is the 1st international workshop on Autonomous Systems Design. The goal of ASD is to explore recent industrial and academic methods and methodologies in autonomous systems design. This includes several areas:

  • Embedded and cyber-physical systems platforms that implement and execute the autonomous system functions including their architectures, hardware, software and communication.
  • The design for autonomous systems including processes, modelling, optimization, verification, validation, and test.
  • All aspects of dependable systems design for autonomous systems including, but not limited to, functional safety concepts, fail-operational systems design, functional safety for applications with machine learning, safe and secure changes and updates, autonomous systems security.
NEW: Proceedings are available for download on the following link:
http://drops.dagstuhl.de/portals/oasics/index.php?semnr=16104

Highlights

The workshop consists of regular sessions with papers selected from an open call for papers, complemented by invited talks, and an exhibition with posters and live demos. In addition, the workshop will feature the two following distinguished keynotes:

Keynote 1: Challenges of Automated and Connected Driving

Speaker: Thomas Form, Head of Electronics and Vehicle Research, Volkswagen AG, Germany

Summary: In recent years, various publications and presentations from a lot of companies have shown the improvements in the sector of automated driving. The vehicle- and mobility-concept SEDRIC is a current example from the Volkswagen AG. However, for a release of these technologies there are several unresolved issues regarding sensor technologies, redundancies as well as verification and validation questions. Regarding sensors, the main objectives are miniaturization and reduction of system costs. Advantages and disadvantages of existing solutions have to be evaluated. In addition to economic aspects, ensuring the redundancy of the system is absolute necessary. Is, for example, Artificial Intelligence able to provide an independent second or third function path? Regarding verification and validation concepts, current discussions are focused on which scenarios have to be tested and how, in order to enable regulatory authorities to approve the release of automated driving functions? It is conceivable, that this is an automotive industry wide task that can only be solved in cooperation with all stakeholders

Biography: Born in 1959, Thomas Form studied Electrical Engineering at the University of Braunschweig, Germany, joined the Institute for Communication Engineering as research fellow in 1987 and received his Ph.D. in 1992. Up to 2002 he worked as a senior engineer in the Centre for Electromagnetic Compatibility of Volkswagen AG. In 2002 Dr. Form was appointed as the head of Telephone-/Telematics and Antenna systems development. He became a professor for "Electronic Vehicle Systems" in the Institute of Control Engineering at the Technical University Braunschweig in 2005 and participated with the CAROLO-Team in the finals of the DARPA URBAN Challenge 2007. From 2007 to 2009 he was responsible for concept development, module- and project management in the VW Electric/Electronic development. In 2009 he was appointed as head of the "Electronics and Vehicle Research" within Volkswagen Group research. Major achievements were the presentation of AUDI "Jack" vehicle driving in L3 automatic mode with Journalists from San Francisco to CES 2015 in Las Vegas and the presentation of the autonomous driving pod "SEDRIC" in 2017. Since 2016 he is the coordinator of the German national funded project PEGASUS which wants to answer the question "L3 Highway Chauffeur - how safe is safe enough and how to prove it". He got the Uni-DAS e.V. ADAS Award for significant influence on the development and introduction of driver assistance systems in 2017.


Keynote 2: AUTOSAR Adaptive - Challenging the Impossible

Speaker: Masaki Gondo, CTO at eSOL Co., Ltd., JP

Summary: The vast researches related to autonomous driving seem steadily progressing - it no longer makes news to just have some research vehicle drive autonomously. However, bringing this technology to the market, with all the associated legal, societal, and ethical responsibilities, with justifiable cost efficiency, is hard at its best, and impossible at its worst. Furthermore, the automotive industry is facing drastic challenges in electric vehicles, connected services, which also heavily impact the whole vehicle architecture.
AUTOSAR (AUTomotive Open System ARchitecture) is a worldwide development partnership of automotive interested parties. One of its latest challenges is to develop the software platform specification for the highly automated and autonomous driving, named AUTOSAR Adaptive Platform. This talk gives an overview of the challenges of such a platform, followed by the solution approach of AUTOSAR reflecting the industrial needs, and the overall architecture of AUTOSAR Adaptive. It also introduces a new multi-kernel OS technology the author develops, describing why such OS architecture is essential for coping with the challenge in the long run.

Biography: Masaki is CTO at eSOL, the company that provides POSIX/AUTOSAR/TRON RTOS, various software development tools, and various engineering services. Graduated from the State University of New York, Masaki has more than 20 years of experience in the field of OS architecture and related technologies for use in wide range of embedded system applications including automotive, industrial, and electronic appliances. He has authored/co-authored multiple popular Japanese books/international articles on OS and embedded systems, and given technical talks in conferences worldwide. While serving as a CTO directing the technology strategy and architecture at eSOL, he acts as one of the architects in Task Force Architecture of Adaptive Platform. His interest in recent years spans from next-generation OS/platform architectures, parallel processing, adaptive systems including machine learning, as well as scrum/kanban based development.


Agenda

TimeLabelSession
07:30W03.1Registration Desk opens
08:30W03.2Keynotes Session

Keynote 1: Challenges of Automated and Connected Driving
Speaker: Thomas Form, Head of Electronics and Vehicle Research, Volkswagen AG, DE

Keynote 2: AUTOSAR Adaptive - Challenging the Impossible
Speaker: Masaki Gondo, CTO at eSOL Co., Ltd., JP

09:50W03.3Interactive Presentations

IP1: A Dependable Detection Mechanism for Intersection Management of Connected Autonomous Vehicles
Rachel Dedinsky, Mohammad Khayatian, Mohammadreza Mehrabian and Aviral Shrivastava
Arizona State University, USA

IP2: A LIDAR Only Perception System for Autonomous Vehicle
Mohamed Yazid Lachachi, Abdelmalik Taleb-Ahmed, Smail Niar and Mohamed Ouslim
Université Polytechnique Hauts-de-France, FR

IP3: Generation of a Reconfigurable Probabilistic Decision Making Engine based on Decision Networks: UAV Case Study
Sara Zermani and Catherine Dezan
Lab-STICC, FR

10:00W03.4Coffee break 1
10:30W03.5Development Approaches for Autonomous Systems

10h30 Bringing the Next Generation Robot Operating System on Deeply Embedded Autonomous Platforms
Ralph Lange, Robert Bosch GmbH, DE

11h00 IDF-Autoware: Integrated Development Framework for ROS-based Self-driving Systems Using MATLAB/Simulink
Shota Tokunaga(1), Yuki Horita(2), Yasuhiro Oda(2) and Takuya Azumi(3)
(1)Graduate School of Engineering Science, Osaka University (2)Hitachi Automotive Systems, Ltd, (3)Graduate School of Science and Engineering, Saitama University, JP

11h20 Feasibility Study and Benchmarking of Embedded MPC for Vehicle Platoons
Inaki Martin Soroa(1), Amr Ibrahim(1), Dip Goswami(1) and Hong Li(2)
(1) Eindhoven University of Technology (2) NXP Semiconductor, NL

11h40 A Multiview Approach Toward Updatable Vehicle Automation Systems
Marcus Nolte, Mischa Möstl, Johannes Schlatow and Rolf Ernst
Technical University of Braunschweig, DE

12:00W03.6Lunch break
13:00W03.7 Dependable Autonomous Systems

13h00 Autonomous Data Center - Feedback Control for Predictable Cloud Computing
Martina Maggio, University of Lund, SE

13h30 Fault-Tolerance by Graceful Degradation for Car Platoons
Mohammed Baha E. Zarrouki(1), Verena Klös(1), Markus Grabowski(2) and Sabine Glesner(1)
(1)Technische Universität Berlin, DE, (2)Assystem Germany GmbH

13h50 Safety and Security Analysis of AEB for L4 Autonomous Vehicle using STPA
Shefali Sharma(1), Adan Flores(1), Chris Hobbs(2), Jeff Stafford(3) and Sebastian Fischmeister(1)
(1)University of Waterloo, CA (2)QNX Software Systems Limited,CA (3) Renesas Electronics America Inc.

14h10 Towards a Formal Model of Recursive Self-Reflection
Axel Jantsch
TU Wien, AT


14:30W03.8Coffee break 2
15:00W03.9Research Clusters on Autonomous Systems

15h00 An approach to automotive service-oriented software architectures in a multi-partner research project
Stefan Kowalewski, RWTH Aachen, DE

15h30 Controlling Concurrent Change- Design Automation for Critical Systems Integration
Rolf Ernst, TU Braunschweig, DE

16h00 Panel Discussion

16:30W03.10Closing & Exhibition
17:30W03.11Workshops end