Full Program – Special Initiative on Autonomous Systems Design at DATE 2023
Monday/Tuesday 17-18 April
Second Opening Keynote: (Monday April 17: 11h – 12h30)
Speaker: Dirk Elias from Robert Bosch GmbH.
Title: The Cyber-Physical Metaverse – Where Digital Twins and Humans Come Together
The concept of Digital Twins (DTs) has been discussed intensively for the past couple of years. Today we have instances of digital twins that range from static descriptions of manufacturing data and material properties to live interfaces to operational data of cyber physical systems and the functions and services they provide.
Currently, there are no standardized interfaces to aggregate atomic DTs (e.g., the twin of the lowest-level function of a machine) to higher-level DTs providing more complex services in the virtual world. Additionally, there is no existing infrastructure to reliably link the DTs in the virtual world to the integrated CPSs in the real world (like a car consisting of many ECUs with even more functions).
This keynote will address how the Metaverse can become the virtual world where DTs of humans and machines live and how to reliably connect DTs to the physical world. Insights in current activities of Bosch Research and its academic partners to move towards this vision will be provided.
Later with the Keynote Speaker: (Monday April 17: 9:45h – 10h30)
Speaker: Dirk Elias from Robert Bosch GmbH.
Prof. Dr.-Ing. Dirk Christian Elias is currently Senior Vice President at Robert Bosch GmbH and heading the Corporate Research Division ‘Advanced Digital’ (CR/AD). The division is focusing its applied research on ICT, SW development and systems engineering and simulation. He also is responsible for the Bosch Research and Technology Center in India. After studying EE at TU Munich he started his career at Fraunhofer Fokus in 1992. In 1996 he received his PhD from TU Berlin. In 2000 he left Fraunhofer and started IVISTAR AG. IVISTAR focused on IoT products for the organization of open offices and provided later also development services for smart home products. In 2008 he left Berlin and started setting up the Fraunhofer Portugal Research Center AICOS and joined the Faculty of Engineering of the University of Porto (FEUP) as invited full professor.
ASD Technical Session 1 (Monday April 17: 11h – 12h30)
Title: Designing Fault tolerant and resilient autonomous systems
Talk 1: MAVFI: An End-to-End Fault Analysis Framework with Anomaly Detection and Recovery for Micro Aerial Vehicles Yu-Shun Hsiao, Zishen Wan, Tianyu Jia, Radhika Ghosal, Abdulrahman Mahmoud, Arijit Raychowdhury, David Brooks, Gu-Yeon Wei and Vijay Janapa Reddi (firstname.lastname@example.org)
Talk 2: Phalanx: Failure-Resilient Truck Platooning System Changjin Koo, jaegeun park, Ahn TaeWook, Hongsuk Kim, Jong-Chan Kim and Yongsoon Eun (email@example.com)
Talk 3: Efficient Software-Implemented HW Fault Tolerance for TinyML Inference in Safety-critical Applications Uzair Sharif, Daniel Mueller-Gritschneder, Rafael Stahl and Ulf Schlichtmann (firstname.lastname@example.org)
Talk 4: Formal Analysis of Timing Diversity for Autonomous Systems Anika Christmann, Robin Hapka and Rolf Ernst (email@example.com)
ASD Special Session 1: (Monday April 17: 14h – 15h30)
Title: Information Processing Factory, Take Two on Self-Aware Systems of MPSoCs
Organizers: Fadi Kurdahi (UCI)
Chair: Bryan Donyanavard, San Diego State University, US
Co-Chair: Smail Niar, UPHF, FR
Abstract: The Information Processing Factory (IPF) project is a collaboration between research teams in the US (UC Irvine) and Germany (TU Munich and TU Braunschweig) looking into Self-aware MPSoCs. IPF 1.0, was first introduced in ESWEEK 2016 as a paradigm to master complex dependable systems. The IPF paradigm applies principles inspired by factory management to the continuous operation and optimization of highly-integrated embedded systems. IPF 2.0 is an extension of the IPF for recent data-centric approaches and decentralization methodologies. While an IPF 1.0 system can operate independently, IPF 2.0 has a system-of-systems structure in which several IPF 1.0 “factories” interact, thus providing an additional layer of abstraction aimed at this data-centric approach. It horizontally extends core concepts such as self-optimization, self-construction, and runtime verification, while maintaining the strengths of the existing IPF methodology. Four talks in this session highlight the various concepts in IPF 2.0 illustrated through a truck platooning exemplar.
The talks outline the challenges introduced when moving from self-organizing local systems in IPF 1.0 to autonomous systems collaboration in IPF 2.0, using commercial vehicle platooning as a use case. The first talk explains how the self-aware truck control systems collaborate towards a platoon-level runtime verification that continuously supervises the state of a platoon, even under a changing platoon formation and external disturbance, e.g., by intersecting traffic participants. The second talk outlines the challenges related to managing enormous amounts of dynamic data in the system, and discusses how self-aware caching can help in mastering the resulting communication and data management requirements. The third talk proposes approaches to mitigate the energy cost of data management across multiple systems. The fourth talk addresses lack of explainability in the underlying machine learning technology in collaborative autonomous systems.
Talk 1: Trust, But Verify: Towards Self-Aware, Safe, Autonomous Self-Driving Systems, Fadi Kurdahi (UCI)
Talk 2: Vehicle as a Cache – A Data Centric Platform for the IPF Paradigm, Rolf Ernst (TUBS)
Talk 3: Computational Self-Awareness for Energy-Efficient Memory Systems, Nikil Dutt (UCI)
Talk 4: Learning Classifier Tables – Turning ML Decision Making Explainable, Andreas Herkersdorf (TUM)
ASD Technical Session 2 (Monday April 17: 16h30 – 18h00)
Title: Autonomy for systems perception, control and optimization
Talk 1: Autonomous Hyperloop Control Architecture Design using MAPE-K, Julian Demicoli, Laurin Prenzel and Sebastian Steinhorst
Talk 2: Reinforcement-Learning-Based Job-Shop Scheduling for Intelligent Intersection Management Shao-Ching Huang, Kai-En Lin, Cheng-Yen Kuo, Li-Heng Lin, Muhammed Omer Sayin and Chung-Wei Lin
Talk 3: Bio-inspired Autonomous Exploration Policies with CNN-based Object Detection on Nano-drones, Lorenzo Lamberti, Luca Bompani, Victor Javier Kartsch Morinigo, Manuele Rusci, Daniele Palossi and Luca Benini
Talk 4: Butterfly Effect Attack: Tiny and Seemingly Unrelated Perturbations for Object Detection Nguyen Anh Vu Doan, Arda Yueksel and Chih-Hong Cheng
ASD Panel Session (Monday April 17: 18h30 – 20h00)
Title: : Autonomous Systems Design as a Driver of Innovation?
Organizers: Rasmus Adler, Peter Liggesmeyer ( Fraunhofer IESE, Germany)
Abstract: Autonomous systems have high potential in many application domains. However, most discussions seem to take place with respect to autonomous road vehicles. Automotive industry promised substantial progress in this field but many predictions have not come true. Companies stepped back and corrected their predictions. Does this mean, systems autonomy is not ready to drive innovation? However, autonomous behavior is obviously not limited to road vehicles. Various kinds of systems can benefit from autonomous behavior in various domains such as health and pharmaceutics, energy, manufacturing, farming, mining and so on. In this session, we will thus take a broader perspective on autonomous system design as a driver of innovation and discuss benefits, challenges, and risks in various application domains.
Prof. Karl-Erik Årzén https://portal.research.lu.se/en/persons/karl-erik-%C3%A5rz%C3%A9n
Prof. Martin Fränzle – OFFIS e. V., Scientific Director Division Transportation
Dr. Arne Hamann- Chief Expert Distributed Intelligent Systems, Corporate Research Robert Bosch GmbH
Prof. Davy Pissoort – KU Leuven Faculty of Engineering Technology
Dr. Claus Bahlmann – Head of R&D Department AI, Siemens, Mobility Division, Technology & Innovation
Dr. Christoph Schulze – Technology Expert from “the Autonomous”
ASD focus Session 1 (Tuesday April 18: 08h30 – 10h00)
Title: : Autonomy-driven Emerging Directions in Software-defined Vehicles
Organizers: Markus Joachim and S. Ramesh (General Motors, USA)
Abstract: Over the past two decades, the volume of electronics and software in cars have grown tremendously. There is now widespread consensus that more than 90% of the innovation in modern vehicles is driven by them. But this growth has also resulted in hardware and software architectures that are proving to be a bottleneck for further innovation and efficient design flows, especially when implementing compute-intensive functions necessary for autonomous features. Another emerging trend in the domain of automotive software is the need for continuous improvement and continuous deployment (CI/CD) of functionality, that is enabled by Over-The-Air (OTA) capability. The goal of this special session is to discuss these new trends, the resulting challenges, and explore emerging solutions and directions in the broad area of design, development, and verification of software-defined vehicles. The three talks will highlight different aspects of software-defined vehicle designs, what research challenges they pose, and how they would impact the future automotive design ecosystem.
Talk1: Impacts of Service Oriented Communication on SDV architectures, Prachi Joshi (General Motors R&D, USA)
Talk 2: “Shift-Left” Development and Validation of Software Defined Vehicles with a Virtual Platform, Unmesh Bordoloi (Siemens, USA)
Talk 3: Design Tools for Assured Autonomy, Samarjit Chakraborty (UNC Chapel Hill, USA)
ASD focus Session 2 (Tuesday April 18: 11h – 12h30)
Title: SelPhys: Self-awareness in Cyber-physical Systems
Organizers: Lukas Esterle (Aarhus University, DK), Axel Jantsch (TU Wien, AT)
Abstract: Computational self-awareness enables autonomous systems to operate in rapidly unfolding situations and conditions that have not been considered during development. Cyber-physical systems, constantly interacting with the physical world, have to deal with an even wider spectrum of potentially unknown situations introduced in their environment, including other (autonomous) systems and humans. Their ability to respond appropriately is vital for these systems not only to achieve their goals but also to ensure the safety of other machines and humans in the process. In this special session, we will have various invited talks on different aspects of computational self-awareness and its contribution to autonomous systems design. Specifically, we aim to have talks ranging from fundamental theory on computational self-awareness, over signal processing, embedded and high-performance computing, towards applications utilising self-aware properties for increased safety and performance. After the short presentations, the presenters will be invited to participate in a panel discussion together with the audience.
Talk1: “Self-Aware Machine Intelligence”, Peter R. Lewis (Ontario University of Technology, Canada)
Talk2: ”Incremental Self-Awareness Based on Free Energy Minimization for Autonomous Agents”, Carlo Regazzoni and and Lucio Marcenaro (University of Genova)
Talk3: “Adaptive, Resilient Computing Platforms through Self-Awareness”, Nikil Dutt (University of California, Irvine)
Talk4: “Cognitive Energy Systems”:, Christian Gruhl (University of Kassel)
ASD Workshop (Tuesday April 18: 14h – 18h00): Can Autonomous Systems Be Safe?
Abstract: Despite the advancement of machine learning and artificial intelligence, safety still constitutes the main hurdle for supporting high levels of autonomy in domains such as self-driving cars, where more than 400* car accidents** involving autonomous functionalities were reported last year*. The design process according to safety standards typically assumes a correct and complete system specification. For autonomous systems, it is often impossible to show that the specification is complete, e.g. due to the underspecified environment and evolving behavior. This extremely challenges current safety engineering practices to reason about risk and uncertainty at operation time for decision making.
**Actually according to the 2021 autonomous vehicle disengagement report, almost 3000 accidents/incidents were reported.