Complex Embedded Software and Systems for Multi-domain Integration | CNCC2021

Li Yahui

Postdoctoral Researcher at Xi’an University of Electronic Science and Technology

Speaker Introduction: Li Yahui, PhD, postdoctoral researcher at Xi’an University of Electronic Science and Technology, researcher, and director of the Pre-research Office at the Aviation Industry Computing Institute. His main research directions are computer system architecture, artificial intelligence, and information security, undertaking several national key research and development projects, and has received awards including two provincial first prizes and two provincial second prizes.

Report Title: On-board Cloud Computing Architecture

Report Abstract: Aiming at the future civil aviation field’s demand for free flight, combined with the intelligent development trend of on-board systems, this report discusses how to organize and manage the resources of heterogeneous on-board platforms through cloud computing, and conducts research on capability requirements, core issues, and development trends based on cloud computing theories, proposing an on-board cloud computing architecture for collaboration between aircraft, and further analyzing the information sharing and functional organization requirements within the cloud system.

Qiao Lei

Researcher at the 502nd Institute of the Fifth Academy of Aerospace Science and Technology

Speaker Introduction: Qiao Lei, PhD, researcher at the 502nd Institute of the Fifth Academy of Aerospace Science and Technology, and a senior member of the CCF. He is a young top talent of the National Ten Thousand Plan and the China Aerospace Science and Technology Corporation. He is the winner of the 2019 NASA Young Software Innovation Award and a visiting scholar at Yale University. He is a committee member of the System Software, Formal Methods, and Industrial Control Computer Committees of the China Computer Federation. He has been committed to the design, verification, and trusted assurance technology research of operating system software. His research results have been directly applied and played a key role in significant model tasks such as China’s Yutu-1 and Yutu-2 lunar rovers. He has published over 50 papers, co-authored a monograph, and applied for 20 national invention patents and one international invention patent. He has hosted or participated in over ten national major research projects, including the National Natural Science Foundation, nuclear high base, 863, and equipment development pre-research. As a core member, he developed China’s first spacecraft operating system, SpaceOS, which has been applied in over 100 spacecraft including manned spaceflight, lunar exploration, and Beidou navigation. He has received multiple awards including the National Defense Science and Technology Innovation Team Award, Beijing Science and Technology Progress First Prize, and Military Science and Technology Progress Third Prize.

Report Title: Research and Practice of Spacecraft Operating Systems

Report Abstract: This report analyzes and summarizes the development history and application status of embedded operating systems in China’s aerospace field. It focuses on introducing the design and verification methods of the SpaceOS embedded operating system with independent intellectual property rights developed by the 502nd Institute of the Fifth Academy of Aerospace Science and Technology and looks forward to the future development prospects of spacecraft operating systems.

Spacecraft are resource-constrained safety-critical systems, with increasing demands for autonomous intelligence. Faced with external environmental and internal state changes without altering hardware resources, the system functionality needs to be dynamically expanded and reconstructed while ensuring correctness and real-time performance. This report studies the real-time scheduling of spacecraft tasks, admission control, and formal verification as key issues, providing real-time assurance mechanisms for the execution of various real-time tasks in spacecraft, and has been applied in actual spacecraft systems.

Chang Wanli

Professor at Hunan University

Speaker Introduction: Chang Wanli, professor at Hunan University, doctoral supervisor, selected for the National Overseas High-level Talent Introduction Program in 2020, Huawei senior expert consultant, secretary-general of the ACM Embedded Systems Committee SIGBED, and executive member of the Design Automation Committee SIGDA. Previously, he served as a tenured professor at the University of York in the UK, deputy director of the National Laboratory for Real-Time Systems, and expert consultant for Bosch. He graduated with a PhD from the Technical University of Munich in Germany, receiving the Best Doctoral Thesis Award, and obtained his bachelor’s degree from Nanyang Technological University in Singapore. In the past three years, he has published 19 CCF A-level papers and has received eight nominations for best paper awards at CCF B-level conferences. He serves as the chair of the DAC Automatic Systems Field and as a member of the RTSS Paper Award Committee, and is an expert reviewer for the British Computer Society Outstanding Paper Award.

Report Title: Hardware/Software Co-Synthesis and Co-Optimization for Autonomous Systems

Report Abstract: With ever more complicated functionalities being integrated in modern autonomous systems, traditional design methods may not remain sufficient to deliver trusted and high-performance systems with stringent temporal, safety, and cost efficiency requirements. In this talk, we discuss the limitations of the traditional design methods with the above requirements enforced, in which hardware and software design are often considered separately. To tackle these limitations, this talk presents a novel design solution that synthesizes both software-level and hardware-level design. First, we highlight and analyze the interconnections between software-level methods (e.g., priority assignment and task allocation) and hardware design (e.g., cache and memory management), in terms of the resulting system performance, e.g., latency. Second, by applying the identified interconnections, we propose an optimization framework to produce high-quality synthesized solutions of both software and hardware design based on a set of candidate design methods. In addition, we describe potential research directions derived from the work and major challenges that can be investigated jointly by engineers and researchers from embedded systems, system safety, and programming languages communities.

Bu Lei

Secretary-General of the CCF System Software Committee, Professor at Nanjing University

Speaker Introduction: Bu Lei, professor at the Department of Computer Science and Technology at Nanjing University, doctoral supervisor. His main research areas are software engineering and formal methods, including model checking technology, real-time hybrid systems, and cyber-physical integration systems. He obtained his PhD in Computer Science from Nanjing University in 2010. He has conducted research and cooperation at research institutions such as CMU, MSRA, UTD, and FBK. His related work has been published in important journals and conferences in the field, such as TCAD, TC, TDSC, TCPS, TPDS, RTSS, CAV, and HSCC. He has been selected for the NASAC Young Software Innovation Award, the Excellent Teacher Award Program for Computer Majors in Higher Education, the China Computer Federation Young Talent Development Program, and the Microsoft Research Asia Star Program.

Report Title: Real-Time System Runtime Safety Monitoring and Control Generation Based on Online Verification

Report Abstract: Real-time hybrid systems are widely present in safety-critical areas such as rail control, aerospace, aviation, and intelligent driving, where the correctness of their behavior is crucial. However, the internal behavior of real-time hybrid systems is extremely complex due to the intertwining of discrete and continuous behaviors; and the dynamic uncertainty of the external environment further exacerbates the complexity of the related system behaviors, making safety assurance difficult. To address this issue, we propose a formal verification-based method for quickly modeling and verifying the system’s short-term behavior at runtime, thus discovering and warning about potential unsafe behaviors. On this basis, we further generate automatic control parameters for fine-grained runtime control, taking over system control when potential dangers are detected, ensuring operational safety.

Yang Pengfei

Associate Professor at Xi’an University of Electronic Science and Technology

Speaker Introduction: Yang Pengfei, associate professor and deputy director of the Computer Engineering Department at Xi’an University of Electronic Science and Technology, and an executive member of the CCF Embedded Systems Committee. He is mainly engaged in research on embedded system architecture and embedded intelligent computing, undertaking over ten projects including the National Natural Science Foundation, the Intelligent Eye Project for Equipment Development, sub-projects of the National Key Research and Development Program, and key research and development projects in Shaanxi Province, as well as industry cooperation projects.

Report Title: Design and Development of Embedded Systems under Multi-Dimensional Capability Objectives

Report Abstract: The “ubiquitous” and “omnipotent” intelligent embedded systems are redefining personal lives, disrupting industry models, and accelerating social development. Future embedded systems must possess multi-dimensional capabilities such as high performance, high safety, self-organization, and intelligence, which puts higher demands on the design and development of embedded systems. This report reviews the current status and challenges of embedded systems in architecture design, system resource management, code compilation and deployment, and intelligent application development, discusses ideas and methods to address new challenges in system development, and introduces the team’s exploratory work in this area.

Jiang Wei

Associate Professor at University of Electronic Science and Technology

Speaker Introduction: Jiang Wei, associate professor and doctoral supervisor at the University of Electronic Science and Technology, and a high-level overseas talent in Sichuan Province. He completed his undergraduate, master’s, and doctoral studies at the School of Computer Science and Engineering at the University of Electronic Science and Technology and graduated with a PhD in Computer Application Technology in 2009. He received the National Foundation for Foreign Students and the Danish Government Cultural Exchange Scholarship, conducting a year-long embedded systems research at DTU University in Denmark from 2011 to 2012. In April 2012, he conducted short-term research at the ESLAB laboratory of Linköping University in Sweden. He serves as the secretary of the IEEE CEDA Chengdu Branch, deputy editor of the JCSC journal, guest editor for journals such as JSA and JEST, and chair of the ICITES 2021 conference.

His research focuses on trustworthy systems, leading multiple research projects on safety-critical real-time systems, trustworthy artificial intelligence, and fault tolerance in deep learning, including natural foundation projects and projects from Tencent in Shenzhen. He has published over 40 academic papers in journals such as IEEE TCAD, ACM TCPS, EMSOFT, and CODES+ISSS, and has received honors such as the IEEE NAS Best Student Paper Award and the Third Prize for Scientific and Technological Progress from the China Electric Power Association.

Report Title: Discussion on Fault-Tolerant Enhancement Technologies for Safety-Critical Areas such as Intelligent Unmanned Systems

Report Abstract: Artificial intelligence technology will be widely applied in safety-critical areas such as intelligent unmanned systems, aerospace, and intelligent industry. Intelligent systems in safety-critical areas face severe “internal and external troubles”; in addition to facing errors and threats from hacker attacks, they must also contend with errors and threats arising from the actual deployment environment and their own characteristics. If errors and threats are not properly handled, it will lead to the failure or collapse of intelligent systems, resulting in severe property and life losses, and even catastrophic disasters. This report will introduce and discuss fault-tolerant enhancement technologies for artificial intelligence, deep learning fault tolerance technologies, and system-level fault-tolerant design technologies suitable for intelligent unmanned systems.