Tutorial on the Arbitrated Networked Control Systems Approach to Cyber-Physical Systems

CPSWeek 2011 (April 11, 2011)

Presenters:
Anuradha Annaswamy (Massachusetts Institute of Technology, USA, aanna@mit.edu)
Samarjit Chakraborty (TU Munich, Germany, samarjit@tum.de)
Dip Goswami (TU Munich, Germany, dip@rcs.ei.tum.de)

Abstract

This tutorial will focus on “control/architecture co‐design” in settings where multiple control applications are implemented on a distributed architecture. Such architectures will be assumed to consist of multiple processors communicating over a system made up of several buses and gateways implementing different protocols. Such a setting arises specifically in the context of automotive cyber-physical architectures that now consist of  80-100 ECUs (electronic control units) connected through multiple buses, such as FlexRay, CAN and Ethernet. In this tutorial, we will discuss implementations of control applications in distributed architectures. We will quantify the semantic gap between control models and their implementations that occur due to underlying assumptions in the modeling process. We will then illustrate several co-design procedures where the control designs are in sync with the implementation architecture and the architecture design efficiently accommodates all control performance specifications, stability and tracking performance in the transient and steady states.

There has been a significant amount of work in the area of networked control systems (NCS), which primarily deal with control applications communicating over wireless networks, where messages suffer loss, delay and jitter. In contrast to such traditional NCS, the cyberphysical systems that we focus on have a distributed architecture where questions related to the arbitration of control tasks (e.g., how to partition and map the control tasks) with a synergistic design of the underlying implementation architecture (e.g., how many ECUs and how are they connected?). The issue of co‐design in such settings becomes considerably more important, instead of merely taking into account the properties of the wireless network in designing the control models. Apart from discussing the necessary analysis and synthesis methods for this co-design, we will also present several examples and realistic case studies from the automotive electronics and software domain in this tutorial.

Tutorial outline

Control Theory Fundamentals [1, 2, 3, 4, 5, 9] (Anuradha Annaswamy): The first part of the tutorial will cover fundamental principles of control theory, control performance adaptive control theory and relevant control tools for  the co-design of control and implementation architecture.

Communication Architectures [8, 9, 10, 11, 12] (Samarjit Chakraborty): The second part of the tutorial will talk about automotive architectures, various communication bus protocols such as FlexRay and CAN, how to formally analyze such architectures and various design and optimization techniques that arise in this domain.

Control/Communication Co-design [6, 7, 8, 9] (Dip Goswami): The final part will discuss various co‐design techniques; more specifically, topics such as how to design architectures for several control applications each with multiple control performance metrics, how to design controllers that can tolerate bounded message deadline violations and how to exploit hybrid time‐ and event‐triggered architectures for designing control applications.

References

  1. K. J. Åström & R. M. Murray, "Feedback Systems: An Introduction for Scientists and Engineers," Princeton University Press, 2008.
  2. R. C. Dorf and R.H. Bishop, "Modern Control Systems," Pearson Prentice Hall, 2008.
  3. K. S. Narendra and A.M. Annaswamy, "Stable Adaptive Systems," Dover Publications, 2004.
  4. W. Zhang, M.S. Branicky, and S.M. Phillips, "Stability of networked control systems," IEEE Control Systems Magazine, 21(1):84-99, 2001.
  5. H. Voit and A.M. Annaswamy, "Adaptive Control of a Networked Control System with Hierarchical Scheduling," Technical Report, Active-adaptive Control Laboratory, MIT, September 2010.
  6. D. Goswami, R. Schneider, S. Chakraborty, "Re-engineering Cyber-Physical Control Applications for Hybrid Communication Protocols," In 14th Proceedings of the Conference for Design, Automation and Test in Europe (DATE), Grenoble, France, 2011.
  7. D. Goswami, R. Schneider, S. Chakraborty, "Co-design of Cyber-Physical Systems via Controllers with Flexible Delay Constraints," In 16th Asia and South Pacific Design Automation Conference (ASP-DAC), Yokohama, Japan, 2011.
  8. R. Schneider, U. Bordoloi, D. Goswami, S. Chakraborty, "Optimized Schedule Synthesis under Real-Time Constraints for the Dynamic Segment of FlexRay," In 8th International Conference on Embedded and Ubiquitous Computing (EUC), Hong Kong SAR, China, 2010.
  9. H. Voit, R. Schneider, D. Goswami, A. Annaswamy, S. Chakraborty, "Optimizing Hierarchical Schedules for Improved Control Performance," In 5th International Symposium on Industrial Embedded Systems (SIES), Trento, Italy, 2010.
  10. S. Chakraborty, S. Künzli, L. Thiele, "A General Framework for Analysing System Properties in Platform-Based Embedded System Designs," DATE, 2003.
  11. F. Zhang, K. Szwaykowska, W. Wolf, V. J. Mooney III "Task Scheduling for Control Oriented Requirements for Cyber-Physical Systems," IEEE Real-Time Systems Symposium 2008: 47-56
  12. H. Yazarel, A. Girard, G. J. Pappas, R. Alur "Quantifying the Gap between Embedded Control Models and Time-Triggered Implementations," RTSS 2005: 111-120

Target audience

This tutorial will be targeted towards an audience who has some background in embedded systems design but is new to control theory, automotive architectures and cyber‐physical systems. The level of the tutorial will be from introductory to intermediate and no background in advanced system‐level design techniques will be assumed. The material to be presented will be useful to researchers from the embedded systems domain, students, and software developers who are interested in automotive‐ and cyber‐physical systems‐related methods and design questions.

Biographies

Anuradha Annaswamy received her Ph.D. in Electrical Engineering from Yale University in 1985. She has been a member of the faculty at Yale, Boston University, and MIT where currently she is the director of the Active‐Adaptive Control Laboratory and a Senior Research Scientist in the Department of Mechanical Engineering. Her research interests pertain to adaptive control theory and applications to aerospace and automotive control, active control of noise in thermo‐fluid systems, active emission control, control of autonomous systems, control of smart grid‐oriented markets and integration with renewables, and co‐design of control and implementation in complex systems. She has authored numerous journal and conference papers and a graduate textbook on adaptive systems. Dr. Annaswamy has received several awards including the George Axelby Outstanding Paper award from the IEEE Control Systems Society, the Presidential Young Investigator award from the National Science Foundation, and the Hans Fisher Senior Fellowship from the Institute for Advanced Study at the Technische Universität München. Dr. Annaswamy is a Fellow of the IEEE and a member of AIAA.

Samarjit Chakraborty is a Professor of Electrical Engineering at TU Munich in Germany, where he heads the Institute for Real‐Time Computer Systems (RCS). Prior to joining TU Munich, from 2003 –2008 he was an Assistant Professor of Computer Science at the National University of Singapore. He obtained his Ph.D. in Electrical and Computer Engineering from ETH Zurich in 2003. His research interests cover all aspects of system‐level design of real‐time embedded systems and software, and his work has attracted funding from companies like Intel, General Motors, Bosch and BMW. Dr. Chakraborty regularly serves on the technical program committees of several conferences in the area of real‐time and embedded systems, such as DATE, CODES+ISSS, ASP‐DAC, RTSS and RTAS. He was the TPC Co‐Chair of EMSOFT 2009, and is currently serving as the General Co‐Chair of Embedded Systems Week 2010. Apart from more than 80 referred journal and conference articles, he has authored a number of patents and regularly gives invited talks and tutorials at various research labs and international conferences (such as ESWeek, VLSI Design, ACM Multimedia and ICME). For his Ph.D. thesis, he received the ETH Medal and the European Design and Automation Association’s “Outstanding Doctoral Dissertation Award” in 2004. His work has also received a HiPEAC Paper Award in 2009 and Best Paper Award nominations at DAC 2005, CODES+ISSS 2006, ECRTS 2007, and CODES+ISSS 2008.

Dip Goswami obtained his Ph.D. in Electrical and Computer Engineering from the National University of Singapore (NUS) in 2008. Subsequently, he was a post‐doctoral fellow at the School of Computing of NUS in 2009, where he worked on a General Motors funded project. He is currently an Alexander von Humboldt Postdoctoral Fellow at TU Munich, Germany and is working on co‐design techniques for cyber‐physical systems with an emphasis on automotive architectures and control software. He has published in several international journals and conferences in the fields of control systems, robotics and cyber‐physical systems. He has also served on the TPCs of the IEEE Real‐Time Systems Symposium (RTSS) 2010, the International Conference on Computational Intelligence, and Robotics and Autonomous Systems (CIRAS) 2010.