You are previewing Co-design Approaches to Dependable Networked Control Systems.
O'Reilly logo
Co-design Approaches to Dependable Networked Control Systems

Book Description

This book describes co-design approaches, and establishes the links between the QoC (Quality of Control) and QoS (Quality of Service) of the network and computing resources. The methods and tools described in this book take into account, at design level, various parameters and properties that must be satisfied by systems controlled through a network. Among the important network properties examined are the QoC, the dependability of the system, and the feasibility of the real-time scheduling of tasks and messages. Correct exploitation of these approaches allows for efficient design, diagnosis, and implementation of the NCS. This book will be of great interest to researchers and advanced students in automatic control, real-time computing, and networking domains, and to engineers tasked with development of NCS, as well as those working in related network design and engineering fields.

Table of Contents

  1. Cover
  2. Title Page
  3. Copyright
  4. Foreword
  5. Introduction and Problem Statement
    1. I.1. Networked control systems and control design challenges
    2. I.2. Control design: from continuous time to networked implementation
    3. I.3. Timing parameter assignment
    4. I.4. Control and task/message scheduling
    5. I.5. Diagnosis and fault tolerance in NCS
    6. I.6. Co-design approaches
    7. I.7. Outline of the book
    8. I.8. Bibliography
  6. Chapter 1: Preliminary Notions and State of the Art
    1. 1.1. Overview
    2. 1.2. Preliminary notions on real-time scheduling
      1. 1.2.1. Some basic results on classic task model scheduling
        1. 1.2.1.1. Fixed priority scheduling
        2. 1.2.1.2. EDF scheduling
        3. 1.2.1.3. Discussion
      2. 1.2.2. (m, k)-firm model
    3. 1.3. Control aware computing
      1. 1.3.1. Off-line approaches
      2. 1.3.2. Quality of Service and flexible scheduling
    4. 1.4. Feedback-scheduling basics
      1. 1.4.1. Control of the computing resource
        1. 1.4.1.1. Control structure
        2. 1.4.1.2. Sensors and actuators
        3. 1.4.1.3. Control design and implementation
      2. 1.4.2. Examples
        1. 1.4.2.1. Feedback scheduling a web server
        2. 1.4.2.2. Optimal control-based feedback scheduling
        3. 1.4.2.3. Feasibility: feedback-scheduler implementation for robot control
          1. 1.4.2.3.1. Plant modeling and control structure
          2. 1.4.2.3.2. Scheduling controller design
          3. 1.4.2.3.3. Implementation of the feedback scheduler
    5. 1.5. Fault diagnosis of NCS with network-induced effects
      1. 1.5.1. Fault diagnosis of NCS with network-induced time delays
        1. 1.5.1.1. Low-pass post-filtering
        2. 1.5.1.2. Structure matrix of network-induced time delay
        3. 1.5.1.3. Robust deadbeat fault filter
        4. 1.5.1.4. Other work
      2. 1.5.2. Fault diagnosis of NCS with packet losses
        1. 1.5.2.1. Deterministic packet losses
        2. 1.5.2.2. Stochastic packet losses
      3. 1.5.3. Fault diagnosis of NCS with limited communication
      4. 1.5.4. Fault-tolerant control of NCS
    6. 1.6. Summary
    7. 1.7. Bibliography
  7. Chapter 2: Computing-aware Control
    1. 2.1. Overview
    2. 2.2. Robust control w.r.t. computing and networking-induced latencies
      1. 2.2.1. Introduction
      2. 2.2.2. What happens when delays appear?
        1. 2.2.2.1. Initial conditions
        2. 2.2.2.2. Infinite dimensional systems
      3. 2.2.3. Delay models
      4. 2.2.4. Stability analysis of TDS using Lyapunov theory
        1. 2.2.4.1. The second method
        2. 2.2.4.2. The Lyapunov–Razumikhin approach
        3. 2.2.4.3. The Lyapunov–Krasovskii approach
      5. 2.2.5. Summary: time-delay systems and networking
    3. 2.3. Weakly hard constraints
      1. 2.3.1. Problem definition
      2. 2.3.2. Notion of accelerable control
      3. 2.3.3. Design of accelerable controllers
      4. 2.3.4. Accelerable LQR design for LTI systems
      5. 2.3.5. Kalman filtering and accelerability
      6. 2.3.6. Robustifying feedback scheduling using weakly hard scheduling concepts
      7. 2.3.7. Application to the attitude control of a quadrotor
    4. 2.4. LPV adaptive variable sampling
      1. 2.4.1. A polytopic discrete-plant model
      2. 2.4.2. Performance specification
      3. 2.4.3. LPV/H∞ control design
      4. 2.4.4. Experimental assessment
    5. 2.5. Summary
    6. 2.6. Bibliography
  8. Chapter 3: QoC-aware Dynamic Network QoS Adaptation
    1. 3.1. Overview
    2. 3.2. Dynamic CAN message priority allocation according to the control application needs
      1. 3.2.1. Context of the study
        1. 3.2.1.1. The considered process control application
        2. 3.2.1.2. Control performance evaluation
        3. 3.2.1.3. The implementation through a network
          1. 3.2.1.3.1. Structure
          2. 3.2.1.3.2. Choice of the sampling period
          3. 3.2.1.3.3. Considering the network CAN
        4. 3.2.1.4. Evaluation of the influence of the network on the behavior of the process control application
          1. 3.2.1.4.1. Dedicated network CAN
          2. 3.2.1.4.2. Shared network CAN: considering static priorities and showing their inadequacy
        5. 3.2.1.5. Idea of hybrid priority schemes: general considerations
          1. 3.2.1.5.1. The identifier (ID) field and the scheduling execution
          2. 3.2.1.5.2. Cohabitation of flows with constant needs and flows of process control applications (variable needs)
          3. 3.2.1.5.3. Toward making dynamic priorities
      2. 3.2.2. Three hybrid priority schemes
        1. 3.2.2.1. hp scheme
        2. 3.2.2.2. (hp+sts) scheme
        3. 3.2.2.3. (hp+dts) scheme
          1. 3.2.2.3.1. Main ideas
          2. 3.2.2.3.2. Algorithm for computing the dynamic priority at any sampling instant tk
      3. 3.2.3. Study of the three schemes based on hybrid priorities
        1. 3.2.3.1. Study conditions
        2. 3.2.3.2. hp scheme
        3. 3.2.3.3. (hp+sts) scheme
        4. 3.2.3.4. (hp+dts) scheme
      4. 3.2.4. QoC visualization
      5. 3.2.5. Comment
    3. 3.3. Bandwidth allocation control for switched Ethernet networks
      1. 3.3.1. NCS performance analysis
      2. 3.3.2. NCS modeling
        1. 3.3.2.1. Introduction
        2. 3.3.2.2. Network modeling
        3. 3.3.2.3. System modeling
        4. 3.3.2.4. Controller modeling
      3. 3.3.3. Network adaptation mechanism
      4. 3.3.4. Example
        1. 3.3.4.1. Maximum delay computation
        2. 3.3.4.2. Results
    4. 3.4. Conclusion
    5. 3.5. Bibliography
  9. Chapter 4: Plant-state-based Feedback Scheduling
    1. 4.1. Overview
    2. 4.2. Adaptive scheduling and varying sampling robust control
      1. 4.2.1. Extended elastic tasks controller
      2. 4.2.2. Case study
    3. 4.3. MPC-based integrated control and scheduling
      1. 4.3.1. Resource constrained systems
      2. 4.3.2. Optimal integrated control and scheduling of resource constrained systems
    4. 4.4. A convex optimization approach to feedback scheduling
      1. 4.4.1. Problem formulation
      2. 4.4.2. Cost function definition and approximation
        1. 4.4.2.1. Cost function definition
        2. 4.4.2.2. Introductory example: quadrotor attitude control
      3. 4.4.3. Optimal sampling period selection
        1. 4.4.3.1. Problem formulation
        2. 4.4.3.2. Problem solving
        3. 4.4.3.3. Feedback-scheduling algorithm deployment
      4. 4.4.4. Application to the attitude control of a quadrotor
    5. 4.5. Control and real-time scheduling co-design via a LPV approach
      1. 4.5.1. A LPV feedback scheduler sensible to the plant’s closed-loop performances
      2. 4.5.2. Application to a robot-arm control
        1. 4.5.2.1. Performance evaluation of the control tasks in view of optimal resource distribution
          1. 4.5.2.1.1. Discussion
        2. 4.5.2.2. Simulation with TrueTime
        3. 4.5.2.3. Feasibility and possible extensions
    6. 4.6. Summary
    7. 4.7. Bibliography
  10. Chapter 5: Overload Management Through Selective Data Dropping
    1. 5.1. Introduction
      1. 5.1.1. System architecture
      2. 5.1.2. Problem statement
    2. 5.2. Scheduling under (m, k)-firm constraint
      1. 5.2.1. Dynamic scheduling policy under (m,k)-firm constraints
      2. 5.2.2. Static scheduling policy under (m,k)-firm constraints and schedulability issue
      3. 5.2.3. Static scheduling under (m, k)-constraints and mechanical words
      4. 5.2.4. Sufficient condition for schedulability assessment under (m,k)-pattern defined by a mechanical word
      5. 5.2.5. Systematic dropping policy in control applications
    3. 5.3. Stability analysis of a multidimensional system
      1. 5.3.1. Generic model
      2. 5.3.2. Example of multidimensional system
        1. 5.3.2.1. Sampling period definition
        2. 5.3.2.2. Controller parameters
      3. 5.3.3. Stability condition
    4. 5.4. Optimized control and scheduling co-design
      1. 5.4.1. Optimal control and individual cost function
      2. 5.4.2. Global optimization
      3. 5.4.3. Case study
        1. 5.4.3.1. Plants and controllers
        2. 5.4.3.2. Scheduling parameters
        3. 5.4.3.3. Optimal controller
        4. 5.4.3.4. Simulation scenario
        5. 5.4.3.5. Simulation results for hard real-time constraints
        6. 5.4.3.6. Simulation results for (m, k)-firm constraints
    5. 5.5. Plant-state-triggered control and scheduling adaptation and optimization
      1. 5.5.1. Closed-loop stability of switching systems
      2. 5.5.2. On-line plant state detection
      3. 5.5.3. Global optimization of control tasks taking into account the plant state
      4. 5.5.4. Case study
        1. 5.5.4.1. Simulation scenario
        2. 5.5.4.2. Observed performance
        3. 5.5.4.3. Summary
    6. 5.6. Conclusions
    7. 5.7. Bibliography
  11. Chapter 6: Fault Detection and Isolation, Fault Tolerant Control
    1. 6.1. Introduction
    2. 6.2. FDI and FTC
      1. 6.2.1. Introduction to diagnosis
      2. 6.2.2. Quantitative model-based residuals
        1. 6.2.2.1. Parity relations
        2. 6.2.2.2. Observers bank
      3. 6.2.3. Example
        1. 6.2.3.1. The system-residual generation
        2. 6.2.3.2. Observer-based residuals
      4. 6.2.4. Diagnostic summary
      5. 6.2.5. Introduction to FTC
    3. 6.3. Networked-induced effects
      1. 6.3.1. Example of network-induced drawbacks
      2. 6.3.2. Modeling data dropouts
      3. 6.3.3. Modeling network delays
    4. 6.4. Pragmatic solutions
      1. 6.4.1. Data synchronization
        1. 6.4.1.1. Clock synchronization
        2. 6.4.1.2. Data reconstruction
        3. 6.4.1.3. Example
      2. 6.4.2. Data loss and diagnostic blocking
    5. 6.5. Advanced techniques
      1. 6.5.1. Residual generation with transmission delay
      2. 6.5.2. Adaptive thresholding
        1. 6.5.2.1. Optimization-based approach for threshold selection
        2. 6.5.2.2. Network calculus-based thresholding
      3. 6.5.3. Fault isolation filter design in the presence of packet dropouts
      4. 6.5.4. Estimation and diagnosis with data loss
        1. 6.5.4.1. Problem formulation
        2. 6.5.4.2. Kalman filter with partial data loss
    6. 6.6. Conclusion and perspectives
    7. 6.7. Bibliography
  12. Chapter 7: Implementation: Control and Diagnosis for an Unmanned Aerial Vehicle
    1. 7.1. Introduction
    2. 7.2. The quadrotor model, control and diagnosis
      1. 7.2.1. The system
      2. 7.2.2. The physical system model
        1. 7.2.2.1. Introduction to quaternions
        2. 7.2.2.2. The quadrotor model
        3. 7.2.2.3. The inertial measurement unit (IMU) model
      3. 7.2.3. The attitude control
        1. 7.2.3.1. Nonlinear control
        2. 7.2.3.2. Linear quadratic control
      4. 7.2.4. The attitude observer
        1. 7.2.4.1. Nonlinear observer
        2. 7.2.4.2. Extended Kalman filter
        3. 7.2.4.3. Simulation results
      5. 7.2.5. The quadrotor diagnosis
        1. 7.2.5.1. Sensor diagnosis
        2. 7.2.5.2. Actuator diagnosis
    3. 7.3. Simulation of the network
      1. 7.3.1. Architecture of the networked control system
      2. 7.3.2. Network design
      3. 7.3.3. Tool implemented in the network simulation
    4. 7.4. Hardware in the loop architecture
      1. 7.4.1. The ORCCAD approach
      2. 7.4.2. Quadrotor simulation setup
    5. 7.5. Experiments and results
      1. 7.5.1. Basic attitude control
      2. 7.5.2. Packet loss
        1. 7.5.2.1. Pragmatic solution
        2. 7.5.2.2. (m, k)-firm solutions
        3. 7.5.2.3. Dynamic priorities
        4. 7.5.2.4. Extended Kalman filter
      3. 7.5.3. Hardware-in-the loop experiment
        1. 7.5.3.1. Basic scenario
        2. 7.5.3.2. Packet loss
        3. 7.5.3.3. Sensor failure
    6. 7.6. Summary
    7. 7.7. Bibliography
  13. Glossary and Acronyms
  14. List of Authors
  15. Index