Architecture and System Design

Prev: Fundamental Limits

Chapter 15 - Architecture and System Design

Next: Bibliography

In this chapter we place the relatively simple feedback loops that have been the focus of the previous chapters in the context of overall system design. We outline a typical design process and discuss the role of architecture and how it can be approached from top-down and bottom-up perspectives. Interaction and adaptation are then reviewed and the chapter ends with a brief overview of control design in some major industrial fields.

Contents

1. Introduction
2. System and Control Design
3. Top-Down Architectures
   • Layered Architectures for Control
   • Online Optimization
   • Discrete-decision making and supervisory control
   • Linking Continuous and Discrete Controllers
   • Model Checking and Program Synthesis
4. Bottom-Up Architectures
   • Cascade Control -- Several Sensors
   • Mid-Range Control -- Many Actuators
   • Selector Control -- Equipment Protection
   • Internal Model Control -- Disturbance Observer
   • The Smith Predictor -- Phase Advance
   • Complementary Filtering -- Sensor Fusion
   • Extremum Seeking or Self-Optimization
5. Interaction
   • The Relative Gain Array
   • Parallel Systems
6. Adaptation and Learning
   • Adaptive Control
   • Learning
   • Neural Networks
   • Deep Learning
7. Control Design in Common Application Fields
   • Aerospace -- High Performance Systems and Highly Skilled Users
   • Automotive -- Complex Systems Used by Ordinary People
   • Process Industry -- Complex Systems with Many Different Users
   • Telecommunication -- Billions of Systems
8. Further Reading