Control Systems Overview

REV II

DR. TAREK A. TUTUNJI

MECHACTRONIC SYSTEM DESIGN
PHILADELPHIA UNIVERSITY
2014
 Control Systems

 The control system is at the heart of mechatronic

systems and its selection is arguably the most critical
decision in the design process.

 The controller selection involves two inter-

dependent parts:
 The control method (i.e. software)
 The physical controller (i.e. hardware)
 Basic Control Concepts

OPEN VS. CLOSED LOOP CONTROL

PROCESS VS. MOTION CONTROL
TRANSIENT AND STEADY STATE SPECIFICATIONS
Open-Loop Control

[Ref] Kilian
Closed-Loop Control

[Ref] Kilian
 Control Systems Classification

 Control systems are classified by application.

 Process control usually refers to an industrial process being
electronically controlled for the purpose of maintaining a
uniform correct output.

 Motion control refers to a system wherein things move. A

servomechanism is a feedback control system that provides
remote control motion of some object, such as a robot arm or a
radar antenna.
Process Control

[Ref] Kilian
Process Control Example

[Ref] Kilian
Motion Control
 Motion Control Examples

CNC Machine Robot Manipulator

[Ref] Kilian
General Control System
First Order Systems
First Order Systems
Second Order Systems
Performance Criteria
 Transient Response

 Transient response is the shape of a signal as it

moves between two steady-state points.
 It is quantified in terms of two parameters:
 The damping ratio, z, pronounced zeta
 The natural undamped frequency, wn.
 Pole Locations

The poles location is the major factor for a systems’ transient response
Step Response Comparisons
 Steady-State Error

 Accuracy (or steady-state tracking error) is the error

between input and output signals in the steady state
for a system.

 Three input signals can be used

 Step

 Ramp

 Parabola
Steady-State Error
Steady-State Error
 Stability

 A stable system is one which produces a bounded, or

finite, response when subjected to a bounded input

 Stability conditions
 A system is stable if the real part of all poles are < 0.

 A system is marginally stable if real part of all poles are <= 0.

 A system is unstable if the real part of any pole is positive.

Control Methods

TAREK A. TUTUNJI
 Control Techniques / Strategies

 Classical Control
 Advanced Control
 Intelligent Control

Dr. Tarek A. Tutunji

 Classical Control

 Classical control design are used for SISO

systems.
 Most popular concepts are:
 Bode plots
 Nyquist Stability

 Root locus.

 PID is widely used in feedback systems.

Dr. Tarek A. Tutunji

 Classical Control: On-Off Control

 This is the simplest method of control. The control

action has three possible outputs: on; off; no change.
This method is usually used for slow-acting
operations (such as a refrigeration unit).

 The advantage is its ease of design and low cost.

However, it cannot vary the controlled variable with
precision.

On-Off Control Example

 Classical Control: PID

 Proportional-Integral-Derivative (PID) is the most

commonly used controller for SISO systems

de( t )
u( t )  K p e( t )  K I  e( t )dt  K D
dt
Dr. Tarek A. Tutunji
PD Design Example
Analog PID Implementation

[Ref] Kilian
 Digital PID Control

Analog

Digital

Tarek A. Tutunji
Digital PID Realization

Required Operations:
•Multiplication
•Addition
•Delay
Discrete PID Implementation
Digital Control Block Diagram
Classical Control: Root Locus
Discrete Systems: Pole Locations
 Advanced Control

 Adaptive control methods modify the control law used by a

controller to cope with time-varying parameters.
 For example, as an aircraft flies, its mass will slowly decrease as
a result of fuel consumption; we need a control law that adapts
itself to such changing conditions.

 Robust control methods deal with uncertainty.

 They guarantee that if the changes are within given bounds the
control law need not be changed.

 Optimal control uses math optimization methods to solve a set

of differential equations. Two such methods are:
 Model Predictive Control (MPC)
 Linear-Quadratic-Gaussian control (LQG).
 Intelligent Control

 Intelligent controllers are used for high-level control

 Intelligent controllers are also used when the system must make
decisions (from several alternatives) based on input data from
sensors.

 Intelligent Control is usually used when the mathematical model

for the plant is unavailable or highly complex.

 The most two commonly used intelligent controllers are

 Artificial Neural Networks
 Fuzzy Logic

Dr. Tarek A. Tutunji

 Intelligent Control: Fuzzy

 Fuzzy set theory provides mathematical tools for carrying

out approximate reasoning processes when available
information is uncertain, incomplete, imprecise, or vague.

 Fuzzy logic controllers manage complex control problems

through heuristics (IF … THEN) and mathematical models
provided by fuzzy logic, rather than via mathematical
models provided by differential equations.

 This is particularly useful for controlling systems whose

mathematical models are nonlinear or for which standard
mathematical models are simply not available
Fuzzy Control
Fuzzy Control
 Intelligent Control: ANN

 Artificial Neural networks (ANN) are nonlinear

mathematical models that are used to mimic the biological
neurons in the brain.

 ANN are used as black box models to map unknown

functions

 ANN can be used for: Identification and Control

Dr. Tarek A. Tutunji

 ANN: Single Neuron

w0
x1
w1
x2 f(net)
wM y
xM

 M 
y  f   xm wm 
 m 1 
 Neural Nets

Plant
Output TDL Weights
Log Log Net
+ Weights + Function Output
Plant Function
Input TDL Weights

First Layer Second Layer

ANN: Identification and Control

Control Identification
ANN: Identification and Control
Intelligent Controllers Applications
Intelligent Controller Application

•Low Level PID Control for velocity control

•High Level Intelligent Control:
•Fuzzy for Decision making
•Neural nets for Image Analysis
Hardware Controllers

DR. TAREK TUTUNJI

 Analog vs. Digital Control Systems

Analog Digital
Time variable Continuous Discrete
Time equations Differential equations Difference equations

Frequency transforms Laplace Z-Transform

Stability Poles on LHS Poles inside unit circle

Controller Hardware: Op-Amps Hardware: Microcontroller

Software: None Software: Program

Dr. Tarek A. Tutunji

 Criteria for Choosing Controller

 Price
 Size and Weight
 Number of Digital Inputs and Outputs
 Number of Analog Inputs and Outputs
 Speed
 Required Interrupt
 Required hardware
 Communication Interface
 Reliability
 Memory
 Programming Capability
 Software Support

Dr. Tarek A. Tutunji

 Hardware Controllers

 Microcontroller
 PLCs
 DSPs
 FPGA
 PC with DAQ

Dr. Tarek A. Tutunji

 Microcontrollers

 Microcontroller is a special type of small computer

that can perform a specific job
 Microcontrollers

 The microcontroller is a computer-on-chip. It is an

integrated circuit that contains microprocessor,
memory, I/O ports and sometimes A/D converters.
It can be programmed using several languages
(such as Assembly or C/C++). It can be used in
manufacturing lines, but requires additional
hardware. Microcontrollers are mainly used in
engineering products such as washing machines and
air-conditioners.

Dr. Tarek A. Tutunji

Microcontrollers Companies
Microcontroller Market Share
 Arduino

 Arduino is an open-source electronics prototyping platform based

on flexible, easy-to-use hardware and software.
 The hardware consists of a simple open source hardware board
designed around an 8-bit Atmel AVR microcontroller, though a new
model has been designed around a 32-bit Atmel ARM
 ARM

 The ARM architecture

describes a family
of RISC-based computer
processors designed and
licensed by British
company ARM Holdings.
 As an IP core business,
ARM Holdings itself does
not manufacture its own
electronic chips, but
licenses its designs to
other
semiconductor manufactu
rers
 PLCs

 A Programmable Logic Controller (or PLC) is a specialized

digital controller that can control machines and processes. it
monitors inputs, makes decisions, and controls outputs in
order to automate machines and processes
 Programmable Logic Controller

 PLC’s are a user-friendly, microprocessor-based,

specialized computer that is used for process control.
It contains input/output (I/O) modules for
appropriate sensors/actuator interfaces. It is
mainly used in automated manufacturing lines. The
PLC is usually used for simple logic operations. It is
considered reliable and easy to program (using
ladder diagrams, instructions, or function blocks).

Dr. Tarek A. Tutunji

PLC Manufacturers
 PLC vs. Microcontroller

 Usually PLCs are used in an industrial environment,

where as the microcontrollers are smaller and well
suited for embedded situations.

 PLCs are programmed with ready made blocks or

programming elements, whereas in Microcontrollers
a programming language must be used to write a
programming code
 PLC Advantages

 They are highly reliable, fast and flexible.

 They can handle severe conditions such as dust,
humidity etc.
 They can communicate with other controllers.
 They are easy to program and troubleshoot.
 They include display units.
 Digital Signal Processors

 Digital Signal Processing (DSP) is the arithmetic processing of discrete-

time signals.
 A/D is needed for analog signals

 Digital signal processors (DSP) are specialized

microprocessors with advanced architectures (such as
multiple buses, parallel processing, hardware multipliers and
fast sampling rate) that are designed to reduce the number of
instructions and operations necessary for efficient processing.

 DSP chips enable developers to implement complex

algorithms and perform computationally efficient and fast
algorithms.
 DSP are preferred over microcontrollers when the need for complex and
iterative control algorithms is required.

Dr. Tarek A. Tutunji

 DSP Operations: Convolution

x(n) h(n) y(n)


y(n)  x(n) * h(n)   x(k)h(n  k)
k -

 Convolution requires:
 Reflection

 Shift

 Multiplication

 Addition
 DSP Architecture Features

 Parallel Processing (Modified Harvard)

 Deep Instructions Pipeline
 Very Fast A/D
 Hardware Multiplier
 Barrel Shifter
 RISC

Dr. Tarek A. Tutunji

 Modified Harvard Architecture

 A Harvard architecture employs separate program and

data buses to access separate data and program memories.

 A modified Harvard architecture.

 DSP use multiple data buses (and multiple associated address buses)
so that the processing of two signals can be done in parallel.
 The address buses are also separate. This multiple bus arrangement
increases speed since instructions and data can move in parallel, and
execute simultaneously rather than sequentially.
 Modified Harvard Architecture

DAGEN DAGEN
A Memory Memory B
A B

ALU
Multiplier
Shifter

Accumulators

DAGEN Memory
C Shifter
C
 Instruction Pipelining

 Up to six levels of pipelining are implemented.

 DSP can execute instructions in parallel
 Overall execution times are accelerated so that high
 Hardware Multiplier

 A 16- by 16-bit hardware multiplier multiplies

and stores results in a 40-bit accumulator (8
guard bits) in a single instruction cycle.

 Thus, multiply and accumulate operations can be

performed in a single clock cycle in a DSP;
conventional processors may require tens of
cycles for this operation.
 Shifters and RISC

 Hardware shifters allow scaling, prevent overflows, and maintain

required precision.

 An on-chip hardware stack reduces interrupt response time and

minimizes stack pointer manipulations.

 DSP use reduced instruction sets tailored to digital signal processing

operations. For example, the MACD command implements four
operations in one instruction:
 multiplies two values
 moves data
 adds the product to a previous result
 transfers the result to an adjacent register.
Digital Signal Controllers Manufacturers

 Texas Instruments.
 TMS320C2000™
DSP Platform

 Microchip.
 dsPIC30F3010

 Motorola
Custom made DSP Engines
 Field Programmable Gate Arrays

 The field-programmable gate array (FPGA) is a semiconductor

device that can be programmed after manufacturing.

 Instead of being restricted to any predetermined hardware function,

an FPGA allows you to program product features and functions,
adapt to new standards, and reconfigure hardware for specific
applications even after the product has been installed in the field—
hence the name "field-programmable".

 FPGAs can be used to implement any logical

function that an application-specific integrated
circuit (ASIC) could perform. One advantage is its
ability to update the functionality after shipping.
 FPGAs vs. Microcontrollers

 FPGAs can perform concurrent operations while the

microcontrollers’ operations are sequential.
 This makes FPGAs better suited for real-time applications such as executing
DSP algorithms.

 FPGA are flexible, you can add subtract the functionality as

required. This can not be done in microcontroller.

 FPGAs are hard-wired and the random attack of alpha rays can not
destroy/corrupt the memory areas hence collapse the device
functionality.

 FPGA based development is longer while microcontrollers change

too often and there is lots re-work required to do in order to keep
pace with changing technology. This is necessary to save the design
from being obsolete.

Dr. Tarek A. Tutunji

 FPGAs vs. Microcontrollers

 The development time for microcontroller is shorter and that of

FPGA

 The microcontroller peripherals are readily available and tested by

the vendor. As for the FPGA, open source soft-peripherals are
available, but still need to be embedded and tested.

 Microcontroller are power efficient.

 Microcontroller are low-cost, much lower than FPGAs. This is

specially true for small applications and large quantities.

 Microcontrollers are available in easy to solder SOIC and QFP

package while FPGAs offer limited sources.

Dr. Tarek A. Tutunji

 Personal Computers

 Personal computers are used when extensive signal

processing and in-depth analysis is required.

 This will require Data Acquisition Cards (DAQs) to interface

the I/O power and signals between the PC and the
environment.

 Advantages include superior graphical and software

flexibility.

 However, the cost is high and, therefore, they are not suitable
for a large number of products

 Another disadvantage is the speed

Dr. Tarek A. Tutunji

PCs and DAQs
 Summary

 The selection of the controller is arguably the most

important issue of the mecahtronics system

 This choice can be divided into two parts:

1. Software/Firmware algorithm
 On-Off, PID, Adaptive, Robust, Optimal, and Intelligent

2. Hardware system
 Microcontroller, PLC, DSP, FPGA, and PC-DAQ

Dr. Tarek A. Tutunji