intelligent vs classical control bax smith en9940

37
Intelligent vs Classical Control Bax Smith EN9940

Upload: melina-atkins

Post on 03-Jan-2016

229 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: Intelligent vs Classical Control Bax Smith EN9940

Intelligent vs Classical Control

Bax Smith

EN9940

Page 2: Intelligent vs Classical Control Bax Smith EN9940

Today’s Topics

Distinguishing Between Intelligent and Classical Control

Methods of Classical Control Methods of Intelligent Control Applications for Both Types of Control Discussion

Page 3: Intelligent vs Classical Control Bax Smith EN9940

Distinguishing b/w Intelligent and Classical Control

Page 4: Intelligent vs Classical Control Bax Smith EN9940

Classical Control

The Mathematicians Approach– Rigidly Modeled System

Software does what it is told– Intelligence comes from the Designer

Page 5: Intelligent vs Classical Control Bax Smith EN9940

Intelligent Control

The Lazymans Approach– System not Rigidly Modeled

Software does what it wants to– Intelligence comes from the Software

Page 6: Intelligent vs Classical Control Bax Smith EN9940

Shifting Intelligence

Software

Designer

Increasing Intelligence

Designer

SoftwareClassical Control

Intelligent Control

Page 7: Intelligent vs Classical Control Bax Smith EN9940

Methods for Classical Control

Page 8: Intelligent vs Classical Control Bax Smith EN9940

Open-Loop Control System

Page 9: Intelligent vs Classical Control Bax Smith EN9940

Closed-Loop Control System

Page 10: Intelligent vs Classical Control Bax Smith EN9940

System Modeling

First-Order System:

Second-Order System:

Page 11: Intelligent vs Classical Control Bax Smith EN9940

Classical Control Examples

PID Control Optimal Control Discrete-Event Control Hybrid Control

Page 12: Intelligent vs Classical Control Bax Smith EN9940

PID Control

Proportional Control– Pure gain adjustment acting on error signal

Integral Control– Adjust accuracy of the system

Derivative Control– Adjust damping of the system

Page 13: Intelligent vs Classical Control Bax Smith EN9940

PID Control

dt

tdeKdeKteKtm D

t

Ip

)()()()(

0

sKs

KKsG D

IpC )(

Page 14: Intelligent vs Classical Control Bax Smith EN9940

Optimal Control (LQR)

Page 15: Intelligent vs Classical Control Bax Smith EN9940

Optimal Control (LQR)

Page 16: Intelligent vs Classical Control Bax Smith EN9940

Inverted Pendulum

Page 17: Intelligent vs Classical Control Bax Smith EN9940

Inverted Pendulum Model

Page 18: Intelligent vs Classical Control Bax Smith EN9940

Methods for Intelligent Control

Page 19: Intelligent vs Classical Control Bax Smith EN9940

Intelligent Control Examples

Fuzzy Logic Control Neural Network Control Genetic Programming Control Support Vector Machines Numerical Learning COMDPs - POMDPs

Page 20: Intelligent vs Classical Control Bax Smith EN9940

No System Modeling

Software learns system model

Page 21: Intelligent vs Classical Control Bax Smith EN9940

Fuzzy Logic Control

Multi-valued Logic– Rather warm/pretty cold vs hot/cold– Fairly dark/very light vs Black/White

Apply a more human-like way of thinking in the programming of computers

Page 22: Intelligent vs Classical Control Bax Smith EN9940

Sets

Set A = {set of young people} = [0,20] Is somebody on his 20th birthday young and

right on the next day not young?

Page 23: Intelligent vs Classical Control Bax Smith EN9940

Fuzzy Sets

Page 24: Intelligent vs Classical Control Bax Smith EN9940

Fuzzy Example – Inverted Pendulum

Page 25: Intelligent vs Classical Control Bax Smith EN9940

Fuzzy Rules

If angle is zero and angular velocity is zero then speed shall be zero

If angle is zero and angular velocity is pos. low then speed shall be pos. low

Page 26: Intelligent vs Classical Control Bax Smith EN9940

Actual Values

Page 27: Intelligent vs Classical Control Bax Smith EN9940

Neural Network Control

Mimic Structure and Function of the Human Nervous System

Page 28: Intelligent vs Classical Control Bax Smith EN9940

Biological Neurons

Dendrites– Connects neurons– Modify signals

Synapses– Connects Dendrites

Neuron– Emits a pulse if input

exceeds a threshold– Stores info in weight

patterns

Page 29: Intelligent vs Classical Control Bax Smith EN9940

Mathematical Representation of a Neuron

Page 30: Intelligent vs Classical Control Bax Smith EN9940

Back-Propagation Neural Network

Page 31: Intelligent vs Classical Control Bax Smith EN9940

Training a Neural Network

Analogous to teaching a child to read– Present some letters and assign values to them– Don’t learn first time, must repeat training– Knowledge is stored by the connection weights

Minimize the error of the output using LMS algorithm to modify connection weights

Page 32: Intelligent vs Classical Control Bax Smith EN9940

Genetic Programming Control

Output of Genetic Programming is another computer program!

Page 33: Intelligent vs Classical Control Bax Smith EN9940

Genetic Programming Steps

Generate a random group of functions and terminals (programs)

– Functions: +, -, *, /, etc…– Terminals: velocity, acceleration, etc…

Execute each program assigning fitness values Create a new population via:

– Mutation– Crossover– Most fit

Which ever program works best is the result

Page 34: Intelligent vs Classical Control Bax Smith EN9940

Crossover Operation

Page 35: Intelligent vs Classical Control Bax Smith EN9940

Mutation Operation

Page 36: Intelligent vs Classical Control Bax Smith EN9940

Applications

In general,– Use Classical Control (Intelligent Control can take long to

train) If problem too complex

– Use Intelligent Control

Page 37: Intelligent vs Classical Control Bax Smith EN9940

Discussion