Download presentation
Presentation is loading. Please wait.
Published byMelina Atkins Modified over 9 years ago
1
Intelligent vs Classical Control Bax Smith EN9940
2
Today’s Topics Distinguishing Between Intelligent and Classical Control Methods of Classical Control Methods of Intelligent Control Applications for Both Types of Control Discussion
3
Distinguishing b/w Intelligent and Classical Control
4
Classical Control The Mathematicians Approach – Rigidly Modeled System Software does what it is told – Intelligence comes from the Designer
5
Intelligent Control The Lazymans Approach – System not Rigidly Modeled Software does what it wants to – Intelligence comes from the Software
6
Shifting Intelligence Software Designer Increasing Intelligence Designer Software Classical Control Intelligent Control
7
Methods for Classical Control
8
Open-Loop Control System
9
Closed-Loop Control System
10
System Modeling First-Order System: Second-Order System:
11
Classical Control Examples PID Control Optimal Control Discrete-Event Control Hybrid Control
12
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
13
PID Control
14
Optimal Control (LQR)
16
Inverted Pendulum
17
Inverted Pendulum Model
18
Methods for Intelligent Control
19
Intelligent Control Examples Fuzzy Logic Control Neural Network Control Genetic Programming Control Support Vector Machines Numerical Learning COMDPs - POMDPs
20
No System Modeling Software learns system model
21
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
22
Sets Set A = {set of young people} = [0,20] Is somebody on his 20th birthday young and right on the next day not young?
23
Fuzzy Sets
24
Fuzzy Example – Inverted Pendulum
25
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 …
26
Actual Values
27
Neural Network Control Mimic Structure and Function of the Human Nervous System
28
Biological Neurons Dendrites – Connects neurons – Modify signals Synapses – Connects Dendrites Neuron – Emits a pulse if input exceeds a threshold – Stores info in weight patterns
29
Mathematical Representation of a Neuron
30
Back-Propagation Neural Network
31
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
32
Genetic Programming Control Output of Genetic Programming is another computer program!
33
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
34
Crossover Operation
35
Mutation Operation
36
Applications In general, – Use Classical Control (Intelligent Control can take long to train) If problem too complex – Use Intelligent Control
37
Discussion
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.