Intelligent vs Classical Control by os8lO1GE

VIEWS: 0 PAGES: 37

									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
Distinguishing b/w Intelligent and
Classical Control
Classical Control

   The Mathematicians Approach
    –   Rigidly Modeled System
   Software does what it is told
    –   Intelligence comes from the Designer
Intelligent Control

   The Lazymans Approach
    –   System not Rigidly Modeled
   Software does what it wants to
    –   Intelligence comes from the Software
Shifting Intelligence

               Classical Control
 Software                             Designer

            Increasing Intelligence

 Designer                             Software
               Intelligent Control
Methods for Classical Control
Open-Loop Control System
Closed-Loop Control System
System Modeling


First-Order System:




Second-Order System:
Classical Control Examples

   PID Control
   Optimal Control
   Discrete-Event Control
   Hybrid Control
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
PID Control

                         t
                                          de(t )
  m(t )  K p e(t )  K I  e( )d  K D
                          0
                                           dt


                      KI
     GC ( s )  K p      KDs
                      s
Optimal Control (LQR)
Optimal Control (LQR)
Inverted Pendulum
Inverted Pendulum Model
Methods for Intelligent Control
Intelligent Control Examples

   Fuzzy Logic Control
   Neural Network Control
   Genetic Programming Control
   Support Vector Machines
   Numerical Learning
   COMDPs - POMDPs
No System Modeling

   Software learns system model
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
Sets

   Set A = {set of young people} = [0,20]
   Is somebody on his 20th birthday young and
    right on the next day not young?
Fuzzy Sets
Fuzzy Example – Inverted
Pendulum
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
   …
Actual Values
Neural Network Control

   Mimic Structure and Function of the Human
    Nervous System
    Biological Neurons

   Dendrites
    –   Connects neurons
    –   Modify signals
   Synapses
    –   Connects Dendrites
   Neuron
    –   Emits a pulse if input
        exceeds a threshold
    –   Stores info in weight
        patterns
Mathematical Representation of a
Neuron
Back-Propagation Neural Network
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
Genetic Programming Control

   Output of Genetic Programming is another
    computer program!
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
Crossover Operation
Mutation Operation
Applications
   In general,
    –   Use Classical Control (Intelligent Control can take long to train)
   If problem too complex
    –   Use Intelligent Control
Discussion

								
To top