ISRC - PowerPoint PPT Presentation

1 / 14
About This Presentation
Title:

ISRC

Description:

Statistical & Neural Network Pattern Recognition, Detection Theory ... Probability, Time Series Analysis, Dynamical Systems Theory ... – PowerPoint PPT presentation

Number of Views:46
Avg rating:3.0/5.0
Slides: 15
Provided by: eceU7
Category:
Tags: isrc | theory

less

Transcript and Presenter's Notes

Title: ISRC


1
ISRC Controls
  • Undergraduate Depth Sequence Workshop
  • March 1, 2006

2
Intelligent Systems
  • Sense the World
  • DSP, Image Processing, Statistical Data
    Collection
  • Extract Information About the World
  • Feature Extraction, Neural Network and Fuzzy
    Learning, Data Compression
  • Recognize the State of the World
  • Statistical Neural Network Pattern Recognition,
    Detection Theory
  • Predict the Behavior of the World
  • Probability, Time Series Analysis, Dynamical
    Systems Theory
  • Reason About the World and Plan Actions
  • Artificial Intelligence, Decision Theory
  • Act on, and Communicate with, the World
  • Control Theory, Robotics, Information
    Communications Theory

3
Example Automated Medical Diagnosis
The contrast enhanced image is sent to an
automatic feature detection stage
Magnetic resonance brain scan. After some
preprocessing
4
Example Automated Medical Diagnosis - continued
  • Informative Features are Extracted and Patterns
    are Recognized
  • A Bayesian Diagnostic Reasoning Engine is
    Invoked

5
Machine Intelligence and Controls Applications
  • Weather and Natural Disaster Prediction
  • Speech and Image Recognition
  • Biomedical imaging and Diagnosis
  • Manufacturing and Robotics
  • Automotive Systems
  • Stock Market and Economic Prediction
  • Aerospace Systems
  • Electromagnetic Device Control Optimization
  • Computer games and Virtual reality
  • Genomics and Bioinformatics
  • Agricultural and Industrial inspection
  • Industrial Optimization

Just to name a few
6
ISRC Depth Requirement
  • Introduction to Intelligent Systems Robotics and
    Machine Intelligence (ECE172A)
  • Neural Networks Fuzzy Logic (ECE173)
  • Introduction to Linear and Nonlinear Optimization
    with Applications (ECE174)
  • Elements of Machine Learning (ECE175)
  • Introduction to DSP (ECE161A)or ECE161A, 187,
    253A, 285, Cogsci108F (Medical Imaging,
    Advanced Neural Networks, )

7
Controls Depth Requirement
  • Classical State-Space Linear Control Systems
    (ECE171AB)
  • Introduction to Linear and Nonlinear Optimization
    with Applications (ECE174)
  • Neural Networks Fuzzy Logic (ECE173)
  • Computer Interfacing (ECE118)

8
ECE 172A Introduction to Intelligent Systems
  • Main topics covered in the course include
  • Introduction to Intelligent Systems and
    Sensor-based Robots
  • Model-Based approach in perception
  • Image segmentation
  • Edge Detection
  • Region growing
  • Texture analysis
  • Object recognition and image understanding
  • Extraction of 3-dimensional cues passive and
    active approaches
  • Project (about 5 weeks long)
  • Vehicle Detection Re-identification
  • Person Detection and Tracking
  • Robust Image Classification

Image Classified as a Day Image
Input Image
9
175 Elements of machine intelligence
  • introduction to machine learning.
  • decision functions.
  • statistical pattern classifiers.
  • generative vs. discriminant methods
  • feature selection.
  • regression.
  • clustering.
  • applications of machinelearning.

10
ECE 174 Intro to Linear Nonlinear Optimization
Benefits
  • Linear Nonlinear Inverse Problems
  • Regularization of Ill-Posed Problems
  • Projection Theorem and the Geometry of Linear
    Operators
  • Least-Squares and the Pseudoinverse
  • Application Computer Projects
  • Many practical, real-world algorithms are based
    on least-squares solutions
  • Project Reports Make Impressive Additions to Job
    Interview Portfolios
  • Students from 174 have been hired by companies
    which do
  • Signal processing
  • Speech compression
  • GPS navigation

11
ECE 171AB Classical and State-Space Control
Systems
Benefits
  • Theory of SISO LTI Systems in the Time
    Frequency Domains
  • Transient and Steady State Behavior
  • Continuous vs Discrete-Time Control Systems
  • Frequency Domain Stabilization Techniques
  • Design of Controllers and Compensators
  • State-Space Modeling of Complex Multivariable
    Systems
  • Used to provide high-performance control of
    electromechanical systems
  • Many important real-world applications and jobs
  • Disk Drive Industry
  • Printer Industry
  • Automotive Industry
  • Aerospace Industry
  • Manufacturing

12
ECE 173 Neural Networks (NNs) and Fuzzy Logic
(FL)
Benefits
  • Theory of Neural Networks and training of NNs
  • Feedforward vs. recurrent NNs
  • NNs for Pattern Classification and Functional
    Approximation
  • NNs for Adaptive Control
  • Theory of Fuzzy Logic and Reasoning
  • Theory of Fuzzy Control
  • Computer Projects
  • Applications to Decision, Modeling, and Control
    of Highly Uncertain and Nonlinear Systems
  • Many important real-world applications and jobs
  • Disk Drive Industry
  • Printer Industry
  • Automotive Industry
  • Aerospace Industry
  • Manufacturingand more

13
ECE 161A Introduction to DSP
  • Discrete-time signals and systems
  • Frequency-domain characterization
  • Analysis of discrete-time systems
  • Z-transform
  • Linear-phase, Minimum-phase and Allpass systems

14
ECE 118 Computer Interfacing
  • Interfacing computers and embedded controllers to
    devices the real world
  • Busses, interrupts, DMA, memory mapping,
    concurrency
  • Serial/parallel communications, real-time I/O,
    closed loop control.
  • Design construct an interfacing project
Write a Comment
User Comments (0)
About PowerShow.com