ECE 595C - Biologically Inspired Engineering

Course Details

Lecture Hours: 3 Credits: 3

Counts as:

Experimental Course Offered:

Fall 2004

Catalog Description:

The objective of this course is to prepare the student to use modern design strategies rooted in biological principles. Examples include: the use of biological genetics as an inspiration for genetic algorithms for the automated design of components and systems; the use of linguistic based controls strategies based on human speech and logic to formulate control systems (a strategy formally known as fuzzy logic); the construction of artificial neural networks modeled after the construction of biological nervous systems for information processing and control; use of protein classification and identification algorithms to solve engineering design problems; and the study of immunology and how it relates to computer viruses and synthetic immune systems.

Course Objectives:

The development of complex, highly reliable integrated systems with greatly reduced design cycle times poses a formidable challenge to today's engineers. Traditional engineering design practices are not adequate to meet this new challenge; engineers are increasingly turning to biology for answers. Engineering has always borrowed from nature to provide conceptual examples (for instance, aerodynamics from birds). However, the extraordinary demands placed on today's engineering designs have resulted in the use of biological concepts in a concrete and mathematically defined way. These concepts include artificial networks that mimic the functioning of neurons in the brain, fuzzy logic that more closely reflects human reasoning, and genetic algorithms that imitate the mechanics of biological genetics. Future challenges facing engineering, science, and technology will require multidisciplinary education including proficiency in biologically inspired engineering. The objective of this course is to make students aware of biologically inspired engineering techniques and to equip them with multidisciplinary breadth - thus making them prepared to collaborate with their colleagues from other disciplines.

Required Text(s):

  1. selected journal papers and book chapters

Recommended Text(s):

None.

Learning Outcomes:

A student who successfully fulfills the course requirements will have demonstrated an ability to:
  1. implement and apply genetic and evolutionary algorithms to parameter identification and design problems. [None]
  2. analyze a simple bioinformatics system. [None]
  3. apply soft computing tools to the design of control systems. [None]
  4. construct and analyze a simple model of immune system or tumor growth. [None]

Lecture Outline:

Lectures Topics
3 Overview of Menelian principles of genetics: basic terns (gene, alleles, genome, genotype, phenotype); chromosomes and reproduction; mutation; genetic processes of evolution; genetic engineering.
8 Genetics as an inspiration for an optimization approach: evolutionary and genetic algorithms; implementing evolutionary and genetic algorithms; Schema theorem. Case studies involving evolutionary and genetic algorithms: genetic algorithm frequency domain parameter identification of a permanent magnet synchronous machine; genetic algorithm time domain parameter identification of a transformer-rectifier; genetic algorithms for the design of an inductor.
6 Introduction to bioinformatics---computational biology: genes, corresponding proteins, and digital symbol sequences; protein function; information content of biological sequences; hidden Markov models.
9 Soft computing: fuzzy systems (fuzzy linear programming, fuzzy classification); architectures and learning in neural networks (backpropogation, brain-state-in-a-box neural models, neural fuzzy systems, protein classification using artificial neural networks); application of genetic algorithms to soft computing.
9 Mathematical modeling and immunology: tumor growth models; the logistic approach; a predator-prey approach; modeling of tumor cell/immune system interaction.
2 Computer viruses and synthetic immune systems
2 Power system as a metaphor of an immune system
2 Case studies and student project presentation
3 Exams

Engineering Design Content:

  • Establishment of Objectives and Criteria

Engineering Design Consideration(s):

  • performance

Assessment Method:

The student will have more than one opportunity to satisfy the course outcomes. The primary means will be through the regular hourly and final exams. Exam questions will be written around each of these course outcomes. The student will satisfy each outcome when his/her score equals or exceeds a specified value representing minimal competency. If the student fails to meet this level of minimal competency on a specific course outcome, he/she will have a second chance by taking a retest. The retest will not affect the original exam score, but it will allow for the second opportunity to demonstrate the student's competency on the course material, thus satisfying the course outcome. Finally, the student may be able to satisfy outcomes on later exams that cover overlapping material.