Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While...
-
date post
20-Dec-2015 -
Category
Documents
-
view
221 -
download
0
Transcript of Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While...
![Page 1: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/1.jpg)
Genetic Programming
![Page 2: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/2.jpg)
Genetic AlgorithmsGenetic Algorithms• 1. Randomly initialize a population of
chromosomes.
• 2. While the terminating criteria have not been satisfied:– a. Evaluate the fitness of each chromosome:
• i. Construct the phenotype (e.g. simulated robot) corresponding to the encoded genotype (chromosome).
• ii. Evaluate the phenotype (e.g. measure the simulated robot’s walking abilities), in order to determine its fitness.
– b. Remove chromosomes with low fitness.– c. Generate new chromosomes, using certain
selection schemes and genetic operators.
![Page 3: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/3.jpg)
Genetic Algorithms
• Fitness function
• Selection scheme
• Genetic operators– Crossover– Mutation
![Page 4: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/4.jpg)
Crossover and mutation.
![Page 5: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/5.jpg)
Agenda• What is Genetic Programming?
• Background/History.
• Why Genetic Programming?
• How Genetic Principles are Applied.
• Examples of Genetic Programs.
• Future of Genetic Programming.
![Page 6: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/6.jpg)
Genetic ProgrammingGenetic Programming
• John Koza, 1992
• Evolve program instead of bitstring
• Lisp program structure is best suited– Genetic operators can do simple replacements of
sub-trees– All generated programs can be treated as legal (no
syntax errors)
Please review about Behavioral systems, Genetic Algorithms and Genetic Programming from the book.
Chapters 20, 21, 22, 23.
![Page 7: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/7.jpg)
Artificial Intelligence
Machine learningMachine learning
What is Genetic Programming(GP)?
GP
Artificial Intelligence
ROBOTICS
ESEP GA
evolutionary
![Page 8: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/8.jpg)
Genetic Algorithms• Most widely used• Robust • uses 2 separate spaces
– search space - coded solution (genotype)– solution space - actual solutions (phenotypes)
Genotypes must be mapped to phenotypes before the quality or fitness of each solution can be evaluated
![Page 9: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/9.jpg)
Evolutionary Strategies• Like GP no distinction between search and
solution space• Individuals are represented as real-valued
vectors.• Simple ES
– one parent and one child– Child solution generated by randomly mutating
the problem parameters of the parent.
• Susceptible to stagnation at local optima
![Page 10: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/10.jpg)
Evolutionary Strategies (cont’d)
• Slow to converge to optimal solution
• More advanced ES – have pools of parents and children
• Unlike GA and GP, ES– Separates parent individuals from child
individuals
– Selects its parent solutions deterministically
![Page 11: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/11.jpg)
Evolutionary Programming
• Resembles ES, developed independently• Early versions of EP applied to the evolution of transition
table of finite state machines• One population of solutions, reproduction is by mutation
only• Like ES operates on the decision variable of the problem
directly (ie Genotype = PhenotypeGenotype = Phenotype)• Tournament selection of parents
– better fitness more likely a parent– children generated until population doubled in size – everyone evaluated and the half of population with lowest fitness
deleted.
![Page 12: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/12.jpg)
General Architecture
of Evolutionary
Algorithms
![Page 13: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/13.jpg)
Genetic Programming• Specialized form of GA• Manipulates a very specific type of
solution using modified genetic operators
• Original application was to design computer program
• Now applied in alternative areas eg. Analog Circuits
• Does not make distinction between search and solution space.
• Solution represented in very specific hierarchical manner.
![Page 14: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/14.jpg)
Background/History• By John R. Koza, Stanford University.• 1992, Genetic Programming Treatise -
“Genetic Programming. On the Programming of Computers by Means of Natural Selection.” - Origin of GP.
• Combining the idea of machine learning and evolved tree structures.
![Page 15: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/15.jpg)
Why Genetic Programming?
• It saves time by freeing the human from having to design complex algorithms.
• Not only designing the algorithms but creating ones that give optimal solutions.
• Again, Artificial Intelligence.
![Page 16: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/16.jpg)
What Constitutes a Genetic Program?
• Starts with "What needs to be done"
• Agent figures out "How to do it"• Produces a computer program - “Breeding Programs”
• Fitness Test
• Code reuse
• Architecture Design - Hierarchies
• Produce results that are competitive with human produced results
![Page 17: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/17.jpg)
How are Genetic Principles Applied?
• “Breeding” computer programs.
• Crossovers.
• Mutations.
• Fitness testing.
![Page 18: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/18.jpg)
Computer Programs as Trees
• Infix/Postfix• (2 + a)*(4 - num) *
+ -
2 a 4 num
![Page 19: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/19.jpg)
“Breeding” Computer Programs
Hmm hmm heh.Hey butthead. Do
computer programs actually score?
![Page 20: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/20.jpg)
“Breeding” Computer Programs
• Start off with a large “pool” of random computer programs.
• Need a way of coming up with the best solution to the problem using the programs in the “pool”
• Based on the definition of the problem and criteria specified in the fitness test, mutations and crossovers are used to come up with new programs which will solve the problem.
![Page 21: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/21.jpg)
The Fitness TestThe Fitness Test• Identifying the way of evaluating how
good a given computer program is at solving the problem at hand.
• How good can a program cope with its environment.
• Can be measured in many ways, i.e. error, distance, time, etc…
![Page 22: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/22.jpg)
Fitness Test Criteria
• Time complexity a good criteria.– i.e. n2 vs. nlogn.
• Accuracy - Values of variables.
• Combinations of criteria may also be tested.
![Page 23: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/23.jpg)
Mutations in Nature• Ultimate source of genetic variation.• Radiation, chemicals change genetic
information.• Causes new genes to be created.• One chromosome.• Asexual.• Very rare.
Before:
acgtactggctaa
After:
acatactggctaa
Properties of mutationsProperties of mutations
![Page 24: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/24.jpg)
Genetic Programming
• 1. Randomly generate a combinatorial set of computer programs.
• 22. Perform the following steps iteratively until a termination criterion is satisfied– a. Execute each program and assign a fitness value to each individual.– b. Create a new population with the following steps:
• i. Reproduction: Copy the selected program unchanged to the new population.
• ii. Crossover: Create a new program by recombining two selected programs at a random crossover point.
• iii. Mutation: Create a new program by randomly changing a selected program.
• 33.. The best sets of individuals are deemed the optimal solution upon termination
![Page 25: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/25.jpg)
Mutations in ProgramsMutations in Programs• Single parental program is probabilistically selected
from the population based on fitness. • MutationMutation point randomly chosen.
– the subtree rooted at that point is deleted, and – a new subtree is grown there using the same random growth
process that was used to generate the initial population.
• Asexual operations (mutation) are typically performed sparingly: – with a low probability of,– probabilistically selected from the population based on fitness.
![Page 26: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/26.jpg)
Crossovers in NatureCrossovers in Nature• Two parental chromosomes exchange part
of their genetic information to create new hybrid combinations (recombinant).
• No loss of genes, but an exchange of genes between two previous chromosomes.
• No new genes creatednew genes created, preexisting old ones mixed together.
![Page 27: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/27.jpg)
Crossovers in Programs• Two parental programs are selected from the population
based on fitness. • A crossover point is randomly chosen in the first and second
parent. – The first parent is called receiving– The second parent is called contributing
• The subtree rooted at the crossover point of the first parent is deleted
• It is replaced by the subtree from the second parent.• CrossoverCrossover is the predominant operation in genetic
programming (and genetic algorithm) research • It is performed with a high probability (say, 85% to 90%).
![Page 28: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/28.jpg)
Applications of GP in robotics
• Wall-following robot – Koza– Behaviors of subsumption architecture of Brooks. Evolved a new
behavior.• Box-moving robot – Mahadevon• Evolving behavior primitives and arbitrators
– for subsumption architecture• Motion planning for hexapod – Fukuda, Hoshino, Levy
PSU.• Evolving communication agents Iba, Ueda.• Mobile robot motion control – Walker.
– for object tracking• Soccer• Car racing
Population sizes from 100 to 2000
![Page 29: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/29.jpg)
This is a very very small subset of This is a very very small subset of
Lisp Lisp
![Page 30: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/30.jpg)
• More functions
• Atom, list, cons, car, cdr, numberp, arithmetic, relations. Cond.
• Copying example
![Page 31: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/31.jpg)
A very important concept – Lisp does not distinguish data and programs
![Page 32: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/32.jpg)
Programs in Lisp are trees that are evaluated (calculated)
Tree-reduction semantics
![Page 33: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/33.jpg)
Lisp allows to define special languages in itself
![Page 34: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/34.jpg)
![Page 35: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/35.jpg)
• This subset of Lisp was defined for a mobile robot• You can define similar subsets for a humanoid robot,
robot hand, robot head or robot theatre
![Page 36: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/36.jpg)
As an exercise, you can think about behaviors of all Braitenberg Vehicles described by such programs
![Page 37: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/37.jpg)
• You can define much more sophisticated mutations that are based on constraints and in general on your knowledge and good guesses (heuristics)
![Page 38: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/38.jpg)
![Page 39: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/39.jpg)
EyeSim simulatorEyeSim simulator allows to simulate robots, allows to simulate robots, environments and learning strategies together with environments and learning strategies together with
no real robotno real robot
Evolution here takes feedback from simulated environment (simulated)
In our project with teens the feedback is from humans who evaluate the robot theatre performance (subjective)
In our previous project with hexapod the feedback was from real measurement in environment (objective)
![Page 40: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/40.jpg)
Evolution principles are the same for all evolutionary algorithms.
• Each individual (Lisp program) is executed on the EyeSim simulator for a limited number of program steps.
• The program performance is rated continually or when finished.
![Page 41: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/41.jpg)
Sample Problem: Ball Tracking
• Find ball in environment• Drive towards it• Stop when close to ball
Similar other tasks that we solved:
• Collect cans
• Shoot goal in soccer
• Robot sumo
• Robot fencing
A single robot is placed at a random position and orientation in a rectangular driving area closed by walls.
A colored ball is also placed at a random position.
Using its camera, the robot has to detect the ball, drive towards it, and stop close to it.
The robot camera is positioned at an angle so the robot can see the wall ahead from any position in the field.
However, the ball is not always visible.
![Page 42: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/42.jpg)
Idea for solving• In the loop, grab an image and analyze it as
follows:– Convert the image from RGB to HSV.– Use the histogram ball detection routine from section
8.6 of the book (this returns a ball position in the range [0..79] (left.. Right) or no ball and a ball size in pixels [0..60])
– If the ball height is 20 pixels or more, then stop and terminate (the robot is close enough to the ball)
– Otherwise• If no ball is detected or the ball position is less than 20, turn
slowly left.• If the ball position is between 20 and 40, drive slowly straight• If the ball position is greater than 40, turn slowly right.
![Page 43: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/43.jpg)
• ColSearch returns the x-position of the ball or -1 if not detected and the ball height in pixels.
• The statement VWDriveWait following either VWDriveTurn or a VWDriveStraight command suspends execution driving or rotation of the requested distance or angle has finished.
![Page 44: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/44.jpg)
The evolution can start from random data or from good hand-coded solutions from humans
Bear in mind that obj_size and obj_pose are in fact calls to the image processing subroutine.
![Page 45: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/45.jpg)
Evolution of tracking behavior
• Koza suggests the following steps for setting up a genetic programming system:
• 1. Establish an objective.• 2. Identify the terminals and functions used in the
inductive programs• 3. Establish the selection scheme and its evolutionary
operations• 4. Finalize the number of fitness cases• 5. Determine the fitness function and hence the range of
raw fitness values• 6. Establish the generation gap G, and the population M• 7. Finalize all control parameters.• 8. Execute the Genetic Paradigm.
![Page 46: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/46.jpg)
EyeSim simulator allows to simulate robots, environments and learning strategies together with no
real robot
Evolution here takes feedback from simulated environment (simulated)
In our project with teens the feedback is from humans who evaluate the robot theatre performance (subjective)
In our previous project with hexapod the feedback was from real measurement in environment (objective)
![Page 47: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/47.jpg)
Summary on GP• Field of study in Machine Learning.
• Created by John Koza in 1992.
• Save time while creating better programs.
• Based on the principles of genetics.
• Symbolic Regression/Circuit Design.
![Page 48: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/48.jpg)
• Example of using Example of using Genetic Programming Genetic Programming in Roboticsin Robotics
• Use of simulationUse of simulation
![Page 49: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/49.jpg)
Robot’s view and driving path in EyeSim simulator
Robot trajectory
![Page 50: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/50.jpg)
You want to maximize this fitness function
![Page 51: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/51.jpg)
![Page 52: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/52.jpg)
![Page 53: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/53.jpg)
Fitness grows in next populations (in general)
Highest fitness
![Page 54: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/54.jpg)
What are the evolved trajectories of the robot to meet the ball?
The best trajectory
![Page 55: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/55.jpg)
• Use of improved GP• Use of parallel processing• Use of FPGA and evolvable hardware• Use of quantum computers in future.
![Page 56: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/56.jpg)
Behavior-Based SystemsBehavior-Based Systems• Rodney Brooks, 1986
– Ronald Arkin, 1998
• Instead of single, monolithic program, use a number of behaviors ( parallel processes)– Improve changeability– Utililize emergence
• Each behavior has access to all sensors and actuators
• Problem: behavior selectionThere are now many variants of behavior-based robots.
Braitenberg is just one of them.
![Page 57: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/57.jpg)
Good old-fashioned hierarchical software architecture for a robot
![Page 58: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/58.jpg)
Modern behavioral software architecture for a robot
![Page 59: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/59.jpg)
Behavior Selection
• Clearly, we cannot have two different behaviors in control of actuators– (e.g. obstacle-avoidance wants to drive left, ball-
detection wants to drive right)
• Solution 1: Always select one behavior as active
• Solution 2: Add actuator output of all behaviors, multiplied with certain coefficients
![Page 60: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/60.jpg)
Behavior Selection
• Remaining problem:– How to determine which behavior to select or
which coefficients to use
• Solution: ??
• How about using GA to evolve a solution!
![Page 61: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/61.jpg)
Behavior SelectionBehavior SelectionParameters
Single chromosome
camera Ball detection
Processed sensor input
Evolution of controller
parameters
Schemas using direct and processed inputs
Actuator output
Schema weighting
Move to ball
Avoid obstacles
schemas
Direct inputs
Trial result
![Page 62: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/62.jpg)
• Many tools of this type are created• In this class the projects are about animatronix editor and state
based editor• Another possibility is to use Artificial Neural Network
![Page 63: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/63.jpg)
• This slide shows hierarchical editor for subsumption-type architecture
![Page 64: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/64.jpg)
Sample Problem: Ball Tracking
• Problem:– Find ball in environment– Drive towards it– [Stop when close to ball]
• Behavior selection:– Use NN– Evolve NN weights using GA
![Page 65: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/65.jpg)
Behavior Selection with NN Behavior Selection with NN and GAand GA
• Fitness function:
fitness = initDist - b_distance();
if (fitness < 0.0) fitness = 0.0;
![Page 66: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/66.jpg)
![Page 67: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/67.jpg)
![Page 68: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/68.jpg)
Both approaches can provide increase in Fitness function. The problem is how good?
![Page 69: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/69.jpg)
Trajectories found by behavioral approaches
![Page 70: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/70.jpg)
![Page 71: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/71.jpg)
AI in robotics (first summary)AI in robotics (first summary)
• Neural Networks– Learning requires complete test set with correct
results
• Genetic Algorithms– Evolving solutions based on fitness function can be
used in real robot or in simulation (often easier/faster)
• Genetic Programming– Requires algorithm coding/interpretation (e.g. Lisp)
• Behavior-Based Systems– Combining parallel processing approach with
intelligent selection scheme
![Page 72: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/72.jpg)
Other areas - ExamplesOther areas - Examples of Genetic Programsof Genetic Programs
• 1. Symbolic Regression - – the process of discovering:
• the functional form of a target function • and all of its necessary coefficients, • or at least an approximation to these.
• 2. Analog circuit design – Embryo circuit is an initial circuit which is modified
to create a new circuit according to functionality criteria.
![Page 73: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/73.jpg)
Genetic Programming in the Genetic Programming in the FutureFuture
• Speculative.• Only been around for 10 years.• Is very successful.• Discovery of new algorithms in
existing projects.• Can be combined with constructive
induction, LISP-based systems, evolutionary hardware, neural networks, fuzzy logic, quantum logic and many other ideas and methods.
Mr. Roboto
![Page 74: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/74.jpg)
End of Show
Shut up Buttmunch.That sucked.
Hey Butthead.That kicked ass.
Oh yeah. Hm hm yeah yeah hm. It sucked.
![Page 75: Genetic Programming. Genetic Algorithms 1. Randomly initialize a population of chromosomes. 2. While the terminating criteria have not been satisfied:](https://reader035.fdocuments.net/reader035/viewer/2022081603/56649d4a5503460f94a27e0b/html5/thumbnails/75.jpg)
Sources• Braunl
• Dan Kiely
• Ran Shoham
• Brent Heigold
• CPSC 533, Artificial Intelligence