Self-organizing operating conditioned response automata and application in realization of intelligent behavior thereof

A technique for operating conditioning reflexes and automata, applied in the field of bionic automata

Inactive Publication Date: 2009-12-09
BEIJING UNIV OF TECH
View PDF2 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

So far, autonomous operant condi...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Self-organizing operating conditioned response automata and application in realization of intelligent behavior thereof
  • Self-organizing operating conditioned response automata and application in realization of intelligent behavior thereof
  • Self-organizing operating conditioned response automata and application in realization of intelligent behavior thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0058] Embodiment 1: A minimal system—a small white mouse with learning ability, simulating Skinner's small white mouse experiment. Briefly describe Skinner's experiment with white mice: Put a white mouse in the Skinner box and set up a lever. The structure of the box excludes all external stimuli as much as possible. The mouse can move freely in the box, and when it presses the lever, a ball of food will fall into the plate under the box, and the mouse can eat the food. There is a device outside the box to record the movements of the animals. The mice will learn to keep pressing the lever and get food rewards through their actions. This experiment implements Skinner's mouse experiment by autonomously operating conditioned reflex automata. The mouse has two operational behaviors, one is to press the lever α 1 , the other is not pressing the lever α 2 , that is, the set of operations Ω={α 1 , α 2}, and the probabilities are denoted by p1 and p2 respectively. Its state se...

Embodiment 2

[0073] Embodiment 2: a robot pigeon with learning ability, simulating Skinner's pigeon experiment. In this experiment, the robot pigeon gets food when it pecks the red button (positive reinforcement stimulus), no stimulus when it pecks the yellow button, and gives electric shock when it pecks the blue button (negative reinforcement stimulus). At the beginning, the pigeon pecks the red, yellow and blue buttons is random. After a while, the pigeons pecked at the red button significantly more than the other two buttons. Define a 3-operation 3-state autonomous operation conditioned reflex automaton for the robot pigeon, whose operation set Ω={α 0 , α 1 , α 2}, whose elements are pecking red button α 0 , peck the yellow button α 1 and peck the blue button α 2 , the probabilities are denoted by p0, p1, p2 respectively. State set S={s 0 ,s 1 ,s 2}, that is, zero starvation state (non-starvation state) s 0 , half-starved state s 1 , starvation state s 2 , the state transi...

Embodiment 3

[0099] Embodiment 3: The balance control of the two-wheeled self-balancing robot is realized through the autonomous operation conditioned reflex automaton. Two-wheeled upright robots can freely move left and right on flat ground. When the deflection angle exceeds ±12°, the robot will lose its balance. The state set of the AOC automaton designed for this purpose is the robot deflection angle, including 6 states: θ=0°, 0°12°, use s respectively 0 , s 1 , s 2 , s 3 , s 4 , s 5 , s 6 to represent, therefore, the state set S={s 0 ,s 1 ,s 2 ,s 2 ,s 3 ,s 4 ,s 5 ,s 6 ,}. Its operation set Ω={α 0 , α 1 , α 2}, including not moving α 0 , move to the right by α 1 , move to the left by α 2 . The state transition rules are as follows:

[0100] δ(s 0 ×α 0 )=s 0 δ(s 0 ×α 1 )=s 3 δ(s 0 ×α 2 )=s 1

[0101] δ(s 1 ×α 0 )=s 2 δ(s 1 ×α 1 )=s 0 δ(s 1 ×α 2 )=s 2

[0102] δ(s 2 ×α 0 )=s 5 δ(s 2 ×α 1 )=s 1 δ(s 2 ×α 2 )=s 5

[0103] δ(s ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

A self-organizing operating conditioned response automata and an application in the realization of intelligent behaviors thereof belong to the biomimetics field. The self-organizing operating conditioned response automata relates to a discrete computer model for describing a self-organizing automaton, comprising operation set, state set, 'condition-operation' rule set, observable state transition and operating conditioned response law of learning; furthermore the behavior entropy based on AOC state orientation value is defined and the recursive operation program of AOC is set. The key character of the AOC is to stimulate the operating condition reflex mechanism of organism so that the AOC has the self-organizing function of bionics, comprising the self-learning and self-adaption functions which can be used for describing, stimulating and designing various self-organizing systems, in particular to being used for describing, stimulating and designing various intelligent behaviors for robot systems.

Description

technical field [0001] The invention relates to an automaton, which is a bionic automaton based on the principle of operating conditioned reflex. Background technique [0002] The automata model for learning systems existed in the 1960s and was called learning automata. In the past few years, the structure of learning automata was mainly changed to meet different application requirements. Generally, it has both input and There is output. The invention is a self-organizing system based on Skinner's operant conditioning theory, and has the functions of self-learning and self-adaptation. Skinner began experimental research on animal learning in the late 1920s and proposed two forms of learning: one is classical conditioned reflex learning, which is used to shape the response behavior of organisms; the other is operant conditioned reflex Learning, used to shape the operational behavior of an organism. Western scholars believe that these two reflexes are two different connecti...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06N3/00
Inventor 阮晓钢戴丽珍蔡建羡陈静郜园园
Owner BEIJING UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap