Intrinsic motivation based self-cognition system for motion balance robot and control method

A technology of cognitive system and control method, applied in the field of autonomous cognitive system and control of motion balance robot based on intrinsic motivation, can solve the problem that the research of motion balance bionic autonomous cognitive model is rare and so on.

Inactive Publication Date: 2015-10-21
TIANJIN UNIV OF TECH & EDUCATION TEACHER DEV CENT OF CHINA VOCATIONAL TRAINING & GUIDANCE
View PDF0 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there are still few studies on the bionic autonomous ...

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
  • Intrinsic motivation based self-cognition system for motion balance robot and control method
  • Intrinsic motivation based self-cognition system for motion balance robot and control method
  • Intrinsic motivation based self-cognition system for motion balance robot and control method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] An autonomous cognition system for a motion balance robot based on intrinsic motivation, the system structure is as follows: figure 1 As shown, the autonomous cognitive system of the kinematic balance robot includes: a cognitive model based on intrinsic motivation, and the cognitive model based on intrinsic motivation is composed of behavior generation, behavior evaluation and orientation mechanisms.

[0038] Among them, the behavior generation is mainly the formation of the "perception-motor" loop in the cognitive system, including: topology design, perceptual behavior mapping, evolutionary learning algorithm, two parts of perceptual behavior mapping and topology design can be realized by neural network , the connection weight of the neural network is the synaptic connection weight.

[0039] Behavior evaluation is mainly the formation of the "movement-result" loop in the cognitive system, including: topology design, behavior exploration, and behavior evaluation based o...

Embodiment 2

[0044] Behavior is output by the cortico-cerebellar system, and is projected to the cerebral cortex via the thalamus, and then transmitted to the muscles via the spinal cord to implement the behavior. The behavior output is U=CB(SC|S CC ), where CB() is the functional function of the cortico-cerebellar system, SC is the sensory cortical afferent, S CC For the cortex-cerebellum synaptic connection weight, use the MLP (multilayer perceptron) network to establish the cortex-cerebellum system, and its synaptic connection weight is expressed as W a ,V a , sensory cortical afferents as feedback system state X a , as shown in Figure 3(b), then:

[0045]

[0046] Among them, T means transpose; f a (z) represents the functional function from the hidden layer to the output in the cortical-cerebellar system network, and z is (V a ) T σ((W a ) T x a ); σ(h) represents the functional function input to the hidden layer in the cortical-cerebellar system network, h is (W a ) T x...

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

The invention discloses an intrinsic motivation based self-cognition system for a motion balance robot and a control method. The system comprises an intrinsic motivation based cognition model composed of behavior generation, behavior evaluation and orientation mechanism; the behavior generation is the formation of a 'perception-motion' loop; the behavior evaluation is the formation of a 'motion-result' loop; and the orientation mechanism is used for connecting the behavior generation with the behavior evaluation. The method comprises the steps that: a cortex-cerebellum system calculates action output amount according to sensory cortex information fed back by an intelligent body; a cortex-striatum system in a basal ganglion obtains an evaluation value by utilizing the sensory cortex information fed back by the intelligent body and motor cortex information calculated by cerebellum; and the cortex-striatum system and the cortex-cerebellum system are subjected to synaptic modification. According to the system and the method, neurophysiology, cognitive psychology and robotology are combined, a cognition mechanism is described and realized in a mathematic mode, and the self-cognition problem of the robot is solved.

Description

technical field [0001] The invention relates to the field of motion balance robots, in particular to an autonomous cognitive system and control method for a motion balance robot based on intrinsic motivation. Background technique [0002] At present, the wheeled robot in the robot family has become an important branch in the field of robot research by virtue of its high theoretical value and practical value. Among them, the two-wheeled self-balancing robot has the characteristics of small footprint and flexible movement, and can be applied in special environments (such as: search and rescue in a small space or crowded places), and can be used as a special means of transportation in Many areas of modern society serve human beings. Different from other mobile robots, the motion balance ability of the two-wheeled self-balancing robot can automatically adjust the motion speed and body tilt angle. Its motion balance mechanism is consistent with the balance principle of human lim...

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): G06F19/00
Inventor 陈静李莉李冰于雅楠李宗帅
Owner TIANJIN UNIV OF TECH & EDUCATION TEACHER DEV CENT OF CHINA VOCATIONAL TRAINING & GUIDANCE
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