CN106126493A - A kind of natural language analytic method based on robot autonomous behavior - Google Patents
A kind of natural language analytic method based on robot autonomous behavior Download PDFInfo
- Publication number
- CN106126493A CN106126493A CN201610429120.2A CN201610429120A CN106126493A CN 106126493 A CN106126493 A CN 106126493A CN 201610429120 A CN201610429120 A CN 201610429120A CN 106126493 A CN106126493 A CN 106126493A
- Authority
- CN
- China
- Prior art keywords
- natural language
- behavior
- robot
- stem
- analytic method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
Abstract
The invention belongs to natural language analytic technique field, be specifically related to a kind of natural language analytic method based on robot autonomous behavior.The behavior of autonomous robot is divided into basis behavior module, i.e. motor control, motion planning, object run module, for each module definition natural language attribute, the natural language analytic method that design combines based on stem knowledge base and logical reasoning rule, each module gives linguistic property, set up corresponding natural language stem knowledge base, in conjunction with stem attribute and the internal relation of logical out rule, it is achieved natural language mates with inner directed behavior.The present invention provides a kind of natural language analytic method based on robot autonomous behavior for autonomous robot, realize in man-machine interaction, by natural language instruction, robot autonomous behavior is assisted, human intelligence and machine intelligence are combined, make man-machine interaction mode more natural, it is possible to complete task more conveniently.
Description
Technical field
The invention belongs to natural language analytic technique field, be specifically related to a kind of natural language based on robot autonomous behavior
Speech analytic method.
Background technology
Artificial its of librarian use intelligent machine by professional training does not services the most direct communication way is natural language,
But the linguistic form that natural language is the most wide in range and changeable sentence pattern structure to robot to the understanding of natural language and decision-making band
Come difficult, and for autonomous robot, it is understood that abundant changeable natural language is extremely difficult.Present nature
The research that language resolves also tends to only pay attention to the understanding of the natural language part of speech meaning of a word itself, it practice, artificial at autonomous machine
In work it is understood that natural language necessarily match with its inner directed behavior ability, be not required to all natural languages are entered
Row understands fractionation, adds the difficulty that natural language resolves and the efficiency reducing man-machine interaction the most beyond doubt, thus single
Single needle can not meet, for the morphological analysis of pure natural language, the requirement that natural language instruction is resolved by autonomous robot.This
Bright it is intended to reduce the parsing difficulty that the most wide in range natural language brings, targetedly to relevant to robot autonomous behavior
Natural language resolves, and designs corresponding analytic method, is efficiently completed natural language and resolves task, assists man-machine interaction.
The inner directed behavior (including object run behavior, motion planning behavior and motor control behavior) of robot is drawn by one aspect of the present invention
Being divided into basis behavior module, and be each basis behavior module definition natural language attribute, on the other hand design is known based on stem
The natural language analytic method that knowledge storehouse and logical reasoning rule combine, it is achieved natural language is effective with robot autonomous behavior
Coupling.The sound instruction of operator can be resolved into executable inner directed behavior sequence and complete instruction task by robot, passes through
Natural language instruction stem is extracted, and scans for coupling in basis stem knowledge base, in conjunction with basis behavior knowledge
Storehouse, it is achieved natural language is effectively matched with robot autonomous behavior.This natural language based on basis behavior linguistic property
Analytic method can solve the problem that the difficult problem that language enrichment brings, it is established that more efficient and ambiguity less man-machine interaction pattern,
It it is the effective way of robot hommization.
Summary of the invention
It is an object of the invention to provide a kind of natural language analytic method based on robot autonomous behavior.
The object of the present invention is achieved like this:
A kind of natural language analytic method based on robot autonomous behavior, is divided into basis by the behavior of autonomous robot
Behavior module, i.e. motor control, motion planning, object run module, for each module definition natural language attribute, design based on
The natural language analytic method that stem knowledge base and logical reasoning rule combine, each module gives linguistic property, sets up phase
The natural language stem knowledge base answered, in conjunction with logical reasoning rule, it is achieved natural language mates, by nature with inner directed behavior
Sound instruction character string extracts all stems with the search of mating of stem knowledge base, and advises according to stem attribute and logical reasoning
Internal relation then, the stem completing natural language instruction decomposes, it is achieved stem and the Proper Match of robot base's behavior, machine
Natural language instruction is resolved to executable inner directed behavior sequence and completes instruction task by device people.
Described natural language analytic method makes robot learn natural language instruction, and the destination of study is abundant
Stem storehouse, basis, behavior storehouse, basis, inferenctial knowledge storehouse, the object knowledge storehouse with linguistic property and environmental map, the hands of study
Section is to evaluate mechanism by natural language AC mode and implementation effect to realize.
The beneficial effects of the present invention is:
The present invention provides a kind of natural language analytic method based on robot autonomous behavior for autonomous robot, it is achieved
In man-machine interaction, by natural language instruction, robot autonomous behavior is assisted, human intelligence is tied mutually with machine intelligence
Close so that man-machine interaction mode is more natural, it is possible to complete task more conveniently.
Accompanying drawing explanation
Fig. 1 is the natural language stem knowledge base model that the present invention decomposes based on inner directed behavior;
Fig. 2 is that natural language of the present invention instructs process of analysis.
Detailed description of the invention
Below in conjunction with the accompanying drawings the present invention is described further.
The invention discloses a kind of natural language analytic method based on robot autonomous behavior, it is intended to use natural language
The behavior to autonomous robot that instructs carries out assist control.The inner directed behavior of robot is divided into basis behavior module, and is
Each basis behavior module definition natural language attribute, classifies for the behavior of autonomous robot basis, by its inner directed behavior
Carry out natural language description, and determine stem natural language attribute.By natural language instruction stem is extracted, and
On basis, stem knowledge base scans for coupling, in conjunction with behavior knowledge storehouse, basis, it is achieved natural language and robot autonomous behavior
Be effectively matched, on this basis, the sound instruction of operator can be resolved into executable inner directed behavior sequence by robot
Complete instruction task.Natural language instruction being extended through study complete, basis behavioral competence study can be had by abundant
The object knowledge storehouse of linguistic property and environmental map realize, and abundant means can be handed over by the instruction of auxiliary environment perception natural language
Streamed realization, the enhancing of basis behavioral competence can abundant with optimized integration stem storehouse, extension class of languages inferenctial knowledge is permissible
Strengthen the adaptation ability that natural language is accustomed to by robot.The destination of study is stem storehouse, abundant basis, basis behavior storehouse, logic
Rule of inference, the object knowledge storehouse with linguistic property and environmental map, the means of study can pass through natural language AC mode
Evaluate mechanism with implementation effect to realize.
The invention discloses a kind of natural language analytic method based on robot autonomous behavior, in order to reduce natural language
Rich and irregularities resolves the difficulty brought to natural language, designs the natural language matched with robot autonomous behavior
Speech, reduces the difficulty that language resolves, and the inner directed behavior of robot is divided into basis behavior module, and is each basis behavior mould
Block definition natural language attribute, by extracting natural language instruction stem, and searches in basis stem knowledge base
Rope mates, in conjunction with the internal relation of stem attribute with logical reasoning rule, it is achieved natural language has with robot autonomous behavior
Effect coupling, on this basis, the natural language Command Resolution of operator can be become executable inner directed behavior sequence by robot
Complete instruction task.The independent learning ability of robot by abundant basis stem storehouse, basis behavior storehouse, logical reasoning rule,
Object knowledge storehouse and the environmental map etc. with linguistic property realize, the means of study can by natural language AC mode and
Implementation effect is evaluated mechanism and is realized.The natural language that present invention achieves robot autonomous behavior resolves, and generates corresponding machine
Device people's inner directed behavior sequence, enriches language knowledge base by the raising of inner directed behavior ability so that robot obtains autonomous row
For learning capacity, and then improve interactive capability.
Refering to Fig. 1, the present invention is directed to the various inner directed behavior of robot and establish a stem knowledge base, first by robot
Object run, motion planning, the three class inner directed behaviors such as motor control be divided into basis behavior module, set up behavior storehouse, basis,
And be each basis behavior module definition inner directed behavior attribute, basis behavior simultaneously all has target, such as the target captured, arrives
The place that reaches, the direction of motion and distance etc., based on behavioral objective define natural language attribute, by natural language defined above
Speech classification is set up stem storehouse, basis and is come behavior storehouse, corresponding corresponding basis.Motor control stem and motor control mesh can be set up out
Mark stem correspondence motor control behavior;Motion planning stem and the corresponding motion planning behavior of motion planning target stem;Target is grasped
Make stem and operation target stem corresponding object run behavior;Objective attribute target attribute stem correspondence attribute objectives operation behavior;Target is closed
The dry corresponding motion planning of copula and object run behavior;Auxiliary environment perception stem correspondence auxiliary environment perception behavior.
The present invention provides the natural language analytic method based on robot autonomous behavior decomposition to be based on basis stem
Knowledge base and logical reasoning rule carry out resolving coupling, in conjunction with behavior knowledge storehouse, basis, robot autonomous behavior are carried out nature
Sound instruction controls, and process of analysis refers to Fig. 2.First pass through natural language instruction character string and stem knowledge base mates search
Extracting all stems, and extract stem respective attributes, the stem completing natural language instruction decomposes, then by setting up
The internal relation of stem attribute and logical reasoning rule realizes the Proper Match of stem and robot base's behavior, generates corresponding
Robot behavior sequence.
Claims (2)
1. a natural language analytic method based on robot autonomous behavior, it is characterised in that: by the behavior of autonomous robot
It is divided into basis behavior module, i.e. motor control, motion planning, object run module, belongs to for each module definition natural language
Property, the natural language analytic method that design combines based on stem knowledge base and logical reasoning rule, each module gives language
Attribute, sets up corresponding natural language stem knowledge base, in conjunction with logical reasoning rule, it is achieved natural language and inner directed behavior
Join, extract all stems by natural language instruction character string with the search of mating of stem knowledge base, and according to stem attribute
With the internal relation of logical reasoning rule, the stem completing natural language instruction decomposes, it is achieved stem and robot base's behavior
Proper Match, natural language instruction is resolved to executable inner directed behavior sequence and completes instruction task by robot.
Natural language analytic method based on robot autonomous behavior the most according to claim 1, it is characterised in that: described
Natural language analytic method make robot that natural language instruction to be learnt, the destination of study be stem storehouse, abundant basis,
Behavior storehouse, basis, inferenctial knowledge storehouse, the object knowledge storehouse with linguistic property and environmental map, the means of study are by nature
Communication pattern and implementation effect are evaluated mechanism and are realized.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610429120.2A CN106126493A (en) | 2016-06-16 | 2016-06-16 | A kind of natural language analytic method based on robot autonomous behavior |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610429120.2A CN106126493A (en) | 2016-06-16 | 2016-06-16 | A kind of natural language analytic method based on robot autonomous behavior |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106126493A true CN106126493A (en) | 2016-11-16 |
Family
ID=57469532
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610429120.2A Pending CN106126493A (en) | 2016-06-16 | 2016-06-16 | A kind of natural language analytic method based on robot autonomous behavior |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106126493A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110187873A (en) * | 2019-06-03 | 2019-08-30 | 秒针信息技术有限公司 | A kind of rule code generation method and device |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104699708A (en) * | 2013-12-09 | 2015-06-10 | 中国移动通信集团北京有限公司 | Self-learning method and device for customer service robot |
CN104950896A (en) * | 2015-07-14 | 2015-09-30 | 上海智臻网络科技有限公司 | Floor sweeping robot, server and floor sweeping robot service system |
CN104965426A (en) * | 2015-06-24 | 2015-10-07 | 百度在线网络技术(北京)有限公司 | Intelligent robot control system, method and device based on artificial intelligence |
-
2016
- 2016-06-16 CN CN201610429120.2A patent/CN106126493A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104699708A (en) * | 2013-12-09 | 2015-06-10 | 中国移动通信集团北京有限公司 | Self-learning method and device for customer service robot |
CN104965426A (en) * | 2015-06-24 | 2015-10-07 | 百度在线网络技术(北京)有限公司 | Intelligent robot control system, method and device based on artificial intelligence |
CN104950896A (en) * | 2015-07-14 | 2015-09-30 | 上海智臻网络科技有限公司 | Floor sweeping robot, server and floor sweeping robot service system |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110187873A (en) * | 2019-06-03 | 2019-08-30 | 秒针信息技术有限公司 | A kind of rule code generation method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107168945A (en) | A kind of bidirectional circulating neutral net fine granularity opinion mining method for merging multiple features | |
CN109492232A (en) | A kind of illiteracy Chinese machine translation method of the enhancing semantic feature information based on Transformer | |
CN109711529A (en) | A kind of cross-cutting federal learning model and method based on value iterative network | |
CN105843381A (en) | Data processing method for realizing multi-modal interaction and multi-modal interaction system | |
CN107704456A (en) | Identify control method and identification control device | |
CN106126493A (en) | A kind of natural language analytic method based on robot autonomous behavior | |
CN110379234A (en) | A kind of study coach method and device | |
Kelly | Apprehending felt futures in Broome, Australia | |
CN110147550A (en) | Pronunciation character fusion method neural network based | |
Wang et al. | Learning to Follow and Generate Instructions for Language-Capable Navigation | |
Petit et al. | Hierarchical action learning by instruction through interactive grounding of body parts and proto-actions | |
Liu et al. | Practice and exploration of artificial intelligence education in universities of political science and law with Python | |
CN106326208B (en) | A kind of system and method that robot is trained by voice | |
CN114880347A (en) | Method for converting natural language into SQL statement based on deep learning | |
Hashimoto et al. | Extraction of editing processes for modifying a program to another purpose based on representation learning | |
Wolf et al. | Linking Speech and Gesture in Multimodal Instruction Systems | |
Zhou | Application Research of Teaching Robot Based on ChatGPT | |
CN107037740A (en) | A kind of Virtual Maintenance Simulation automatic generating method and device | |
Mendoza et al. | OBOC: Ontology based object categorisation for robots | |
Wang | What have we done and where will we go?-Brief Review and Outlook of Education for Sustainable Development in China | |
Formoso | Jean-Pierre Bocquet-Appel and the interdisciplinary dialogue Jean-Pierre Bocquet-Appel et le dialogue interdisciplinaire | |
Ogino et al. | How can humanoid acquire lexicon?-active approach by attention and learning biases based on curiosity | |
Xue et al. | A Hierarchical Motion Retrieval Algorithm for Complex Manipulation Tasks Planning with An Encoded Knowledge Base | |
Filippo Chiara et al. | Goal-driven Self-Attentive Recurrent Networks for Trajectory Prediction | |
Cui et al. | The research between cognitive linguistics and metaphor translation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161116 |