CN108646550A - A kind of multiple agent formation method of Behavior-based control selection - Google Patents
A kind of multiple agent formation method of Behavior-based control selection Download PDFInfo
- Publication number
- CN108646550A CN108646550A CN201810286164.3A CN201810286164A CN108646550A CN 108646550 A CN108646550 A CN 108646550A CN 201810286164 A CN201810286164 A CN 201810286164A CN 108646550 A CN108646550 A CN 108646550A
- Authority
- CN
- China
- Prior art keywords
- behavior
- income
- intelligent body
- based control
- control selection
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of multiple agent formation methods of Behavior-based control selection, include the following steps:Step 1: setting and the relevant detection income calculation method in position;Step 2: defining the concrete behavior of multiple intelligent bodies;Step 3: determining the number of channels of basal ganglion, basal ganglion channel pattern is established, initializes relevant parameter;Step 4: the channel pattern parameter of correction basal ganglion.Through the above way, the multiple agent formation method of Behavior-based control selection of the present invention, introduce a kind of dynamic and intelligent body formation method of Behavior-based control selection, it is not strong to efficiently solve robustness in existing formation detection method, the shortcomings of reliability is not strong passes through open parameter and every weight modification channel so that the technical solution of the application has more extensive versatility, target detection income variable is introduced, the income that multiple agent is detected task can be promoted on the whole.
Description
Technical field
The present invention relates to the formation fields of intelligent body, more particularly to a kind of multiple agent formation side of Behavior-based control selection
Method.
Background technology
Intelligent body can replace people to complete, in repetitive operation complicated, under hazardous environment, to accelerate the development of society.Closely
Nian Lai, market environment propose the function of intelligent body new demand, and intelligent body needs to complete increasingly complex Detection task very
To job task, but when task facing complexity, needing high efficiency, complete parallel, single intelligent body can not be competent at, and need
Multiple agent is wanted to cooperate with.
Intelligent body formation control is a kind of multiple intelligent bodies during reaching target formation, can form target team
Shape, and it is adapted to a kind of control method of specific environmental constraints.Existing common method mainly has based on navigator-follower
Method, virtual architecture method, the method for Behavior-based control, based on graph theory, the method for potential energy method.
The basic thought of wherein pilotage people-follower's method is that may exist in the fleet system that multiple agent forms
One or more navigator's intelligent body, other non-pilotage people's intelligent bodies are to follow intelligent body, follow intelligent body with its opposite neck
It navigates and is used as input control quantity in the position relative distance of intelligent body, relative angle so that follow intelligent body and navigator's intelligent body
Relative position infinitely approaches desired value.The control structure of navigator's follower method is fairly simple, but due to navigator's intelligent body and with
With not having position feedback between intelligent body, and it is the control of pilotage people's intelligent body single-point, causes to be susceptible to intelligent body and fall behind
The robustness of situation, system is poor.
The basic thought of virtual architecture method is that the system for forming multiple agent regards an imaginary rigid structure as, each
Coordinate of the intelligent body under reference frame is constant, i.e., the relative position between intelligent body is constant.
The basic thought of the method for Behavior-based control is that the links that multiple agent is formed into columns are regarded as by single intelligent body
Multiple basic acts are constituted, it is only necessary to which the control method for studying each basic act can be more by the combination of basic act control
Intelligent body, which is formed, forms into columns.Basic act generally comprises target following, avoids obstacle, avoids collision, form into columns generation and holding of forming into columns
Deng.The method of Behavior-based control is controlled by each intelligent body mutual perception, systematic comparison distributed AC servo system easy to implement, is had fine
Robustness, but since accurate mathematics model analysis can not be combined, cause cannot highly effective guarantee system it is reliable
Property.
Potential energy method is that another formation different from conventional method described above forms control method, it forms formation
All steps are combined into together, and close potential function is closed by building, and determine the control law of intelligent body by other intelligence by potential field function
Energy body acts on its different force, in conjunction with desired formation figure, so that intelligent body is moved towards expected formation figure, is arrived when respectively
When scheming up to formation, the potential energy glass of whole system is small.In formation forming process, formation target point is dynamic change, belongs to dynamic
State formation forming method, intelligent body are to reach an equilibrium state by constantly adjusting mutual distance;But formation is formed
Rapidity and controllability can not hold, due to being to form formation according to the formation figure of relative distance, then illustrating intelligent body
Between relative position be pre-determined in fact, this just makes the target location of intelligent body may not be best, and due to whole
The fixed formation figure of a relative position is kept in a formation target, it is assumed that an intelligent body fails to reach specified formation point,
His intelligent body will always be in the state of dynamical research minimum potential energy point and move always.
Invention content
The invention mainly solves the technical problem of providing a kind of multiple agent formation methods of Behavior-based control selection, solve
The problem of reliability of system can not ensure with robustness during multiple agent is formed into columns.
In order to solve the above technical problems, one aspect of the present invention is:A kind of selection of Behavior-based control is provided
Multiple agent formation method, includes the following steps:
Step 1: setting and the relevant detection income calculation method in position:
(1) income area in target acquisition position is set:The specifically following condition of target acquisition position income area setting:Income
It is circular several piece sector region that area, which is with target acquisition center,;
(2) income calculation method is detected:In the sector closer apart from income area, the detection for setting the intelligent body is checked and accepted
Benefit be and the relevant fixed income of target point significance level;It should in income area annulus farther out, set the intelligent body
Test point income is the dynamic income that distance is inversely proportional between intelligent body and target point, and upon completion of the assays, current work
The detection income in period will be reset;
Step 2: defining the concrete behavior of multiple intelligent bodies:Intelligent body is defined according to the concrete structure of intelligent body and function
After the completion of action selection it is possible that concrete behavior, inhibit relationship to carry out limitation and set point the excitation of each behavior
Analysis, if a certain intelligent body behavior can be excited, is verified using basal ganglion channel pattern;If the intelligent body being excited
Behavior can then execute this behavior by verification;
Step 3: determining the number of channels of basal ganglion, basal ganglion channel pattern is established, initializes related ginseng
Number:Basal ganglion mathematical model includes corpus straitum, globus pallidus outer core, subthalamic nucleus and core group globus pallidus kernel;
(1) mathematical model in behavior channel is established:Each channel integrates neuron with a leakage and indicates:
Wherein, x is the state of neuron, and u is the input of neuron, and y is the output of neuron, and H is jump function, remaining
For model parameter;
yi C=Si
Cerebral cortex is integrated, processing obtains the synthesis importance index S in conjunction with each factor in step 1i, wherein i is logical
Taoist monastic name, yi CCharacterize the output in i-th of channel in cerebral cortex;
The state of corpus straitum D1 can be described as:
Wherein, ui SD1It is corpus straitum D1, the neuron input in the channels i, ai SD1It is the state of the channel neuron, yi SD1It is
The output of the channel neuron, wCSD1It is weight of the cerebral cortex to corpus straitum D1;1+ λ-descriptions are dopamines to corpus straitum D1
Incentive action, εSD1For the output threshold value of corpus straitum D1;
The state of corpus straitum D2 can be described as:
Wherein, ui SD2It is corpus straitum D2, the neuron input in the channels i, ai SD2It is the state of the channel neuron, yi SD2It is
The output of the channel neuron, wCSD2It is weight of the cerebral cortex to corpus straitum D2,1- λ-descriptions are dopamines to corpus straitum D2
Inhibiting effect, εSD2For the output threshold value of corpus straitum D2;
Globus pallidus outer core can be described as:
Wherein, ui GPeIt is globus pallidus outer core, the neuron input in the channels i, ai GPeIt is the state of the channel neuron, yi GPe
It is the output of the channel neuron, wSD2GPeIt is weights of the corpus straitum D2 to globus pallidus outer core, εGPeIt is the output of globus pallidus outer core
Threshold value.
The model of subthalamic nucleus can be described as:
Wherein, ui STNIt is subthalamic nucleus, the neuron input in the channels i, ai STNIt is the state of the channel neuron, yi STNIt is
The output of the channel neuron, wGPeSTNIt is weight of the globus pallidus outer core to subthalamic nucleus, εSTNIt is the output threshold value of subthalamic nucleus;
According to anatomical research, subthalamic nucleus very disperses the projection of globus pallidus kernel, therefore grey in description
When the mathematical model of Archon kernel, it includes the subthalamic nucleus input in other left and right channels to enable the output of globus pallidus kernel;
It is described as:
Wherein, ui GPiIt is subthalamic nucleus, the neuron input in the channels i, ai GPiIt is the state of the channel neuron, yi GPiIt is
The output of the channel neuron, wSD1GPeIt is weights of the corpus straitum D1 to globus pallidus kernel, wSTNGPiFor subthalamic nucleus to globus pallidus
The output weight of kernel, εGPiIt is the output threshold value of globus pallidus kernel, value is it should be noted that wCSD1、wCSD2、 wSTNGPiFor positive value
Characterize incentive connection, wSD1GPi、wSD2GPe、wGPeSTNFor negative value, characterization inhibition connection;
Step 4: the channel pattern parameter of correction basal ganglion, when the overall operation situation of intelligent body is inclined with user
When well to be deviated under conditions of standard, the correction to model parameters in step 3 is carried out.
The beneficial effects of the invention are as follows:A kind of multiple agent formation method for Behavior-based control selection that the present invention points out, draws
The dynamic and intelligent body formation method for having entered a kind of selection of Behavior-based control efficiently solves in existing formation detection method robustness not
By force, the shortcomings of reliability is not strong passes through open parameter and every weight modification channel so that the technical solution of the application has more
Add extensive versatility, introduces target detection income variable, the receipts that multiple agent is detected task can be promoted on the whole
Benefit improves running efficiency of system indirectly.
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing, wherein:
Fig. 1 is the detection income in a kind of one preferred embodiment of multiple agent formation method of Behavior-based control selection of the present invention
Range schematic diagram, figure label 3 are the detection target that intelligent body needs detect, and label 2 is fixed test income area, and label 1 is
Dynamic income area;
Fig. 2 is 4 channel bases in a kind of one preferred embodiment of multiple agent formation method of Behavior-based control selection of the present invention
Bottom neuromere schematic diagram.
Specific implementation mode
The technical scheme in the embodiments of the invention will be clearly and completely described below, it is clear that described implementation
Example is only a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, this field is common
All other embodiment that technical staff is obtained without making creative work belongs to the model that the present invention protects
It encloses.
Intelligent body is considered to the sum of the distance of movement for forming formation target position dissum, the suitable shifting of intelligent body
Dynamic speed vi, intelligent body is adjusted to the time T of object-oriented postureadjust, intelligent body moves to the time T of target pointmove,
The detection income of intelligent body in-position after the next sampling period carries out action selection, Tcost=Tmove+Tadjust。
~2 are please referred to Fig.1, the embodiment of the present invention includes following steps:
Step 1: setting and the relevant detection income calculation method in position:
(1) income area in target acquisition position is set:Target acquisition position income area is set specifically according to following condition:It receives
Beneficial area is using target acquisition center as the several piece sector region in the center of circle;
(2) income calculation method is detected:In the sector closer apart from income area, the detection for setting the intelligent body is checked and accepted
Benefit be and the relevant fixed income of target point significance level;It should in income area annulus farther out, set the intelligent body
Test point income is the dynamic income that distance is inversely proportional between intelligent body and target point, and upon completion of the assays, current work
The detection income in period is reset;
Wherein ηprofitFor the dynamic income of current target point, ηkFor the fixed income of current target point, r be intelligent body away from
With a distance from target detection point, r1For the shortest distance in fixed test income area and target detection point, r2For dynamic income area and mesh
Mark the longest distance of test point;
Step 2: defining the concrete behavior of intelligent body:Intelligent body is defined according to the concrete structure of intelligent body with function to be expert at
For after the completion of selection it is possible that concrete behavior;The excitation of each behavior is inhibited relationship to carry out limitation and sets analysis,
If a certain intelligent body behavior can be excited, verified using basal ganglion channel pattern;If the intelligent body row being excited
Can then to execute this behavior by verification;
Design intelligent body posture behavior is rotated clockwise, rotated clockwise, advancing, retreating, searching for and six kinds of backhaul:
Wherein siCharacterize the significance level of each behavior, ηi kFor constant to be adjusted;
Step 3: determining the number of channels of basal ganglion, basal ganglion channel pattern is established, initializes related ginseng
Number;
Basal ganglion mathematical model includes corpus straitum, globus pallidus outer core, subthalamic nucleus and core group globus pallidus kernel;
(1) mathematical model in behavior channel is established:Each channel integrates neuron with a leakage and indicates:
Wherein, x is the state of neuron, and u is the input of neuron, and y is the output of neuron, and H is jump function, remaining
For model parameter.
yi C=Si
Cerebral cortex is integrated, processing obtains the synthesis importance index S in conjunction with each factor in step 1i, wherein i is logical
Taoist monastic name, yi CCharacterize the output in i-th of channel in cerebral cortex;
The state of corpus straitum D1 can be described as:
Wherein, ui SD1It is corpus straitum D1, the neuron input in the channels i, ai SD1It is the state of the channel neuron, yi SD1It is
The output of the channel neuron, wCSD1It is weight of the cerebral cortex to corpus straitum D1.1+ λ-descriptions are dopamines to corpus straitum D1
Incentive action, εSD1For the output threshold value of corpus straitum D1;
The state of corpus straitum D2 can be described as:
Wherein, ui SD2It is corpus straitum D2, the neuron input in the channels i, ai SD2It is the state of the channel neuron, yi SD2It is
The output of the channel neuron, wCSD2It is weight of the cerebral cortex to corpus straitum D2,1- λ-descriptions are dopamines to corpus straitum D2
Inhibiting effect, εSD2For the output threshold value of corpus straitum D2;
Globus pallidus outer core can be described as:
Wherein, ui GPeIt is globus pallidus outer core, the neuron input in the channels i, ai GPeIt is the state of the channel neuron, yi GPe
It is the output of the channel neuron, wSD2GPeIt is weights of the corpus straitum D2 to globus pallidus outer core, εGPeIt is the output of globus pallidus outer core
Threshold value;
The model of subthalamic nucleus can be described as:
Wherein, ui STNIt is subthalamic nucleus, the neuron input in the channels i, ai STNIt is the state of the channel neuron, yi STNIt is
The output of the channel neuron, wGPeSTNIt is weight of the globus pallidus outer core to subthalamic nucleus, εSTNIt is the output threshold value of subthalamic nucleus;
According to anatomically studying, subthalamic nucleus very disperses the projection of globus pallidus kernel, therefore grey in description
When the mathematical model of Archon kernel, it includes the subthalamic nucleus input in other left and right channels to enable the output of globus pallidus kernel;
It is described as:
Wherein, ui GPiIt is subthalamic nucleus, the neuron input in the channels i, ai GPiIt is the state of the channel neuron, yi GPiIt is
The output of the channel neuron, wSD1GPeIt is weights of the corpus straitum D1 to globus pallidus kernel, wSTNGPiFor subthalamic nucleus to globus pallidus
The output weight of kernel, εGPiIt is the output threshold value of globus pallidus kernel, value is it should be noted that wCSD1、wCSD2、 wSTNGPiFor positive value
Characterize incentive connection, wSD1GPi、wSD2GPe、wGPeSTNFor negative value, characterization inhibition connection;
Step 4:Correct the channel pattern parameter of basal ganglion:When the overall operation situation of intelligent body is inclined with user
When well to be deviated under conditions of standard, the correction to model parameters in step 3 is carried out.Finally according in globus pallidus
Intelligent body behavior is selected in core output.
In conclusion a kind of multiple agent formation method for Behavior-based control selection that the present invention points out, by introducing substrate
Neuromere action selection is formed into columns strategy, and the reliability and robustness of system are enhanced, can be in different function, different number scale, no
It is applied in intelligent body cluster with detection target, Different Exercise Mode and target detection mode.
Example the above is only the implementation of the present invention is not intended to limit the scope of the invention, every to utilize this hair
Equivalent structure or equivalent flow shift made by bright description is applied directly or indirectly in other relevant technology necks
Domain is included within the scope of the present invention.
Claims (6)
1. a kind of multiple agent formation method of Behavior-based control selection, which is characterized in that include the following steps:
Step 1: setting and the relevant detection income calculation method in position;
Step 2: defining the concrete behavior of multiple intelligent bodies;
Step 3: determining the number of channels of basal ganglion, basal ganglion channel pattern is established, initializes relevant parameter;
Step 4: the channel pattern parameter of correction basal ganglion.
2. the multiple agent formation method of Behavior-based control selection according to claim 1, which is characterized in that the step 1
In, income area is using target acquisition center as the several piece sector region in the center of circle, and in the sector closer apart from income area, setting should
The test point income of intelligent body be and the relevant fixed income of target point significance level;In the annulus apart from income area farther out
It is interior, the dynamic income that test point income distance as between intelligent body and target point of the intelligent body is inversely proportional is set, is being examined
After the completion of survey, the detection income in current work period will be reset.
3. the multiple agent formation method of Behavior-based control selection according to claim 1, which is characterized in that the step 2
In, the excitation of each behavior is inhibited relationship to carry out limitation and sets analysis, if a certain intelligent body behavior can be excited, is used
Basal ganglion channel pattern is verified;If the intelligent body behavior being excited can execute this behavior by verification.
4. the multiple agent formation method of Behavior-based control selection according to claim 1, which is characterized in that the step 3
In, basal ganglion mathematical model includes corpus straitum, globus pallidus outer core, subthalamic nucleus and core group globus pallidus kernel.
5. the multiple agent formation method of Behavior-based control selection according to claim 1, which is characterized in that the step 4
In, when the overall operation situation of intelligent body is deviated under conditions of using user preference as standard, carry out in step 3
The correction of model parameters.
6. the multiple agent formation method of Behavior-based control selection according to claim 1, which is characterized in that multiple intelligent bodies
Allow the functional characteristic for having different, is controlled by action selection and promote overall operation efficiency.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810286164.3A CN108646550B (en) | 2018-04-03 | 2018-04-03 | Multi-agent formation method based on behavior selection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810286164.3A CN108646550B (en) | 2018-04-03 | 2018-04-03 | Multi-agent formation method based on behavior selection |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108646550A true CN108646550A (en) | 2018-10-12 |
CN108646550B CN108646550B (en) | 2022-03-22 |
Family
ID=63745435
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810286164.3A Active CN108646550B (en) | 2018-04-03 | 2018-04-03 | Multi-agent formation method based on behavior selection |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108646550B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108897315A (en) * | 2018-06-12 | 2018-11-27 | 南京理工大学 | A kind of Multi-agent Team Formation |
CN109725532A (en) * | 2018-12-24 | 2019-05-07 | 杭州电子科技大学 | One kind being applied to relative distance control and adaptive corrective method between multiple agent |
CN113268893A (en) * | 2021-07-19 | 2021-08-17 | 中国科学院自动化研究所 | Group trapping method and device based on communication maintenance constraint |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8379967B1 (en) * | 2008-12-04 | 2013-02-19 | Lockheed Martin Corporation | Swarm control systems and methods |
CN104925059A (en) * | 2015-06-02 | 2015-09-23 | 奇瑞汽车股份有限公司 | Automatic car-following method and device |
CN105205533A (en) * | 2015-09-29 | 2015-12-30 | 华北理工大学 | Development automatic machine with brain cognition mechanism and learning method of development automatic machine |
CN105467981A (en) * | 2015-12-24 | 2016-04-06 | 中南大学 | Method and device for formation aiming at multiple intelligent agents |
JP2016095754A (en) * | 2014-11-17 | 2016-05-26 | 日本電信電話株式会社 | Action control system, and method, and program, for the same |
CN106843269A (en) * | 2017-01-22 | 2017-06-13 | 北京航空航天大学 | A kind of unmanned plane formation method based on small birds cluster fly mechanics |
CN107121986A (en) * | 2017-05-24 | 2017-09-01 | 浙江大学 | The method that a kind of unmanned plane flight pattern of Behavior-based control is kept |
CN109144047A (en) * | 2018-06-12 | 2019-01-04 | 南京理工大学 | A kind of finite time formation control method based on multi-agent system |
-
2018
- 2018-04-03 CN CN201810286164.3A patent/CN108646550B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8379967B1 (en) * | 2008-12-04 | 2013-02-19 | Lockheed Martin Corporation | Swarm control systems and methods |
JP2016095754A (en) * | 2014-11-17 | 2016-05-26 | 日本電信電話株式会社 | Action control system, and method, and program, for the same |
CN104925059A (en) * | 2015-06-02 | 2015-09-23 | 奇瑞汽车股份有限公司 | Automatic car-following method and device |
CN105205533A (en) * | 2015-09-29 | 2015-12-30 | 华北理工大学 | Development automatic machine with brain cognition mechanism and learning method of development automatic machine |
CN105467981A (en) * | 2015-12-24 | 2016-04-06 | 中南大学 | Method and device for formation aiming at multiple intelligent agents |
CN106843269A (en) * | 2017-01-22 | 2017-06-13 | 北京航空航天大学 | A kind of unmanned plane formation method based on small birds cluster fly mechanics |
CN107121986A (en) * | 2017-05-24 | 2017-09-01 | 浙江大学 | The method that a kind of unmanned plane flight pattern of Behavior-based control is kept |
CN109144047A (en) * | 2018-06-12 | 2019-01-04 | 南京理工大学 | A kind of finite time formation control method based on multi-agent system |
Non-Patent Citations (6)
Title |
---|
TUCKER BALCH 等: "Behavior-based Formation Control for Multi-robot Teams", 《IEEE》 * |
吴建国编著: "第六章 多智能体及行为", 《仿生机器人及其智能运动控制》 * |
王义萍 等: "基底神经节的尖峰神经元网络模型及其在机器人中的应用", 《南京理工大学学报(自然科学版)》 * |
王义萍 等: "机器人行为选择机制综述", 《机器人》 * |
王义萍: "基于基底神经节的机器人行为选择与行为序列学习方法", 《中国博士学位论文全文数据库 信息科技辑》 * |
王兢: "水下机器人编队系统研究", 《中国优秀博硕士学位论文全文数据库 (硕士) 信息科技辑》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108897315A (en) * | 2018-06-12 | 2018-11-27 | 南京理工大学 | A kind of Multi-agent Team Formation |
CN109725532A (en) * | 2018-12-24 | 2019-05-07 | 杭州电子科技大学 | One kind being applied to relative distance control and adaptive corrective method between multiple agent |
CN109725532B (en) * | 2018-12-24 | 2021-11-16 | 杭州电子科技大学 | Relative distance control and self-adaptive correction method applied to multiple intelligent agents |
CN113268893A (en) * | 2021-07-19 | 2021-08-17 | 中国科学院自动化研究所 | Group trapping method and device based on communication maintenance constraint |
Also Published As
Publication number | Publication date |
---|---|
CN108646550B (en) | 2022-03-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Bremer | The GLOBUS model: Computer simulation of worldwide political and economic developments | |
Schartau et al. | Reviews and syntheses: parameter identification in marine planktonic ecosystem modelling | |
CN106781489B (en) | A kind of road network trend prediction method based on recurrent neural network | |
CN108897315A (en) | A kind of Multi-agent Team Formation | |
CN108646550A (en) | A kind of multiple agent formation method of Behavior-based control selection | |
CN108596335A (en) | A kind of adaptive crowdsourcing method based on deeply study | |
CN109360604A (en) | A kind of oophoroma molecule parting forecasting system | |
Lin et al. | Machine learning templates for QCD factorization in the search for physics beyond the standard model | |
CN103605711B (en) | Construction method and device, classification method and device of support vector machine | |
CN104835181B (en) | A kind of method for tracking target based on ranking and fusing study | |
CN107481218A (en) | Image aesthetic feeling appraisal procedure and device | |
CN110299008A (en) | A kind of traffic flow multistep forecasting method based on intensified learning | |
CN108732931A (en) | A kind of multi-modal batch process modeling method based on JIT-RVM | |
CN109885916A (en) | A kind of bulk testing on-time model update method based on LSSVM | |
CN111126220A (en) | Real-time positioning method for video monitoring target | |
CN112405542A (en) | Musculoskeletal robot control method and system based on brain inspiring multitask learning | |
CN103605493B (en) | Sorting in parallel learning method based on Graphics Processing Unit and system | |
CN110490324A (en) | A kind of gradient decline width learning system implementation method | |
Hazenfratz et al. | Neural networks (SOM) applied to INAA data of chemical elements in archaeological ceramics from Central Amazon | |
CN114186711A (en) | Industrial raw material consumption prediction method based on multitask time sequence learning | |
Tailor et al. | Comparative analysis of software cost and effort estimation methods: a review | |
CN103675010B (en) | The industrial melt index soft measurement instrument of support vector machine and method | |
CN105787265A (en) | Atomic spinning top random error modeling method based on comprehensive integration weighting method | |
Gu et al. | Quantifying the effects of different data streams on the calibration of building energy simulation | |
CN107168066A (en) | A kind of greenhouse self-adaptation control method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |