CN107053184A - Multi-Agent Cooperation processing system and method based on private clound - Google Patents
Multi-Agent Cooperation processing system and method based on private clound Download PDFInfo
- Publication number
- CN107053184A CN107053184A CN201710482720.XA CN201710482720A CN107053184A CN 107053184 A CN107053184 A CN 107053184A CN 201710482720 A CN201710482720 A CN 201710482720A CN 107053184 A CN107053184 A CN 107053184A
- Authority
- CN
- China
- Prior art keywords
- robot
- private clound
- image data
- module
- agent cooperation
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1679—Programme controls characterised by the tasks executed
- B25J9/1682—Dual arm manipulator; Coordination of several manipulators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
-
- 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
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/39—Robotics, robotics to robotics hand
- G05B2219/39146—Swarm, multiagent, distributed multitask fusion, cooperation multi robots
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention provides a kind of Multi-Agent Cooperation processing system based on private clound, its collaborative task being used between medium-sized group of Robocup multiple Soccer robots, it is characterised in that including:Private clound console module, for handling the real time information that the multiple Soccer robot is obtained and sent;Robot computer vision module, for gathering scene information in real time by panoramic vision and body feeling interaction formula equipment, the body feeling interaction formula equipment catches the action of human body and analyzes the posture of human body;Motion planning and robot control module, for detecting the positional information of the Soccer robot and controlling the behavior of the Soccer robot;Display content update module, the state for showing each Soccer robot in real time.Additionally provide the Multi-Agent Cooperation processing method based on private clound.
Description
Technical field
The present invention relates to the Multi-Agent Cooperation processing system based on private clound and method.
Background technology
Seafile is domestic team's exploitation, enterprise's cloud storage software of increasing income leading in the world, and private clound ring is provided for enterprise
Dropbox solution under border, meets the demands such as centralized file management, multiple terminal access, shared collaboration.
Robocup is RoboCup, is worldwide largest at present, participates in national and number at most, most has shadow
Ring the robot match of power.
Kinect is the name that Microsoft formally issues on June 14th, 2010 to XBOX360 body-sensings periphery peripheral hardware.It is a kind of
Body feeling interaction formula equipment, can catch the action of human body, analyze the posture of human body.
OpenCV full name is Open Source Computer Vision Library.OpenCV is one and is based on (opening
Source) distribution cross-platform computer vision library, may operate in Linux, Windows and Mac OS operating systems, realize
Many general-purpose algorithms in terms of image procossing and computer vision.
VS2015 is the IDE that Microsoft Corporation is produced.
The prior art data related to the application also includes:Soft project term (GB/T 11457-1995), computer
Software development specification (GB 8566), computer software product exploitation documentation guide (GB 8567), software kit quality requirement
With test (GB/T 17544-1998).
The content of the invention
According to an aspect of the invention, there is provided a kind of Multi-Agent Cooperation processing system based on private clound, it is used
Collaborative task between medium-sized group of Robocup multiple Soccer robots, it is characterised in that including:
Private clound console module, for handling the real time information that the multiple Soccer robot is obtained and sent,
Robot computer vision module, believes for gathering scene in real time by panoramic vision and body feeling interaction formula equipment
Breath, the body feeling interaction formula equipment catches the action of human body and analyzes the posture of human body,
Motion planning and robot control module, for detecting the positional information of the Soccer robot and controlling the football machine
The behavior of people,
Display content update module, the state for showing each Soccer robot in real time.
According to the further aspect of the present invention, above-mentioned Multi-Agent Cooperation processing system is characterised by:
The body feeling interaction formula equipment is Kinect.
According to another aspect of the present invention there is provided a kind of Multi-Agent Cooperation processing method based on private clound, its
For the collaborative task between medium-sized group of Robocup multiple Soccer robots, it is characterised in that including:
A private clound console module) is utilized, the real time information that the multiple Soccer robot is obtained and sent is handled,
B robot computer vision module) is utilized, gathering scene in real time by panoramic vision and body feeling interaction formula equipment believes
Breath, the body feeling interaction formula equipment catches the action of human body and analyzes the posture of human body,
C motion planning and robot control module) is utilized, the positional information of the Soccer robot is detected and controls the football machine
The behavior of device people,
D display content update module) is utilized, the state of each Soccer robot is shown in real time.
Brief description of the drawings
Fig. 1 shows the Multi-Agent Cooperation system according to an embodiment of the invention based on private clound.
Fig. 2 is the software block diagram of the Multi-Agent Cooperation system according to an embodiment of the invention based on private clound.
Fig. 3 is the high in the clouds program flow of the Multi-Agent Cooperation system according to an embodiment of the invention based on private clound
Cheng Tu.
Fig. 4 is the decision-making core program of the Multi-Agent Cooperation system according to an embodiment of the invention based on private clound
Flow chart.
Fig. 5 is the robot program of the Multi-Agent Cooperation system according to an embodiment of the invention based on private clound
Flow chart.
Fig. 6 shows the operation of private clound console module according to an embodiment of the invention.
Fig. 7 shows the operation of vision module according to an embodiment of the invention.
Embodiment
The invention provides a kind of private cloud framework suitable for Multi-Agent Cooperation.By the present inventor right
Medium-sized group of Soccer robot of Robocup go deep into technical research and combine to the understanding of traditional multi intelligent agent on the basis of,
Demand combing is carried out to Multi computer cooperation task, it is proposed that the solution of the private cloud framework suitable for Multi-Agent Cooperation of the invention
Scheme, it includes computer vision algorithms make, the self-positioning algorithm of intelligent robot, robot platform upgrading, forms private clound
Framework.
Following running environment is applied to according to the private cloud framework suitable for Multi-Agent Cooperation of the present invention:
Hardware environment:Medium-sized group of Soccer robot platform, cloud hardware platform, Intel (R) Core (TM) i7-2600K
3.40GHz processors, 8G internal memories, the special buses of USB 3.0 (Intel and Rui Sa controllers), 64 (x64) processors;
Software environment:Operating system MS Windows 8, Visual Studio 2015 (C++), Seafile, OpenCV
Storehouse, Kinect for Windows SDK v2.0.
Detection in real time, feedback, cloud data are met according to the private cloud framework suitable for Multi-Agent Cooperation of the present invention
Interaction, the time response requirement of display.
Embodiments of the invention are illustrated below in conjunction with accompanying drawing.
As shown in figure 1, the Multi-Agent Cooperation system according to an embodiment of the invention based on private clound includes showing
Show layer, logical layer, data Layer, the big layer of mechanical floor four, wherein:
● mechanical floor:Cloud Server, the network equipment, storage device;Three-wheel Omni-mobile platform, holding mechanism, kicking machine
Structure, panoramic vision structure, kinect.
● data Layer:It is responsible for obtaining the required machine surrounding environment image data of program operation and machine from hardware device
Device positional information, then transfers data to private clound.
● logical layer:Private clound in designed algorithm process color depth view data by identifying key therein
Information, judges obstacle information in environment etc.;Interaction logic, the keyboard mutuality operation of management platform.
● display layer:Robot location and environmental information are shown on screen.
Software platform module and function
As shown in table 1, the software of the Multi-Agent Cooperation system according to an embodiment of the invention based on private clound
Platform includes private clound console module, robot computer vision module, motion planning and robot control module, display content and updates mould
Block.
The software platform module of table 1 constitutes explanation
Private clound console module
(1) function
Design optimization algorithm, the real time information that handling machine people sends reduces unit computational complexity, improves system fortune
Line efficiency.
(2) performance
Every machine reduces by 20% data processing amount.
(3) input
The image information of five robots, positional information on arena.
(4) export
Competition field real time information after processing, fusion.
Fig. 6 shows the operation of private clound console module according to an embodiment of the invention.As shown in fig. 6, privately owned
Cloud Server (such as by WiFi) receives colour and depth image data from Duo Tai robots, according to what is stored in advance
Algorithm is handled image, obtains the information such as field wire, place, ball, barrier in place.
Robot computer vision module
(1) function
Scene information is gathered by panoramic vision and Kinect2.0 in real time.
(2) performance
Reaction time<0.1 second.
(3) input
Image with colour information and depth information.
(4) export
The information such as place, field wire, ball, barrier in scene.
Fig. 7 shows the operation of vision module according to an embodiment of the invention.As shown in fig. 7, Kinect passes through
Colour TV camera and thermal camera obtain (such as resolution ratio 1280 × 960) coloured image with (for example resolution ratio 640 ×
480) depth image data, obtain the color image information and the distance of each position to camera in place.
Motion planning and robot control module
(1) function
Detect robot positional information, control machine people's behavior
(2) performance
Detect elapsed time<0.1 second.
(3) input
The coloured image that robot code-disc information, machine vision head are collected.
(4) export
Value of statistical indicant (indicates whether there is experiencer);
Customer position information.
Display content update module
(1) function
All robotaries are shown in real time
(2) performance
Elapsed time<0.1 second.
(3) input
The IP of all machines and port.
(4) export
Robot location's information and ambient condition information after display fusion.
Interface design
Man-machine interface:The human-computer interaction interface made of MFC of the man-machine interface of the system.Interface is by showing football machine
Device people position and ambient condition information in place, clear and definite tell situation on user field.
Internal interface:The internal interface of the system mainly includes Data Transport Protocol, opencv functions, robot control letter
Number etc., refers to software document data.
Operating energy loss
Fig. 2 is the software block diagram of the Multi-Agent Cooperation system according to an embodiment of the invention based on private clound.
The vision data being each collected into and positional information etc. are uploaded to high in the clouds data station, subsequent cloud by five machines competed on competition field
The algorithm that end data station is transferred data in decision-making core, decision-making core will return to high in the clouds data station after Data Fusion, most
Decision information is sent to every machine by high in the clouds data station afterwards.
Fig. 3 is the high in the clouds program flow of the Multi-Agent Cooperation system according to an embodiment of the invention based on private clound
Cheng Tu.After cloud server starts, historical data, initiating communications data are loaded first;The data for waiting machine to send afterwards
Bag, receive carry out successively after packet " multimachine data fusion ", " screening decision recommendation from historical data ", " data are beaten
Bag ", " sending to decision-making core ";Decision-making core judges whether to have produced result afterwards, produce after result " being data scaling result ",
" being added to historical data ".
Fig. 4 is the decision-making core program of the Multi-Agent Cooperation system according to an embodiment of the invention based on private clound
Flow chart.After decision-making core starts, decision model, initiating communications port are loaded first;Wait what cloud server was sent afterwards
Fused data bag, receives the current situation of post analysis, generation current strategies and is contrasted with historical data;If current decision is present
In historical data, then implementation strategy, obtains implementing result, and packing data is simultaneously sent to high in the clouds;Gone through if current decision is present
It is not present, is performed if current decision success rate in historical success rates if highest in history data, if not then performs history success
Rate highest decision-making.
Fig. 5 is the robot program of the Multi-Agent Cooperation system according to an embodiment of the invention based on private clound
Flow chart.After start machine people, each hardware device is started working.1. chassis communication apparatus opening->Obtain code-disc number
According to->Resolve machine direction and speed;2. the communication of opening industrial camera->Acquisition bianry image->Extract field wire special
Levy->Robot self-localization->Extraction barrier feature->Calculate Obstacle Position;3.Kinect communications->Obtain three-dimensional point
Cloud->Extraction football feature->Analyze ball flight path.These data fusions are transmitted to cloud server afterwards.
The program operation of Multi-Agent Cooperation system according to an embodiment of the invention based on private clound can reach
, can be with stable operation in 20 frames/second to real-time.
Claims (8)
1. a kind of Multi-Agent Cooperation processing system based on private clound, it is used for medium-sized group of Robocup multiple football machines
Collaborative task between people, it is characterised in that including:
Private clound console module, for handling the real time information that the multiple Soccer robot is obtained and sent,
Robot computer vision module, for gathering scene information in real time by panoramic vision and body feeling interaction formula equipment, institute
The posture that body feeling interaction formula equipment catches the action of human body and analyzes human body is stated,
Motion planning and robot control module, for detecting the positional information of the Soccer robot and controlling the Soccer robot
Behavior,
Display content update module, the state for showing each Soccer robot in real time.
2. Multi-Agent Cooperation processing system according to claim 1, it is characterised in that:
The body feeling interaction formula equipment is Kinect.
3. Multi-Agent Cooperation processing system according to claim 1, it is characterised in that:
Color image data and depth of the privately owned cloud server of private clound console module from each Soccer robot
View data, is handled according to the color image data and depth image data that have stored in advance, is obtained in place
Relevant information, the relevant information include field wire, place, ball, barrier.
4. Multi-Agent Cooperation processing system according to claim 2, it is characterised in that
The robot computer vision module obtains color image data and by red by Kinect by colour TV camera
Outer video camera obtains depth image data, so as to obtain the color image information and the distance of each position to camera in place.
5. a kind of Multi-Agent Cooperation processing method based on private clound, it is used for medium-sized group of Robocup multiple football machines
Collaborative task between people, it is characterised in that including:
A private clound console module) is utilized, the real time information that the multiple Soccer robot is obtained and sent is handled,
B robot computer vision module) is utilized, scene information is gathered by panoramic vision and body feeling interaction formula equipment in real time,
The body feeling interaction formula equipment catches the action of human body and analyzes the posture of human body,
C motion planning and robot control module) is utilized, the positional information of the Soccer robot is detected and controls the Soccer robot
Behavior,
D display content update module) is utilized, the state of each Soccer robot is shown in real time.
6. Multi-Agent Cooperation processing method according to claim 5, it is characterised in that:
The body feeling interaction formula equipment is Kinect.
7. Multi-Agent Cooperation processing method according to claim 6, it is characterised in that the step A) include:
By the privately owned Cloud Server of private clound console module, receive color image data from each Soccer robot and
Depth image data, is handled according to the color image data and depth image data that have stored in advance, must be shown up
Relevant information in ground, the relevant information includes field wire, place, ball, barrier.
8. Multi-Agent Cooperation processing system according to claim 6, it is characterised in that the step B) include:
Color image data is obtained by colour TV camera by Kinect and depth image data is obtained by thermal camera,
So as to obtain the color image information and the distance of each position to camera in place.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710482720.XA CN107053184A (en) | 2017-06-22 | 2017-06-22 | Multi-Agent Cooperation processing system and method based on private clound |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710482720.XA CN107053184A (en) | 2017-06-22 | 2017-06-22 | Multi-Agent Cooperation processing system and method based on private clound |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107053184A true CN107053184A (en) | 2017-08-18 |
Family
ID=59613236
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710482720.XA Pending CN107053184A (en) | 2017-06-22 | 2017-06-22 | Multi-Agent Cooperation processing system and method based on private clound |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107053184A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109617968A (en) * | 2018-12-14 | 2019-04-12 | 启元世界(北京)信息技术服务有限公司 | Communication means between Multi-Agent Cooperation system and its intelligent body, intelligent body |
CN110007671A (en) * | 2019-03-12 | 2019-07-12 | 南方科技大学 | Robot cluster cooperative positioning system and method |
CN111131353A (en) * | 2018-10-31 | 2020-05-08 | 华为技术有限公司 | Collaboration method and intelligent device group |
WO2020199690A1 (en) * | 2019-03-29 | 2020-10-08 | 深圳先进技术研究院 | Cloud platform-based sharing learning system and method, sharing platform and method, and medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104835059A (en) * | 2015-04-27 | 2015-08-12 | 东华大学 | Somatosensory interaction technology-based intelligent advertisement delivery system |
CN106773768A (en) * | 2017-01-17 | 2017-05-31 | 张汇泉 | Robot wisdom service for life system |
-
2017
- 2017-06-22 CN CN201710482720.XA patent/CN107053184A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104835059A (en) * | 2015-04-27 | 2015-08-12 | 东华大学 | Somatosensory interaction technology-based intelligent advertisement delivery system |
CN106773768A (en) * | 2017-01-17 | 2017-05-31 | 张汇泉 | Robot wisdom service for life system |
Non-Patent Citations (1)
Title |
---|
王雪雁等: "《Water Team 秘密技术》", 《机器人技术与应用》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111131353A (en) * | 2018-10-31 | 2020-05-08 | 华为技术有限公司 | Collaboration method and intelligent device group |
CN109617968A (en) * | 2018-12-14 | 2019-04-12 | 启元世界(北京)信息技术服务有限公司 | Communication means between Multi-Agent Cooperation system and its intelligent body, intelligent body |
CN109617968B (en) * | 2018-12-14 | 2019-10-29 | 启元世界(北京)信息技术服务有限公司 | Communication means between Multi-Agent Cooperation system and its intelligent body, intelligent body |
CN110007671A (en) * | 2019-03-12 | 2019-07-12 | 南方科技大学 | Robot cluster cooperative positioning system and method |
WO2020199690A1 (en) * | 2019-03-29 | 2020-10-08 | 深圳先进技术研究院 | Cloud platform-based sharing learning system and method, sharing platform and method, and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107053184A (en) | Multi-Agent Cooperation processing system and method based on private clound | |
CN105407259B (en) | Virtual image capture method | |
CN106598226A (en) | UAV (Unmanned Aerial Vehicle) man-machine interaction method based on binocular vision and deep learning | |
CN109218619A (en) | Image acquiring method, device and system | |
CN104820418B (en) | A kind of embedded vision system and its application method for mechanical arm | |
CN111696216B (en) | Three-dimensional augmented reality panorama fusion method and system | |
CN105301971B (en) | A kind of method and its system controlling smart machine based on image recognition | |
US20200372715A1 (en) | Real-world object recognition for computing device | |
CN106815555B (en) | Augmented reality method and system for distributed scene target recognition | |
CN109760045B (en) | Offline programming track generation method and double-robot cooperative assembly system based on same | |
CN105183269B (en) | Method for automatically identifying screen where cursor is located | |
CN115903541A (en) | Visual algorithm simulation data set generation and verification method based on twin scene | |
Voulodimos et al. | A threefold dataset for activity and workflow recognition in complex industrial environments | |
CN109407709A (en) | A kind of meeting camera shooting automatic tracking system based on Kinect bone track algorithm | |
CN110275532B (en) | Robot control method and device and visual equipment control method and device | |
CN106131407A (en) | Shooting synchronous method and synchronizer | |
CN108972593A (en) | Control method and system under a kind of industrial robot system | |
CN106468917A (en) | A kind of tangible live real-time video image remotely assume exchange method and system | |
WO2024007485A1 (en) | Aerial-ground multi-vehicle map fusion method based on visual feature | |
WO2013025011A1 (en) | Method and system for body tracking for recognizing gestures in a space | |
CN114821408A (en) | Method, device, equipment and medium for detecting parcel position in real time based on rotating target detection | |
CN113569849B (en) | Automobile charging pile interface detection intelligent interaction system based on computer vision | |
CN108460354A (en) | Unmanned aerial vehicle (UAV) control method, apparatus, unmanned plane and system | |
CN109760067B (en) | Intelligent robot system and equipment capable of playing cards | |
CN104123008B (en) | A kind of man-machine interaction method and system based on static gesture |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170818 |