CN108171394A - Multi-robot Task Allocation based on hierachical structure and resource consolidation - Google Patents

Multi-robot Task Allocation based on hierachical structure and resource consolidation Download PDF

Info

Publication number
CN108171394A
CN108171394A CN201711168423.4A CN201711168423A CN108171394A CN 108171394 A CN108171394 A CN 108171394A CN 201711168423 A CN201711168423 A CN 201711168423A CN 108171394 A CN108171394 A CN 108171394A
Authority
CN
China
Prior art keywords
robot
manager
resource
layer
group
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
Application number
CN201711168423.4A
Other languages
Chinese (zh)
Other versions
CN108171394B (en
Inventor
曹志强
任亮
于莹莹
庞磊
谭民
喻俊志
周超
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Automation of Chinese Academy of Science
Original Assignee
Institute of Automation of Chinese Academy of Science
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Institute of Automation of Chinese Academy of Science filed Critical Institute of Automation of Chinese Academy of Science
Priority to CN201711168423.4A priority Critical patent/CN108171394B/en
Publication of CN108171394A publication Critical patent/CN108171394A/en
Application granted granted Critical
Publication of CN108171394B publication Critical patent/CN108171394B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06315—Needs-based resource requirements planning or analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Manipulator (AREA)

Abstract

The present invention relates to multi-robotic tasks to distribute field, and in particular to a kind of multi-robot Task Allocation based on hierachical structure and resource consolidation.The multi-robot Task Allocation of the present invention carries out task distribution based on pyramidal hierachical structure.When building pyramid, from bottom to top, using robot as the bottom, remaining each layer is manager, and manager's number is successively reduced, and to top layer, only there are one managers；Then, based on each robot for meeting task restriction condition, the resource that each manager possesses successively is calculated from bottom to top.In the task of distribution, each lowermost layer manager of mission requirements can be met by finding out from top to bottom, robot that is that each lowermost layer manager is directly or indirectly managed and meeting task restriction condition is combined, and selects robot alliance of the best alliance of robot of matching degree as execution task.The computational methods of the present invention are efficient, can effectively improve the real-time of task distribution.

Description

Multi-robot Task Allocation based on hierachical structure and resource consolidation
Technical field
The present invention relates to multi-robotic tasks to distribute field, and in particular to a kind of based on hierachical structure and resource consolidation Multi-robot Task Allocation.
Background technology
In recent years, multi-robot system with the features such as its flexibility, concurrency, robustness by common concern.Multirobot Systematic research includes group's architecture, task distribution, Partial global planning mechanism etc., wherein, task distribution is wherein heavy The research contents wanted.
The quality that the result of task distribution will directly influence the completion of multi-robot system task, it is suitable to assign the task to Robot go to perform, be advantageously implemented the optimization of system resource.Method for allocating tasks traditional in early days considers each task It can be completed by a robot, this is commonly available to the fairly simple situation of task.With the increase of task complexity, single machine The resource of device people will be unable to meet the needs of task, at this time, it is necessary to assign the task to the alliance of multiple robot compositions.So And it is existing based on the method for allocating tasks of alliance since computation complexity is higher, carried out in large-scale multi-robot system The real-time of task distribution need to be improved.
In this case, the validity of human society layer-stepping administrative mechanism provides a kind of resolving ideas.By mankind society The inspiration of meeting layer-stepping administrative mechanism, researcher begins to focus on and layer-stepping structure is carried out to multi-robot system, and then is dividing Task distribution is carried out under the structure of laminar.But the existing multi-robot Task Allocation based on hierachical structure is ground Study carefully the resource consolidation problem not yet efficiently solved under hierachical structure, this efficiency that will influence task distribution.It is meanwhile existing The real-time of the multi-robot Task Allocation based on hierachical structure also need to further improve.
Invention content
In order to solve the above problem of the prior art, the present invention proposes a kind of based on hierachical structure and resource consolidation Multi-robot Task Allocation, improve multi-robotic task distribution efficiency so that the real-time of distribution is had The raising of effect.
The present invention proposes a kind of multi-robot Task Allocation based on hierachical structure and resource consolidation, including following Step：
Step S10, based on pyramidal hierachical structure, according to task restriction condition and the resource of each robot Vector filters out the robot for meeting task restriction condition from the pyramidal bottom；
Step S20, according to the resource vector of the robot for meeting task restriction condition and each robot, under And the resource vector that each manager possesses in the upper each layer for calculating the pyramidal hierachical structure；
Step S30, according to the resource vector that mission requirements resource vector and each manager possess, from top to bottom, Find out each lowermost layer manager for meeting mission requirements in the pyramidal hierachical structure；
Step S40 searches what each lowermost layer manager for meeting mission requirements directly or indirectly managed, and meets The robot of task restriction condition, selection disclosure satisfy that the robot combination of mission requirements, as candidate alliance of robot, and then Alliance of composition candidate robot set；
Step S50, the best alliance of robot of selection matching degree from alliance of candidate robot set, as selected Send the alliance of robot of execution task；
Wherein,
The pyramidal hierachical structure, construction method are：
Step A10 using all robots as the pyramidal bottom, and presses preset robot group member number by machine Device people is grouped, and a manager is set respectively for each group robot；
By preset manager group member number, manager is grouped by step A20, is distinguished for each group manager One senior author is set；The senior author is grouped, and to each group by the preset manager group member number Upper level manager again is set respectively；The rest may be inferred, and until pyramidal top, only there are one managers；
The manager is computer.
Preferably, step A10 is specially：
Step A11, using the robot that total quantity is robotNum as the pyramidal bottom, the note number of plies is j=1；It is right Robot is grouped, if (robotNum%P)=0, organizes number groupNum1=robotNum/P, every group of P robot； Otherwise, group number groupNum1=robotNum/P+1, the 1st to groupNum1Every group of P robot in -1 group, the groupNum1There are robotNum- (groupNum in group1-1)*P robot；
Step A12 is that every group of robot of bottom sets a manager, as the 2nd layer of member, and by the manager Each robot in the group managed is known as the subordinate member of the manager；Manager's number is denoted as managerNum2= groupNum1；
Wherein,
P is the preset robot group member number；groupNum1For the pyramidal hierachical structure Group number after the grouping of 1 Ceng Zhong robots；managerNum2It is the number of manager in the 2nd layer.
Preferably, step A20 is specially：
Step A21, number of plies j=2；
Step A22, to jth, layer-management person is grouped, if (managerNumj%Q)=0, then number groupNum is organizedj= managerNumj/ Q, every group of Q manager；Otherwise, group number groupNumj=managerNumj/ Q+1, the 1st to groupNumjEvery group of Q manager, groupNum in -1 groupjThere is managerNum in groupj-(groupNumj-1)*Q management Person；
Step A23 is that every group of manager of jth layer sets a senior author, as+1 layer of member of jth, and will Each manager in the group that the senior author is managed is known as the subordinate member of the senior author；Senior author's number managerNumj+1=groupNumj；
Step A24, j=j+1 if j layer-management persons number is more than 1, go to step A22；Otherwise, step A25 is gone to；
Step A25, it is N=j to remember pyramidal total number of plies；
Wherein,
Q is the preset manager group member number；groupNumjIt is in the pyramidal hierachical structure Group number after j layers of manager's grouping；managerNumjThe number of manager for jth layer.
Preferably, the robot for meeting task restriction condition, for the resource vector for being in idle condition and being possessed More than or equal to the robot of task restriction resource vector.
Preferably, the comparison between resource vector is carried out by the comparison to wherein every resource, if resource vector F (a) numerical value of every resource is all higher than or the numerical value equal to corresponding resource in resource vector F (b), then F (a) >=F (b)；
Wherein, F (a) and F (b) are respectively two different resource vectors in resource space；Resource space is described with F, Its dimension is dim (F), and the resource vector in F is expressed as F=[f1f2...fdim(F)], fkRepresent the kth item resource of resource vector Numerical value,K=1,2 ..., dim (F)；The dimension of F (a) and F (b) is dim (F), F (a) and F (b) Respectively：
F (a)=[f1(a)f2(a)…fdim(F)(a)]
F (b)=[f1(b)f2(b)…fdim(F)(b)]。
Preferably, step S20 is specially：
Based on the robot for meeting task restriction condition, according to from the 2nd layer, the 3rd layer until n-th layer sequence from It implements resource integration on down, successively updates managerResource vector
Wherein,
To being located at the 2nd layer of each manager, the robot for meeting task restriction condition that is directly managed according to the manager Resource vector, integrated according to the following formula, so as to fulfill the update to manager's resource vector：
It is manager in the 2nd layerResource vector, subscript 2, g2、i2Respectively layer-management person Level number, group number, serial number；For the 1st Ceng Zhong robotsResource vector, subscript 1, g1、i1Respectively robot Level number, group number, serial number；ΩrlSet for all robots for meeting task restriction condition；
To the 3rd layer to n-th layer of each manager, according to the resource vector for the subordinate member that the manager directly manages, press It is integrated according to following formula, and then updates the resource vector of the manager：
For n-th layer managerResource vector, subscript n, gn、inThe respectively layer of layer-management person Number, group number, serial number；N=3,4 ..., N.
Preferably, the integration of resource vector to each single item resource by being integrated to complete；
To each resource, corresponding integration rules are taken according to the type of this resource；Resource type includes：Cumulative type, Maximum value type, minimum value type；The integration rules of cumulative type resource are that respective resources are carried out summation process；Maximum value type resource Integration rules are that respective resources are asked for maximum value；The integration rules of minimum value type resource are that respective resources are asked for minimum value.
Preferably, step S30 is specially：
Since the n-th layer of the pyramidal hierachical structure, the task is needed according to top-down sequence The resource vector that resource vector possesses with each manager is asked to be compared；When the resource vector that a manager possesses is more than Or equal to mission requirements resource vector, show that the manager meets mission requirements, then continue the mission requirements resource vector It is compared with the resource vector of each member of subordinate that the manager directly manages, if there are still meet under mission requirements Belong to member, repeat the comparison procedure until the 2nd layer, the 2nd layer of member of mission requirements will be met as lowermost layer manager；Such as One manager positioned at the 3rd layer or higher of fruit meets mission requirements but its each member of subordinate directly managed is unsatisfactory for appointing Business demand, and not yet there is the lowermost layer manager for meeting mission requirements, then using the manager as lowermost layer manager.
Preferably, step S40 is specially：
Step S41 judges each number of plies met residing for the lowermost layer manager of mission requirements, if in the 2 layers, then go to step S42；Otherwise, step S43 is gone to；
Step S42 finds out all robots for meeting task restriction condition that lowermost layer manager directly manages； And to all robots for meeting task restriction condition, exhaustive all possible combination, when the money that a robot combination possesses When source vector is greater than or equal to mission requirements resource vector, shows that robot combination meets mission requirements, task need will be met The robot combination asked is as alliance of candidate robot AT, it is added to candidate alliance of robot set omegarcIn；
Step S43 finds out all robots for meeting task restriction condition of lowermost layer manager institute indirect control, And by all robots for meeting task restriction condition in combination as candidate alliance of robot AT, it is added to candidate robot Alliance's set omegarcIn.
Preferably, step S50 is specially：
Based on candidate alliance of robot set omegarc, select to meet matching degree constraints
σ1≤|F(T),F(AT)|≤σ2
And the alliance of robot that matching degree is best：
Wherein, | F (T), F (AT) | for mission requirements resource vector F (T) and alliance of candidate robot ATResource vector F (AT) between matching degree, weighed with weighted euclidean distance between the two；σ1And σ2Respectively preset smallest match degree threshold Value and maximum matching degree threshold value；To be selected and appointed the best alliance of robot of execution task T.
Preferably, the matching degree computational methods between resource vector are：
F (a)=[f1(a)f2(a)…fdim(F)(a)]
F (b)=[f1(b)f2(b)…fdim(F)(b)]
Wherein, resource space described with F, dimension is dim (F), and the resource vector in F is represented by F= [f1f2...fdim(F)]；fkRepresent the numerical value of the kth item resource of resource vector,K=1,2 ..., dim (F)；F (a) and two different resource vectors that F (b) is respectively in resource space F.
Beneficial effects of the present invention：
The multi-robot Task Allocation of the present invention carries out task distribution based on pyramidal hierachical structure. When building pyramid, from bottom to top, using robot as the bottom, remaining each layer is manager, and manager's number successively subtracts Few, to top layer, only there are one managers；Then, it based on each robot for meeting task restriction condition, successively calculates from bottom to top The resource that each manager possesses.In the task of distribution, the manager that disclosure satisfy that mission requirements is successively searched from top to bottom, until The lowermost layer manager of mission requirements can be met by finding, which is directly or indirectly managed and meet task restriction item The robot of part is combined, and selects robot alliance of the best alliance of robot of matching degree as execution task.This Invention efficiently solves the problems, such as the resource consolidation under hierachical structure, and passes through the resource in hierachical structure from bottom to top It integrates and top-down resource compares, quickly determine to be selected and appointed the best alliance of robot of execution task, with the prior art It compares, there is better real-time.
Description of the drawings
Fig. 1 is the flow diagram of the multi-robot Task Allocation embodiment of the present invention；
Fig. 2 is the schematic diagram of the pyramid hierachical structure of the embodiment of the present invention.
Specific embodiment
The preferred embodiment of the present invention described with reference to the accompanying drawings.It will be apparent to a skilled person that this A little embodiments are only used for explaining the technical principle of the present invention, it is not intended that limit the scope of the invention.
Multi-robotic task distribution can be regarded as between the resource that multi-robot system possesses and mission requirements resource Matching process.We describe resource space with F, and dimension is dim (F), and the resource vector in F is represented by F= [f1f2...fdim(F)], wherein, fkRepresent the numerical value of the kth item resource of resource vector, k=1,2 ..., dim (F), fkNumerical value Ranging from
For multi-robot system, resource and the resource of mission requirements that individual robot is possessed use resource space Between resource vector in F be described, for the unfavorable robot of performance is avoided to be selected, it is also contemplated that task is to individual machine The resource constraint of people is characterized with task restriction resource vector.
When individual robot does not have a certain resource, the numerical value of the corresponding resource items is 0 in resource vector；When appoint When not needing to a certain resource in business demand, the numerical value of the corresponding resource items is 0 in mission requirements resource vector；When task not When constraining a certain resource of individual robot, the numerical value of the corresponding resource items is 0 in task restriction resource vector. Robot resource vector, mission requirements resource vector, task restriction resource vector all have identical dimension, and in each dimension Correspond to identical resource.
It can be integrated between resource vector, the integration between resource vector is by being integrated to each single item resource It is completed.Each single item resource has corresponding integration rules according to resource type.Resource type has cumulative type resource, maximum value Type resource, minimum value type resource.Respective resources are exactly carried out summation process by the integration rules of cumulative type resource；Maximum value type provides Respective resources are exactly asked for maximum value by the integration rules in source；The integration rules of minimum value type resource exactly ask for respective resources Minimum value.Using dimension as 3 resource vector F (as) and F (bs) for, the two resource vectors are respectively such as formula (1), (2) institute Show：
F(as)=[f1(as)f2(as)f3(as)] (1)
F(bs)=[f1(bs)f2(bs)f3(bs)] (2)
The the 1st, the 2nd and the 3rd resource in resource vector is respectively cumulative type resource, maximum value type resource and minimum value type Resource, then F (as) and F (bs) integration, i.e. F (as)+F(bs), as shown in formula (3)：
F(as)+F(bs)=[f1(as)+f1(bs)max(f2(as),f2(bs))min(f3(as),f3(bs))] (3)
Wherein, max (f2(as),f2(bs)) to ask for f2(as) and f2(bs) maximum value, min (f3(as),f3(bs)) be Ask for f3(as) and f3(bs) minimum value.
It can also be compared between resource vector, note F (a) and F (b) they are resource vector of the dimension for dim (F), this Two resource vectors are respectively as shown in formula (4), (5)：
F (a)=[f1(a)f2(a)…fdim(F)(a)] (4)
F (b)=[f1(b)f2(b)…fdim(F)(b)] (5)
When the numerical value of every resource of F (a) be all higher than or equal to F (b) in corresponding resource numerical value, just have F (a) >=F (b), as shown in formula (6)：
Matching degree between resource vector can use weighted euclidean distance | F (a), F (b) | it is indicated, computational methods As shown in formula (7)：
The embodiment of the multi-robot Task Allocation based on hierachical structure and resource consolidation of the present invention, such as Fig. 1 It is shown, include the following steps：
Step S10, based on pyramidal hierachical structure, according to task restriction condition and the resource of each robot Vector filters out the robot for meeting task restriction condition from the pyramidal bottom；
Step S20, according to the resource vector of the robot for meeting task restriction condition and each robot, under And the resource vector that each manager possesses in the upper each layer for calculating the pyramidal hierachical structure；
Step S30, according to the resource vector that mission requirements resource vector and each manager possess, from top to bottom, Find out each lowermost layer manager for meeting mission requirements in the pyramidal hierachical structure；
Step S40 searches what each lowermost layer manager for meeting mission requirements directly or indirectly managed, and meets The robot of task restriction condition, selection disclosure satisfy that the robot combination of mission requirements, as candidate alliance of robot, and then Alliance of composition candidate robot set；
Step S50, the best alliance of robot of selection matching degree from alliance of candidate robot set, as selected Send the alliance of robot of execution task；
Wherein：
Manager described above is computer, and in hierachical structure, what manager undertook is storage, calculating, communication Function.According to actual deployment needs, each manager can be integrated on same computer, can also use different meters respectively Calculation machine is integrated into wherein certain several manager on same computer.
Pyramidal hierachical structure described above, construction method are：
In step A10, using all robots as the pyramidal bottom, and by preset robot group member number Robot is grouped, one manager is set respectively for each group robot.
Subordinate member of the every group of robot as corresponding manager, each manager have each robot for oneself being managed Serial number；These managers also subordinate member as high layer-management person simultaneously.
In step A20, by preset manager group member number, manager is grouped, for each group manager One senior author is set respectively；The senior author is grouped by the preset manager group member number, And upper level manager again is set respectively to each group；The rest may be inferred, and until pyramidal top, only there are one managers.
This manager of top is exactly chief executive, so far, completes the layer-stepping knot of multi-robot system The structure of structure.Here the hierachical structure of the multi-robot system built is determined in itself by multi-robot system, not because appointing The difference of business and change.Therefore, this structure can complete structure in advance before task distribution, as long as multi-robot system The resource vector of constant, each robot of number of middle robot is constant, and in the task of distribution, there is no need to rebuild this gold The hierachical structure of word turriform.
Specifically, step A10 can be in the present embodiment：
Step A11, using the robot that total quantity is robotNum as the pyramidal bottom, the note number of plies is j=1；It is right Robot is grouped, if (robotNum%P)=0, organizes number groupNum1=robotNum/P, every group of P robot； Otherwise, group number groupNum1=robotNum/P+1, the 1st to groupNum1Every group of P robot in -1 group, the groupNum1There are robotNum- (groupNum in group1-1)*P robot；
Step A12 is that every group of robot of bottom sets a manager, as the 2nd layer of member, and by the manager Each robot in the group managed is known as the subordinate member of the manager；Manager's number is denoted as managerNum2= groupNum1；
Wherein：
P is the preset robot group member number, i.e., after being grouped to bottom robot, each group of robot Interior robot number, but P may be less than in last group；groupNum1The 1st for pyramid hierachical structure Group number after the grouping of Ceng Zhong robots；managerNum2It is the number of manager in the 2nd layer.
The thinking of previous step A11 to A12 is：
Using each robot in multi-robot system as the member of the bottom of hierachical structure, according to P machine The standard that one group of people is grouped, if last remaining less than P robot is assigned to, using these robots independently as one group； For every group of robot, a manager is set, by the corresponding robot of the manager be known as the subordinate of the manager into Member.
Specifically, step A20 can be in the present embodiment：
Step A21, number of plies j=2；
Step A22, to jth, layer-management person is grouped, if (managerNumj%Q)=0, then number groupNum is organizedj= managerNumj/ Q, every group of Q manager；Otherwise, group number groupNumj=managerNumj/ Q+1, the 1st to groupNumjEvery group of Q manager, groupNum in -1 groupjThere is managerNum in groupj-(groupNumj-1)*Q management Person；
Step A23 is that every group of manager of jth layer sets a senior author, as+1 layer of member of jth, and will Each manager in the group that the senior author is managed is known as the subordinate member of the senior author；Senior author's number managerNumj+1=groupNumj；
Step A24, j=j+1 if j layer-management persons number is more than 1, go to step A22；Otherwise, step A25 is gone to；
Step A25, it is N=j to remember pyramidal total number of plies；
Wherein：
Q is the preset manager group member number, i.e., after being grouped to the manager of same layer, each manager Membership in group, but Q may be less than in last group；groupNumjFor pyramidal hierachical structure Group number after manager's grouping of middle jth layer；managerNumjThe number of manager for jth layer.
The thinking of previous step A21 to A25 is：
The manager of same layer is continued to be grouped according to Q one group of standard, if assigning to last remaining manager less than Q It is a, then using these managers independently as one group；To one senior author of each group of setting, each manager is corresponding The subordinate member of senior author；These senior authors are stored with the serial number of oneself subordinate member；These managers are also simultaneously Subordinate member as higher level manager.Cycle performs grouping, setting manager, until only there are one managers for a certain layer Until, which is designated as chief executive.
As shown in Fig. 2, 37 robots in total, according to P=5, Q=2 has been built into the pyramidal layer-stepping of N=5 Structure.5th layer has 1 chief executive M5,1,1；4th layer has 2 managers, respectively M4,1,1、M4,1,2；3rd layer has 4 Manager, respectively M3,1,1、M3,1,2、M3,2,3、M3,2,4；2nd layer has 8 managers, respectively M2,1,1、M2,1,2、M2,2,3、 M2,2,4、M2,3,5、M2,3,6、M2,4,7、M2,4,8；1st Ceng You37Ge robots are divided into 8 groups, before respectively have 5 machines in 7 groups People, last group have 2 robots.
Each layer is respectively to the hierachical structure of multi-robot system from top to bottom it can be seen from construction method above N layers, N-1 layers ..., the 1st layer.All robots are located at the bottom i.e. the 1st layer, and the member of other each layers is in addition to the 1st layer Manager, chief executive are located at n-th layer.Robot in hierachical structure from can be between manager, different managers It is transmitted by WLAN etc. into row information.We useDescription is located at the 1st layer of robot, wherein, subscript 1, g1With i1Level number, group number and serial number, the resource resource vector that each robot is possessed are represented respectivelyIt represents, i1=1, 2 ..., robotNum, robotNum be robot total quantity；
One Boolean type variable of the state of robotIt represents, when robot is carrying out task or failure WhenWhen robot being capable of normal operation and during current idleSince the 2nd layer, until most High-rise manager usesIt is described, wherein subscript n, gnAnd inLevel number, group number and serial number, the value pair of n are represented respectively It should be 2,3 ..., N, managerThe resource resource vector possessedIt represents.
Specifically, meet the robot of task restriction condition in the present embodiment described in step S10, can be in idle shape State and the resource vector possessed are greater than or equal to the robot of task restriction resource vector.
For task T, noteFor task restriction resource vector, when individual robotIt is in idle condition i.e.And meetWhen, show that the robot meets task restriction condition, that is, be eligible to participate in this Subtask is distributed, and all robots for being eligible to participate in the distribution of this subtask form set omegarl。
Specifically, step S20 can be in the present embodiment：
Based on set omegarlIn robot, according to from the 2nd layer, the 3rd layer until the sequence of n-th layer carries out from bottom to top Resource consolidation successively updates managerResource vectorThe value of n is respectively 2,3 ..., N, and method is such as Under：
The money for the robot for being eligible to participate in the distribution of this subtask directly managed according to oneself positioned at the 2nd layer of manager Source vector is integrated according to formula (8), so as to fulfill the update of oneself resource vector：
The 3rd layer of resource vector of subordinate member directly managed according to oneself up to the manager of n-th layer, according to formula (9) it is integrated, so as to fulfill the update of oneself resource vector：
Specifically, step S30 can be in the present embodiment：
Since the n-th layer of the pyramidal hierachical structure, the task is needed according to top-down sequence The resource vector that resource vector possesses with each manager is asked to be compared；When a manager meets mission requirements, i.e., should The resource vector that manager possesses be greater than or equal to mission requirements resource vector, then continue by the mission requirements resource vector with The resource vector of each member of subordinate that the manager directly manages is compared, if there are still the subordinaties for meeting mission requirements Member repeats the comparison procedure until the 2nd layer, will meet the 2nd layer of member of mission requirements as lowermost layer manager；If One manager positioned at the 3rd layer or higher meets mission requirements but its each member of subordinate directly managed is unsatisfactory for task Demand, and not yet there is the lowermost layer manager for meeting mission requirements, then using the manager as lowermost layer manager.
It is exemplified below and how to search the lowermost layer manager for meeting mission requirements：
What a assuming that pyramidal hierachical structure as shown in Figure 2 had been built up, and completed in each layer The resource consolidation of each manager.
We describe mission requirements resource vector with F (T), when pyramidal hierachical structure is searched from top to down, are One tree-like branched structure, search meet the lowermost layer manager of mission requirements when, need to traverse each and may meet and appoint The branch of business demand.
Example 1：
(1) if F (M5,1,1)≥F(T)；
(2) M is checked4,1,1, find F (M4,1,1)≥F(T)；
(3) M is checked3,1,1, find F (M3,1,1) >=F (T), and F (M2,1,1)≥F(T)、F(M2,1,2)<F (T), then M2,1,1 The lowermost layer manager that be exactly its respective branches one meets mission requirements；
(4) M is checked3,1,2, find F (M3,1,2) >=F (T), and F (M2,2,3)≥F(T)、F(M2,2,4) >=F (T), then M2,2,3、 M2,2,4It is a lowermost layer manager for meeting mission requirements of its respective branches respectively；
(5) M is checked4,1,2, find F (M4,1,2)≥F(T)；
(6) M is checked3,2,3, but F (M3,2,3)<F(T)；
(7) M is checked3,2,4, but F (M3,2,4)<F(T)；
So far, example 1 has looked into the entire hierachical structure that is over, and finds out 3 lowermost layer pipes for meeting mission requirements altogether Reason person：M2,1,1、M2,2,3、M2,2,4。
Example 2：
(1) if F (M5,1,1)≥F(T)；
(2) M is checked4,1,1, find F (M4,1,1)≥F(T)；
(3) M is checked3,1,1, but F (M3,1,1)<F(T)；
(4) M is checked3,1,2, but F (M3,1,2)<F(T)；
(5) meet the minimum of mission requirements because up to the present not finding any one also in entire hierachical structure Layer-management person, so we are by M4,1,1One as its respective branches lowermost layer manager for meeting mission requirements；
(6) M is checked4,1,2, find F (M4,1,2)≥F(T)；
(7) M is checked3,2,3, find F (M3,2,3) >=F (T), and F (M2,3,5)≥F(T)、F(M2,3,6) >=F (T), then M2,3,5、 M2,3,6It is the lowermost layer manager for meeting mission requirements of its respective branches respectively；
(8) M is checked3,2,4, find F (M3,2,4) >=F (T), and F (M2,4,7)≥F(T)、F(M2,4,8)<F (T), then M2,4,7It is The lowermost layer manager for meeting mission requirements of its respective branches；
So far, example 2 has looked into the entire hierachical structure that is over, and finds out 4 lowermost layer pipes for meeting mission requirements altogether Reason person：M4,1,1、M2,3,5、M2,3,6、M2,4,7。
Specifically, step S40 can be in the present embodiment：
Step S41 judges each number of plies met residing for the lowermost layer manager of mission requirements, if in the 2 layers, then go to step S42；Otherwise, step S43 is gone to；
Step S42 finds out all robots for meeting task restriction condition that lowermost layer manager directly manages； And to all robots for meeting task restriction condition, exhaustive all possible combination, when the money that a robot combination possesses When source vector is greater than or equal to mission requirements resource vector, shows that robot combination meets mission requirements, this is met into task The robot combination of demand is as alliance of candidate robot AT, it is added to candidate alliance of robot set omegarcIn；
Step S43 finds out all robots for meeting task restriction condition of lowermost layer manager institute indirect control, And by all robots for meeting task restriction condition in combination as candidate alliance of robot AT, it is added to candidate robot Alliance's set omegarcIn.
Explanation " directly managing " and the concept of " indirect control " below：
In the hierachical structure for being N in a total number of plies, for the manager of n-th (2≤n≤N) layer, have with it and be subordinate to pass (n-1)th layer of member of system, is exactly the subordinate member that it is directly managed；The n-th -2 layers, the n-th -3 layers ..., have with it in the 1st layer and be subordinate to The member of relationship is the subordinate member of its indirect control.Wherein, the manager for the 2nd layer, the subordinate only directly managed Member, i.e. robot, without the subordinate member of indirect control.
For example, for the 2nd layer in Fig. 2 of manager M2,1,1For, there is the bottom robot R of membership with it1,1,1、 R1,1,2、…、R1,1,5It is exactly the subordinate member that it is directly managed；In another example for the 4th layer in Fig. 2 of manager M4,1,2For, There is the manager M of membership in 3rd layer with it3,2,3And M3,2,4The subordinate member that it is directly managed, and in the 2nd layer with it There is the member M of membership2,3,5、M2,3,6、M2,4,7And M2,4,8It is by its indirect control, has membership with it in the 1st layer Member R1,5,21、R1,5,22、…、R1,8,37(bottom is from R namely in figure1,5,21Start to turn right until R1,8,37All machines People) it is also by its indirect control.
The thinking of previous step S30, S41 to S43 is：
We describe mission requirements resource vector with F (T), when a manager meetsMeaning It the manager and possesses sufficient resource, disclosure satisfy that mission requirements；Since hierachical structure n-th layer, according to from top to bottom Sequence F (T) resource vectors possessed with each layer-management person are compared, when a manager meets mission requirements, continue F (T) is compared with the resource vector of each member of subordinate of the manager, if there are still meet the subordinaties of mission requirements into Member repeats above-mentioned comparison procedure until the 2nd layer, for each the 2nd layer of manager for meeting mission requirements, with reference to oneself Subordinate is eligible to participate in each robot of this subtask distribution, exhaustive all possible robot combination, for each machine People combines Cr, its resource vector F (C are calculated using formula (10)r)：
If F (CrCorresponding robot is then combined C by) >=F (T)rAs candidate alliance of robot, A is usedTIt is described, By ATIt is added to candidate alliance of robot set omegarcIn.
If a manager positioned at the 3rd layer or higher meets mission requirements but each member of its subordinate is unsatisfactory for appointing Business demand, in ΩrcIn the case of for empty set, the member with membership is successively searched downwards since the manager up to the 1 layer, all robots for meeting task restriction condition that the 1st layer is related to are in combination as candidate alliance of robot AT, It is added to candidate alliance of robot set omegarcIn.
Specifically, step S50 can be in the present embodiment：
Based on candidate alliance of robot set omegarc, select both to have met the matching degree constraints shown in formula (11)：
σ1≤|F(T),F(AT)|≤σ2 (11)
Meet the best alliance of robot of the matching degree of formula (12) again：
Wherein, | F (T), F (AT) | for mission requirements resource vector F (T) and alliance of candidate robot ATResource vector F (AT) between matching degree, weighed with weighted euclidean distance between the two；σ1And σ2Respectively preset smallest match degree threshold Value and maximum matching degree threshold value；To be selected and appointed the best alliance of robot of execution task T.
Specifically, the matching degree computational methods in the present embodiment between resource vector, as shown in formula (13)：
Wherein：
F (a) and F (b) resource vector that be respectively two different, as shown in formula (14), (15)：
F (a)=[f1(a)f2(a)…fdim(F)(a)] (14)
F (b)=[f1(b)f2(b)…fdim(F)(b)] (15)
Dim (F) is the dimension of F (a) and F (b)；fkRepresent the numerical value of the kth item resource of resource vector,K=1,2 ..., dim (F)；
Below for a specific example, come further illustrate multi-robotic task distribution method：
Such as in multi-robot system, one shares 10 robots, takes P=5,10 robots is divided into 2 groups, often Group sets 1 manager, then the 2nd layer of a total of 2 manager；To this 2 managers, Q=2 is taken, i.e., is managed according to every group 2 The standard of person is grouped, and can only divide 1 group, then sets 1 senior author；This 1 senior author is highest layer-management Person.
So far, structure completes one 3 layers of pyramid structure, i.e. N=3.1st layer by being equally divided into 2 groups 10 A robot is formed, wherein, the 1st group of 5 robots use R respectively1,1,1、R1,1,2、R1,1,3、R1,1,4、R1,1,5Description, the 2nd group 5 robots use R respectively1,2,6、R1,2,7、R1,2,8、R1,2,9、R1,2,10It is described；2nd layer is made of 2 managers, is used M2,1,1And M2,1,2It is described, manages the 1st group of 5 robots and the 2nd group of 5 robots in the 1st layer respectively；3rd layer by One manager is formed, and uses M3,1,1It is described, the M of the 2nd layer of management2,1,1And M2,1,2。
This 3 managers of layers 2 and 3 are fully integrated realizes all storages, calculating, communication on one computer Function.WLAN is built based on Huawei honor router Pro (WS851).The dimension of the resource vector of each robot Number for 4, respectively with locomitivity resource (unit：Meter per second), visual sensor resolution ratio resource (unit：Pixel), laser sensing Device detection angle resource (unit：Degree), laser sensor measurement distance resource (unit：Rice) it is corresponding, this 4 resources are most Big value type resource.The maximum value and minimum value of the numerical value of this 4 resources are given below respectively：
Assuming that this 10 robots all can normal operation and all in idle state, specific resource vector is respectively such as Shown in formula (16)-(25)：
F(R1,1,1)=[0.5640 × 48027010] (16)
F(R1,1,2)=[0.71280 × 72027010] (17)
F(R1,1,3)=[0.61280 × 72027010] (18)
F(R1,1,4)=[0.61280 × 7203608] (19)
F(R1,1,5)=[0.61280 × 72027020] (20)
F(R1,2,6)=[0.7640 × 4803606] (21)
F(R1,2,7)=[0.51280 × 7203606] (22)
F(R1,2,8)=[0.51280 × 7203608] (23)
F(R1,2,9)=[0.51920 × 108027030] (24)
F(R1,2,10)=[0.61920 × 108027030] (25)
For task T, mission requirements resource vector such as formula (26) is shown：
F (T)=[0.6 1280 × 720 270 10] (26)
Shown in task restriction resource vector such as formula (27)：
Multi-robot Task Allocation according to the present invention first screens individual robot, determines qualified There is R in the robot for participating in the distribution of this subtask1,1,2、R1,1,3、R1,1,4、R1,1,5And R1,2,10, these robots, which are formed, participates in this The collection of bots Ω of task distributionrl；It implements resource integration according to the sequence from the 2nd layer to the 3rd layer, successively updates from bottom to top Manager M2,1,1、M2,1,2、M3,1,1Resource vector, shown in updated resource vector such as formula (28)-(30)：
F(M2,1,1)=[0.7 1280 × 720 360 20] (28)
F(M2,1,2)=[0.6 1920 × 1,080 270 30] (29)
F(M3,1,1)=[0.7 1920 × 1,080 360 30] (30)
Then successively the resource vector that mission requirements resource vector F (T) and each layer-management person possess is carried out from top to bottom Compare, due to F (M3,1,1) >=F (T) continues F (T) with M3,1,1Subordinate member M2,1,1、M2,1,2Resource vector compared Compared with due to F (M2,1,1) >=F (T), F (M2,1,2) >=F (M), therefore respectively from M2,1,1And M2,1,2Management is eligible to participate in this In the subordinate robot of task distribution, exhaustive all possible robot combination combines C for each robotr, calculate it Resource vector F (Cr), if F (Cr) >=F (T) is set up, then corresponding robot is combined CrAs candidate alliance of robot, use ATIt is described, by ATIt is added to candidate alliance of robot set omegarcIn；On this basis, according toAnd combine matching degree constraints σ1≤|F(T),F(AT)|≤σ2, In, σ1=0.5, σ2=1, obtain being selected and appointed best alliance of the robot { R of execution task T1,1,2,R1,1,4,R1,1,5}。
The present invention is compared by the resource consolidation in hierachical structure from bottom to top and top-down resource, can be quick Determine to be selected and appointed the best alliance of robot of execution task, real-time is good, for multi-robot system task distribution etc. Using offer technical support.
Those skilled in the art should be able to recognize that, each exemplary side described with reference to the embodiments described herein Method step can realize with the combination of electronic hardware, computer software or the two, in order to clearly demonstrate electronic hardware and The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These Function is performed actually with electronic hardware or software mode, specific application and design constraint depending on technical solution. Those skilled in the art can realize described function to each specific application using distinct methods, but this reality Now it is not considered that beyond the scope of this invention.
So far, it has been combined preferred embodiment shown in the drawings and describes technical scheme of the present invention, still, this field Technical staff is it is easily understood that protection scope of the present invention is expressly not limited to these specific embodiments.Without departing from this Under the premise of the principle of invention, those skilled in the art can make the relevant technologies feature equivalent change or replacement, these Technical solution after changing or replacing it is fallen within protection scope of the present invention.

Claims (11)

1. a kind of multi-robot Task Allocation based on hierachical structure and resource consolidation, which is characterized in that including following Step：
Step S10, based on pyramidal hierachical structure, according to task restriction condition and the resource vector of each robot, The robot for meeting task restriction condition is filtered out from the pyramidal bottom；
Step S20, according to the resource vector of the robot for meeting task restriction condition and each robot, from bottom to top Calculate the resource vector that each manager possesses in each layer of the pyramidal hierachical structure；
Step S30 according to the resource vector that mission requirements resource vector and each manager possess, from top to bottom, is searched Go out in the pyramidal hierachical structure to meet each lowermost layer manager of mission requirements；
Step S40 searches what each lowermost layer manager for meeting mission requirements directly or indirectly managed, and meets task The robot of constraints, selection disclosure satisfy that the robot combination of mission requirements, as candidate alliance of robot, and then form Candidate alliance of robot set；
Step S50, the best alliance of robot of selection matching degree, holds as being selected and appointed from alliance of candidate robot set The alliance of robot of row task；
Wherein,
The pyramidal hierachical structure, construction method are：
Step A10 using all robots as the pyramidal bottom, and presses preset robot group member number by robot It is grouped, one manager is set respectively for each group robot；
By preset manager group member number, manager is grouped, is set respectively for each group manager by step A20 One senior author；The senior author is grouped by the preset manager group member number, and each group is distinguished Upper level manager again is set；The rest may be inferred, and until pyramidal top, only there are one managers；
The manager is computer.
2. multi-robot Task Allocation according to claim 1, which is characterized in that step A10 is specially：
Step A11, using the robot that total quantity is robotNum as the pyramidal bottom, the note number of plies is j=1；To machine People is grouped, if (robotNum%P)=0, organizes number groupNum1=robotNum/P, every group of P robot；Otherwise, Group number groupNum1=robotNum/P+1, the 1st to groupNum1Every group of P robot, groupNum in -1 group1Group In have robotNum- (groupNum1- 1) * P robots；
Step A12 is that every group of robot of bottom sets a manager, is managed as the 2nd layer of member, and by the manager Each robot in the group of reason is known as the subordinate member of the manager；Manager's number is denoted as managerNum2=groupNum1；
Wherein,
P is the preset robot group member number；groupNum1For in the 1st layer of the pyramidal hierachical structure Group number after robot grouping；managerNum2It is the number of manager in the 2nd layer.
3. multi-robot Task Allocation according to claim 2, which is characterized in that step A20 is specially：
Step A21, number of plies j=2；
Step A22, to jth, layer-management person is grouped, if (managerNumj%Q)=0, then number groupNum is organizedj= managerNumj/ Q, every group of Q manager；Otherwise, group number groupNumj=managerNumj/ Q+1, the 1st to groupNumjEvery group of Q manager, groupNum in -1 groupjThere is managerNum in groupj-(groupNumj- 1) * Q are managed Reason person；
Step A23 is that every group of manager of jth layer sets a senior author, as+1 layer of member of jth, and will be on this Each manager in group that grade manager is managed is known as the subordinate member of the senior author；Senior author's number managerNumj+1=groupNumj；
Step A24, j=j+1 if j layer-management persons number is more than 1, go to step A22；Otherwise, step A25 is gone to；
Step A25, it is N=j to remember pyramidal total number of plies；
Wherein,
Q is the preset manager group member number；groupNumjFor jth layer in the pyramidal hierachical structure Group number after manager's grouping；managerNumjThe number of manager for jth layer.
4. multi-robot Task Allocation according to claim 3, which is characterized in that described to meet task restriction condition Robot, the robot of task restriction resource vector is greater than or equal to for the resource vector that is in idle condition and is possessed.
5. multi-robot Task Allocation according to claim 4, which is characterized in that the comparison between resource vector, It is carried out by the comparison to wherein every resource, if the numerical value of every resource of resource vector F (a) is all higher than or equal to resource The numerical value of corresponding resource in vectorial F (b), then F (a) >=F (b)；
Wherein, F (a) and F (b) are respectively two different resource vectors in resource space；Resource space is described with F, is tieed up It is dim (F) to count, and the resource vector in F is expressed as F=[f1 f2...fdim(F)], fkRepresent the number of the kth item resource of resource vector Value,K=1,2 ..., dim (F)；The dimension of F (a) and F (b) is dim (F), F (a) and F (b) difference For：
F (a)=[f1(a) f2(a)…fdim(F)(a)]
F (b)=[f1(b) f2(b)…fdim(F)(b)]。
6. multi-robot Task Allocation according to claim 5, which is characterized in that step S20 is specially：
Based on the robot for meeting task restriction condition, according to from the 2nd layer, the 3rd layer until n-th layer sequence from lower and On implement resource integration, successively update managerResource vector
Wherein,
To being located at the 2nd layer of each manager, the money for the robot for meeting task restriction condition that is directly managed according to the manager Source vector is integrated according to the following formula, so as to fulfill the update to manager's resource vector：
It is manager in the 2nd layerResource vector, subscript 2, g2、i2The respectively layer of layer-management person Number, group number, serial number；For the 1st Ceng Zhong robotsResource vector, subscript 1, g1、i1Respectively robot Level number, group number, serial number；ΩrlSet for all robots for meeting task restriction condition；
To the 3rd layer to n-th layer of each manager, according to the resource vector for the subordinate member that the manager directly manages, under Formula is integrated, and then updates the resource vector of the manager：
For n-th layer managerResource vector, subscript n, gn、inRespectively the level number of layer-management person, Group number, serial number；N=3,4 ..., N.
7. multi-robot Task Allocation according to claim 6, which is characterized in that the integration of resource vector by pair Each single item resource is integrated to complete；
To each resource, corresponding integration rules are taken according to the type of this resource；Resource type includes：Cumulative type, maximum Value type, minimum value type；The integration rules of cumulative type resource are that respective resources are carried out summation process；The integration of maximum value type resource Rule is that respective resources are asked for maximum value；The integration rules of minimum value type resource are that respective resources are asked for minimum value.
8. multi-robot Task Allocation according to claim 7, which is characterized in that step S30 is specially：
Since the n-th layer of the pyramidal hierachical structure, the mission requirements are provided according to top-down sequence The resource vector that source vector possesses with each manager is compared；When the resource vector that a manager possesses is more than or waits In mission requirements resource vector, show that the manager meets mission requirements, then continue the mission requirements resource vector with this The resource vector of each member of subordinate that manager directly manages is compared, if there are still meet the subordinaties of mission requirements into Member repeats the comparison procedure until the 2nd layer, will meet the 2nd layer of member of mission requirements as lowermost layer manager；If one A manager for being located at the 3rd layer or higher meets mission requirements but its each member of subordinate directly managed is unsatisfactory for task need It asks, and the lowermost layer manager for meeting mission requirements not yet occurs, then using the manager as lowermost layer manager.
9. multi-robot Task Allocation according to claim 8, which is characterized in that step S40 is specially：
Step S41 judges each number of plies met residing for the lowermost layer manager of mission requirements, if being in the 2nd layer, Then go to step S42；Otherwise, step S43 is gone to；
Step S42 finds out all robots for meeting task restriction condition that lowermost layer manager directly manages；It is and right All robots for meeting task restriction condition, exhaustive all possible combination, when the resource that a robot combination possesses to When amount is greater than or equal to mission requirements resource vector, shows that robot combination meets mission requirements, mission requirements will be met Robot combination is as alliance of candidate robot AT, it is added to candidate alliance of robot set omegarcIn；
Step S43 finds out all robots for meeting task restriction condition of lowermost layer manager institute indirect control, and will All robots for meeting task restriction condition are in combination as candidate alliance of robot AT, it is added to candidate alliance of robot Set omegarcIn.
10. multi-robot Task Allocation according to claim 9, which is characterized in that step S50 is specially：
Based on candidate alliance of robot set omegarc, select to meet matching degree constraints
σ1≤|F(T),F(AT)|≤σ2
And the alliance of robot that matching degree is best：
Wherein, | F (T), F (AT) | for mission requirements resource vector F (T) and alliance of candidate robot ATResource vector F (AT) it Between matching degree, weighed with weighted euclidean distance between the two；σ1And σ2Respectively preset smallest match degree threshold value and most Big matching degree threshold value；To be selected and appointed the best alliance of robot of execution task T.
11. according to the multi-robot Task Allocation described in any one of claim 1-10, which is characterized in that resource vector Between matching degree computational methods be：
F (a)=[f1(a) f2(a)…fdim(F)(a)]
F (b)=[f1(b) f2(b)…fdim(F)(b)]
Wherein, resource space described with F, dimension is dim (F), and the resource vector in F is represented by F=[f1 f2...fdim(F)]；fkRepresent the numerical value of the kth item resource of resource vector,K=1,2 ..., dim (F)；F (a) with F (b) it is respectively two different resource vectors in resource space F.
CN201711168423.4A 2017-11-21 2017-11-21 Multi-robot task allocation method based on layered structure and resource integration Active CN108171394B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711168423.4A CN108171394B (en) 2017-11-21 2017-11-21 Multi-robot task allocation method based on layered structure and resource integration

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711168423.4A CN108171394B (en) 2017-11-21 2017-11-21 Multi-robot task allocation method based on layered structure and resource integration

Publications (2)

Publication Number Publication Date
CN108171394A true CN108171394A (en) 2018-06-15
CN108171394B CN108171394B (en) 2021-02-02

Family

ID=62527177

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711168423.4A Active CN108171394B (en) 2017-11-21 2017-11-21 Multi-robot task allocation method based on layered structure and resource integration

Country Status (1)

Country Link
CN (1) CN108171394B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110070289A (en) * 2019-04-19 2019-07-30 苏州达家迎信息技术有限公司 Method for allocating tasks, device, equipment and storage medium
CN112859887A (en) * 2019-11-28 2021-05-28 中国科学院沈阳自动化研究所 Multi-underwater robot autonomous task allocation method based on space-based center

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100145514A1 (en) * 2008-12-08 2010-06-10 Electronics And Telecommunications Research Institute Apparatus and method for controlling multi-robot linked in virtual space
CN101945492A (en) * 2010-08-09 2011-01-12 哈尔滨工程大学 Clustering-based multi-robot task allocation method
CN102033536A (en) * 2010-12-22 2011-04-27 大连理工大学 Scheduling, organization and cooperation system and method for multi-robot system
CN102831319A (en) * 2012-08-27 2012-12-19 北京科技大学 Task allocation algorithm of fine-grained multi-robot collaboration system
CN103064743A (en) * 2012-12-27 2013-04-24 深圳先进技术研究院 Resource scheduling method for multiple robots and resource scheduling system thereof

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100145514A1 (en) * 2008-12-08 2010-06-10 Electronics And Telecommunications Research Institute Apparatus and method for controlling multi-robot linked in virtual space
CN101945492A (en) * 2010-08-09 2011-01-12 哈尔滨工程大学 Clustering-based multi-robot task allocation method
CN102033536A (en) * 2010-12-22 2011-04-27 大连理工大学 Scheduling, organization and cooperation system and method for multi-robot system
CN102831319A (en) * 2012-08-27 2012-12-19 北京科技大学 Task allocation algorithm of fine-grained multi-robot collaboration system
CN103064743A (en) * 2012-12-27 2013-04-24 深圳先进技术研究院 Resource scheduling method for multiple robots and resource scheduling system thereof

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
JIAN CHEN等: "Resource Constrained Multirobot Task Allocation with A Leader-Follower Coalition Method", 《THE 2010 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS》 *
周菁: "自组织多机器人系统任务分配建模与设计", 《中国博士学位论文全文数据库(电子期刊)信息科技辑》 *
张嵛等: "多机器人任务分配的研究与进展", 《智能系统学报》 *
苏丽颖等: "多机器人系统任务分配问题的建模与求解", 《中南大学学报（自然科学版）》 *
黎萍等: "多机器人系统任务分配的研究进展", 《计算机工程与应用》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110070289A (en) * 2019-04-19 2019-07-30 苏州达家迎信息技术有限公司 Method for allocating tasks, device, equipment and storage medium
CN112859887A (en) * 2019-11-28 2021-05-28 中国科学院沈阳自动化研究所 Multi-underwater robot autonomous task allocation method based on space-based center
CN112859887B (en) * 2019-11-28 2022-06-14 中国科学院沈阳自动化研究所 Multi-underwater robot autonomous task allocation method based on space-based center

Also Published As

Publication number Publication date
CN108171394B (en) 2021-02-02

Similar Documents

Publication Publication Date Title
CN104461740B (en) A kind of cross-domain PC cluster resource polymerization and the method for distribution
CN106875090A (en) A kind of multirobot distributed task scheduling towards dynamic task distributes forming method
CN104391918B (en) The implementation method of distributed networks database query priority management based on equity deployment
CN109491790A (en) Industrial Internet of Things edge calculations resource allocation methods and system based on container
CN108009016A (en) A kind of balancing resource load control method and colony dispatching device
Sourirajan et al. Hybrid decomposition heuristics for solving large-scale scheduling problems in semiconductor wafer fabrication
Yazdani et al. Evolutionary algorithms for multi-objective dual-resource constrained flexible job-shop scheduling problem
CN104461748B (en) A kind of optimal localization tasks dispatching method based on MapReduce
CN110086650A (en) The cloud resource on-line scheduling method and device of Based on Distributed machine learning task
CN104657795A (en) To-be-observed task determination method and device of multi-satellite earth synergetic observation
CN108171394A (en) Multi-robot Task Allocation based on hierachical structure and resource consolidation
CN109409773A (en) A kind of earth observation resource dynamic programming method based on Contract Net Mechanism
CN105978711B (en) A kind of best exchange side lookup method based on minimum spanning tree
CN113256125B (en) Heterogeneous robot cluster task allocation method, system, equipment and storage medium
CN107066322B (en) A kind of online task allocating method towards self-organizing intelligent perception system
CN109202895A (en) The medium of design support apparatus, design support method and design Storage support program
El-Bouri et al. A neural network for dispatching rule selection in a job shop
CN104680238A (en) To-be-observed task determination method for multi-satellite synergistic earth observation
CN106599291B (en) Data grouping method and device
CN108009012A (en) A kind of multiple agent dynamic task allocation method of task based access control model
Soto et al. Boosting autonomous search for CSPs via skylines
CN110009161A (en) Water supply forecast method and device
Fidanova ACO algorithm for MKP using various heuristic information
CN102082737A (en) Method for replacing Web services based on service priority
CN108881442B (en) A kind of mobile cloud network resource allocation methods based on service-aware

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