CN112862270A - Individual task selection method, device and system for distributed multiple robots - Google Patents

Individual task selection method, device and system for distributed multiple robots Download PDF

Info

Publication number
CN112862270A
CN112862270A CN202110079611.XA CN202110079611A CN112862270A CN 112862270 A CN112862270 A CN 112862270A CN 202110079611 A CN202110079611 A CN 202110079611A CN 112862270 A CN112862270 A CN 112862270A
Authority
CN
China
Prior art keywords
task
unsuccessful
candidate
robot
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
Application number
CN202110079611.XA
Other languages
Chinese (zh)
Other versions
CN112862270B (en
Inventor
段俊花
朱怡安
钟冬
姚烨
李联
张黎翔
史先琛
郭玉峰
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.)
Northwestern Polytechnical University
Original Assignee
Northwestern Polytechnical University
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 Northwestern Polytechnical University filed Critical Northwestern Polytechnical University
Priority to CN202110079611.XA priority Critical patent/CN112862270B/en
Publication of CN112862270A publication Critical patent/CN112862270A/en
Application granted granted Critical
Publication of CN112862270B publication Critical patent/CN112862270B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • 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
    • G06Q10/06311—Scheduling, planning or task assignment for a person or group
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

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 invention discloses a distributed individual task selection method, a device and a system for multiple robots, wherein the method comprises the following steps: acquiring a task set comprising at least one task, and respectively assigning each task into a successful task set and an unsuccessful task set; calculating the current response function value of each unsuccessful task, and determining the unsuccessful task with the maximum response function value as a candidate task; calculating the number of potential participants of the candidate task; judging whether the potential number of participants of the candidate task is smaller than the current number of robots required by the candidate task; if yes, determining the candidate task as a final selection task; otherwise, the candidate tasks are classified into the successful task set, and the unsuccessful task with the largest response function value in the rest unsuccessful tasks in the unsuccessful task set is determined as the final selection task. The invention improves the overall intelligent level of the system, can complete more complex cooperative tasks and is beneficial to being effectively applied to a multi-robot application system in a new period.

Description

Individual task selection method, device and system for distributed multiple robots
Technical Field
The invention relates to the technical field of cluster robot control, in particular to a distributed individual task selection method, device and system for multiple robots.
Background
With the rise and rapid development of technologies such as artificial intelligence, big data, internet of things, automation technology and the like, the research of intelligent robots is more and more concerned by people. Compared with a single robot, the multi-robot system has more advantages, and the adoption of distributed autonomous control is the main direction of future research.
In the current multi-robot system, when each robot in the cluster selects a task, each robot cannot be ensured to dynamically participate in the most appropriate task, and the problems of inconsistent and conflicting task selection exist, so that the overall intelligence level of the system is not high, more complex cooperative tasks are difficult to complete, and the system is difficult to effectively apply in a multi-robot application system in a new period.
It is noted that this section is intended to provide a background or context to the embodiments of the disclosure that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
Disclosure of Invention
The embodiment of the invention provides a distributed individual task selection method, a distributed individual task selection device and a distributed individual task selection system for multiple robots, and aims to solve the problems that each robot cannot be ensured to dynamically participate in the most appropriate task and task selection is inconsistent and conflicted in the prior art.
In a first aspect, an embodiment of the present invention provides a distributed multi-robot individual task selection method, including:
acquiring a task set comprising at least one task, and respectively allocating each task into a successful task set and an unsuccessful task set;
calculating the current response function value of each unsuccessful task included in the unsuccessful task set, and determining the unsuccessful task with the maximum response function value as a candidate task;
calculating the potential participant number of the candidate task according to the predicted selection condition of other peer robots not reaching the candidate task to the candidate task;
judging whether the potential number of participants of the candidate task is smaller than the current number of required robots of the candidate task;
if yes, determining the candidate task as a final selection task; otherwise, the candidate task is classified into the successful task set, and the unsuccessful task with the largest response function value in the rest unsuccessful tasks in the unsuccessful task set is determined as the final selection task.
As a preferred mode of the first aspect of the present invention, after the acquiring a task set including at least one task and allocating each task into a successful task set and an unsuccessful task set, the method further includes:
and acquiring the position information of each unsuccessful task included in the unsuccessful task set and the deterministic participant number of each unsuccessful task.
As a preferred mode of the first aspect of the present invention, the calculating a current response function value of each unsuccessful task included in the unsuccessful task set includes:
calculating a current response function value for each unsuccessful task included in the set of unsuccessful tasks according to the following equation:
Figure BDA0002907145340000021
Figure BDA0002907145340000022
wherein, TjFor unsuccessful task tjThe current response function value Rq is the current position of the robot r, RqjFor unsuccessful task tjPosition of (2), RnjFor unsuccessful task tjNumber of machines currently required, CnjFor unsuccessful task tjDeterministic number of participants, k1And k2Is a first tuning parameter associated with the multi-robot system, alpha is a second tuning parameter that prevents the denominator from being zero,
Figure BDA0002907145340000031
is a piecewise function of x, x being Rnj-Cnj。
As a preferred aspect of the first aspect of the present invention, the calculating the number of potential participants of the candidate task according to the predicted selection condition of the candidate task by each peer robot not reaching the candidate task includes:
acquiring a task response value matrix of each peer robot which does not reach the candidate task to the current task of each task included in the task set, wherein the task response value matrix is used for indicating the task response value of each peer robot to each task included in the task set;
acquiring a distance change matrix from each companion robot to each task according to the historical information of each companion robot, wherein the distance change matrix is used for indicating whether each companion robot approaches to each task or not;
acquiring a pre-estimation selection task vector of each companion robot according to the task response value matrix and the distance change matrix, wherein the pre-estimation selection task vector is used for indicating a pre-estimation selection task of each companion robot;
acquiring a prediction selection matrix of each task according to the task response value matrix and each prediction selection task vector;
and calculating the potential participant number of the candidate task according to the prediction selection matrix.
As a preferred mode of the first aspect of the present invention, after the determining that the candidate task is the final selection task or the determining that the unsuccessful task with the largest response function value among the remaining unsuccessful tasks in the unsuccessful task set is the final selection task, the method further includes:
advancing to the final selection task and judging whether the final selection task is reached;
if yes, executing the final selection task; otherwise, continuing to execute the steps of acquiring a task set comprising at least one task and respectively allocating each task into a successful task set and an unsuccessful task set.
In a second aspect, an embodiment of the present invention provides an individual task selection device for distributed multiple robots, including:
the task set acquisition unit is used for acquiring a task set comprising at least one task and dividing each task into a successful task set and an unsuccessful task set respectively;
the candidate task determining unit is used for calculating the current response function value of each unsuccessful task included in the unsuccessful task set and determining the unsuccessful task with the maximum response function value as a candidate task;
the potential participant determining unit is used for calculating the number of potential participants of the candidate task according to the predicted selection condition of other peer robots which do not reach the candidate task on the candidate task;
the first judgment unit is used for judging whether the number of potential participants of the candidate task is smaller than the number of machines currently required by the candidate task;
the final selection task determining unit is used for determining the candidate task as a final selection task if the number of potential participants of the candidate task is less than the number of machines required by the candidate task currently; and if the number of potential participants of the candidate task is more than or equal to the number of machines required by the candidate task currently, dividing the candidate task into the successful task set, and determining the unsuccessful task with the largest response function value in the rest unsuccessful tasks of the unsuccessful task set as a final selection task.
As a preferred mode of the second aspect of the present invention, the potential participant determination unit is specifically configured to:
acquiring a task response value matrix of each peer robot which does not reach the candidate task to the current task of each task included in the task set, wherein the task response value matrix is used for indicating the task response value of each peer robot to each task included in the task set;
acquiring a distance change matrix from each companion robot to each task according to the historical information of each companion robot, wherein the distance change matrix is used for indicating whether each companion robot approaches to each task or not;
acquiring a pre-estimation selection task vector of each companion robot according to the task response value matrix and the distance change matrix, wherein the pre-estimation selection task vector is used for indicating a pre-estimation selection task of each companion robot;
acquiring a prediction selection matrix of each task according to the task response value matrix and each prediction selection task vector;
and calculating the potential participant number of the candidate task according to the prediction selection matrix.
In a third aspect, an embodiment of the present invention provides a distributed individual task selection system for multiple robots, including:
the cluster robot comprises at least two robots which communicate with each other through a communication interface;
the server end is communicated with the clustered robots and used for transmitting a task set containing at least one task to each robot in the clustered robots;
each of the cluster robots is configured to perform the steps of: acquiring a task set comprising at least one task, and respectively allocating each task into a successful task set and an unsuccessful task set; calculating the current response function value of each unsuccessful task included in the unsuccessful task set, and determining the unsuccessful task with the maximum response function value as a candidate task; calculating the potential participant number of the candidate task according to the predicted selection condition of other peer robots not reaching the candidate task to the candidate task; judging whether the potential number of participants of the candidate task is smaller than the current number of required robots of the candidate task; if yes, determining the candidate task as a final selection task; otherwise, the candidate task is classified into the successful task set, and the unsuccessful task with the largest response function value in the rest unsuccessful tasks in the unsuccessful task set is determined as the final selection task.
In a fourth aspect, an embodiment of the present invention provides a computing device, including a processor and a memory, wherein the memory has stored therein execution instructions, and the processor reads the execution instructions in the memory for executing the steps of any one of the above distributed multi-robot individual task selection methods or the preferred modes thereof.
In a fifth aspect, embodiments of the present invention provide a computer-readable storage medium containing computer-executable instructions for performing the steps as described in any one of the above distributed multi-robot individual task selection methods or its preferred forms.
According to the individual task selection method, device and system for the distributed multiple robots, the obtained multiple tasks are divided into the successful task set and the unsuccessful task set respectively, the unsuccessful task with the largest response function value is selected from the unsuccessful task set as the candidate task, then the potential participants of the candidate task are analyzed and selected, if the number of the potential participants is larger than the number of machines required by the candidate task currently, the selection of the potential participants is adjusted, the second selection is performed from the unsuccessful task set, and the selection conflict is avoided.
The invention not only considers the self-organized behavior selection of the individual robots, thereby ensuring that other individual selection tasks are not influenced under the condition that some individual robots fail, ensuring that the system has certain robustness, but also timely adjusting the selection strategy when the problem that more robots advance to the same task is predicted, thereby enabling the robots to participate in more proper tasks, improving the overall intelligence level of the system, completing more complex cooperative tasks and being beneficial to being effectively applied to a multi-robot application system in a new period.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart illustrating an implementation of a distributed multi-robot individual task selection method according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of an individual task selection device of a distributed multi-robot according to an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of a distributed multi-robot individual task selection system according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a computing device according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
With the development of robotics and the demand for production time, the demand for robots is no longer limited to a single robot. Multi-robot systems have become an important direction for robotics research, with many advantages not found with single robot systems.
For example, in the goods field, a multi-robot system is often used to perform logistics management of receiving, transporting and unloading goods by using an AGV (automated Guided Vehicle) scheduling decision system composed of a central control server and a plurality of AGVs (automated Guided vehicles). The central control server uniformly schedules and decides the actions of the AGVs, so that the purpose of logistics management is achieved.
However, in the existing distributed multi-robot system, when each robot in the cluster performs task selection, it cannot be ensured that each robot dynamically participates in the most appropriate task, and there are also problems of inconsistent and conflicting task selection, so that the overall intelligence level of the system is not high, more complex cooperative tasks are difficult to complete, and effective application in a multi-robot application system in a new period is difficult.
Fig. 1 is a flowchart illustrating an implementation of the individual task selection method for distributed multiple robots according to an embodiment, which shows an individual task selection method for distributed multiple robots, so as to solve the above technical problems.
Referring to fig. 1, an embodiment of the present invention discloses a distributed individual task selection method for multiple robots, which mainly includes:
101. acquiring a task set comprising at least one task, and respectively assigning each task into a successful task set and an unsuccessful task set;
102. calculating the current response function value of each unsuccessful task included in the unsuccessful task set, and determining the unsuccessful task with the maximum response function value as a candidate task;
103. calculating the potential participant number of the candidate task according to the predicted selection condition of the candidate task by other peer robots not reaching the candidate task;
104. judging whether the potential number of participants of the candidate task is smaller than the current number of robots required by the candidate task;
105. if the potential participant number of the candidate task is smaller than the current required robot number of the candidate task, determining the candidate task as a final selection task; and if the potential participant number of the candidate task is more than or equal to the current required robot number of the candidate task, dividing the candidate task into a successful task set, and determining the unsuccessful task with the largest response function value in the rest unsuccessful tasks in the unsuccessful task set as the final selection task.
In this embodiment, the execution subject of the method may be any one of the cluster robots.
It should be noted that the distributed multi-robot individual task selection method provided in this embodiment can be used for cluster robot selection in any field and executing tasks.
It should be noted that the clustered robots described in this embodiment include, but are not limited to, two robots. Each robot can be a mobile robot with a small computer as a core and with autonomous computing capability and autonomous navigation capability. In addition, a plurality of communication interfaces can be installed inside each robot, and each robot can communicate with the communication interfaces on other robots through one communication interface on the robot.
In particular, the plurality of communication interfaces may include a WiFi network interface and a 4G IoT network interface. The WiFi network interface can be used for connection communication among the multiple robots, and the 4G IoT network interface can be used for connection communication between the server side and any one of the multiple robots.
In addition, a task set is a data set that contains at least one task and can be transmitted via a network. Wherein, the task set comprises at least one task, which can be understood as a task group, and the task group can include a navigation task, a moving task, a picking task, a prompt task, and the like.
In step 101, any robot in the cluster robots acquires a task set including at least one task from the server, where the tasks in the task set are all tasks that the robot observes at the same time at a certain time, and the robot is the current robot that executes the individual task selection method for distributed multiple robots described in this embodiment.
Each robot acquires external information through a sensor, and different robots use different types of sensors, some cameras, some radars or other sensors when observing tasks, and the sensors are collectively called as position sensing sensors in the embodiment.
The cluster robots respectively store the acquired task sets, and the task states of the task sets acquired by each robot are completely consistent. Of course, the task set may be directly implanted into each robot, and the related description of the task may be input into the robot.
After the current robot acquires a task set containing at least one task, the current robot respectively sorts the tasks into a successful task set and an unsuccessful task set according to the execution condition of each task, so that the successful task set contains a plurality of successful tasks, and the unsuccessful task set contains a plurality of unsuccessful tasks. And each successful task in the successful task set is executed, so that the current robot can perform reactive behavior selection judgment according to the characteristics of the tasks and select an unsuccessful task from the unsuccessful task set to execute.
In this embodiment, { r } for cluster robot1，r2，……，rnDenotes that the total set of tasks that need to be performed is denoted by t1，t2，……，tmThe task set acquired by the robot is a subset of the whole task set, and is represented by { t }1，t2，……，tsWherein s is less than or equal to m. In addition, successful task sets are denoted by TS and unsuccessful task sets are denoted by TU.
After step 101, the following steps are also included:
and acquiring the position information of each unsuccessful task included in each unsuccessful task set and the deterministic participant number of each unsuccessful task.
In this step, when the robot selects a task to be executed from an unsuccessful task set, two factors are mainly considered: the task distance factor is that the closer the current robot is to a certain task, the larger the instinct of selecting the task to execute is; secondly, the urgency degree of the task is the urgency degree, if a certain task needs a plurality of robots to complete, and a plurality of robots are involved, the smaller the number of the needed robots is, the stronger the urgency that the task needs to be completed is, and the stronger the willingness of the current robot to execute is.
Therefore, after the current robot acquires the task set and correspondingly divides each task, it is further required to acquire the position information of each unsuccessful task included in the unsuccessful task set and the deterministic number of participants of each unsuccessful task.
In this embodiment, if the robot reaches a certain task and participates in the execution of the task, the robot is called a deterministic participant of the task.
In step 102, after the current robot acquires the task set and correspondingly divides each task into a successful task set and an unsuccessful task set, and acquires the position information of each unsuccessful task and the deterministic number of participants of each unsuccessful task included in the unsuccessful task set, the current response function value of each unsuccessful task can be further calculated according to the position information.
At present, the robot individual firstly selects an unsuccessful task with the largest response function value as a candidate task from the perspective of instinct.
In an alternative embodiment provided by the present application, the current response function value for each unsuccessful task included in the set of unsuccessful tasks is calculated according to the following equation:
Figure BDA0002907145340000111
Figure BDA0002907145340000112
wherein, TjFor unsuccessful task tjThe current response function value Rq is the current position of the robot r, RqjFor unsuccessful task tjPosition of (2), RnjFor unsuccessful task tjNumber of machines currently required, CnjFor unsuccessful task tjDeterministic number of participants, k1And k2Is a first tuning parameter associated with the multi-robot system, alpha is a second tuning parameter that prevents the denominator from being zero,
Figure BDA0002907145340000113
is a piecewise function of x, x being Rnj-Cnj。
In the above formula, two main influence factors of the current robot selecting the unsuccessful tasks to be executed from the unsuccessful task set are comprehensively considered, and the formula is an inverse proportion function of the distance between the current robot and one of the unsuccessful tasks and is also an inverse proportion function of the number of robots required by the unsuccessful tasks. The larger the calculated response function value is, the closer the unsuccessful task is to the current robot, and the higher the urgency for execution is, and the unsuccessful task with the largest response function value should be preferentially executed.
In this example, k1Typically set as a function related to the boundary length, such as may be set as: k1 ═ border ^2, where border represents the length of the border; k is a radical of2Typically set to the total number of robots required to complete the task; a is typically set to a fraction less than 1.
In step 103, if the behavior selection process based on the robot idiom is a self-organizing task selection process, there is a problem that too many robots select the same task. Although the above formula considers a deterministic participant who has reached a certain unsuccessful task, it is still impossible to detect other peer robots approaching the unsuccessful task, and there is still a problem that many robots advance to the same task. Therefore, there is a need to perform predictive assessment of other peer robots, analysis and selection of potential participants of the unsuccessful task, by further increasing the intelligence of the individual robots.
It should be noted that, in this embodiment, the robot is said to be a potential participant of the task if it is predicted that the robot may participate in the execution of the task according to the position or the movement trend of the robot.
Specifically, the set CI ═ { c for companion robots that the current robot r finds at a certain time t1，c2，……，clMeans if a companion robot c is presentiIf the following formula is satisfied, the companion robot c is considered to beiIs a potential participant in task select (t).
Figure BDA0002907145340000121
Wherein, distt(ciAnd select (t)) represents a companion robot ciDistance, dist, from task select (t) at time tt-1(ciAnd select (t)) represents a companion robot ciDistance, dist, to task select (t) at time t-1t(r, select (t)) represents the distance from the current robot r to the task select (t) at time t.
Condition distt-1(ci，Select(t))≥distt(ciAnd select (t)) represents a companion robot ciIn the move to task select (t).
Condition distt(ci，Select(r))＜distt(r, select (t))iThe distance to task select (t) is smaller than the distance from the current robot r to task select (t).
The meaning of the expression in the above formula is that when the companion robot ciThe distance to task select (t) is continuously shortened, and the companion robot ciIf the distance to task select (t) is smaller than the distance from current robot r to task select (t), then the companion robot c is considerediIs a potential participant in task select (t).
When the distance between the peer robot and the two tasks is very short during the counting of the number of potential participants, the peer robot cannot be counted as the potential participants of the two tasks respectively, and the number needs to be determined specifically according to the following method.
In this embodiment, if a certain robot has reached a certain task, the robot is a deterministic participant and is not listed as a subject of investigation, and does not belong to the companion robot according to the present invention.
In an alternative embodiment provided by the present application, step 103 may be specifically implemented according to the following steps:
103-1, acquiring a task response value matrix of each peer robot not reaching the candidate task for each task in the task set, wherein the task response value matrix is used for indicating the task response value of each peer robot for each task in the task set.
In the step, the current task response value of each companion robot which does not reach the candidate task to each task in the task set is obtained, and the task response value matrix Total _ T is used for expressing:
Figure BDA0002907145340000131
wherein, TijRobot c showing peersiFor a certain task t in the task setjThe task response value of.
103-2, acquiring a distance change matrix from each companion robot to each task according to the historical information of each companion robot, wherein the distance change matrix is used for indicating whether each companion robot approaches to each task.
In the step, the distance change situation from each companion robot to each task is counted according to the historical information of each companion robot, and a variable D is usedijAnd (4) showing. Let distt(ci，tj) Robot c showing peersiAt time t to task tkIf dist, if distt-1(ci，tj)≥distt(ci，tj) Then, it indicates the companion robot ciTo task tjApproach, this time order DijIf not, let Dij＝0。
And (3) representing the distance change situation of each peer robot to each task in the task set by using a distance change matrix Total _ D:
Figure BDA0002907145340000141
wherein D isijRobot c showing peersiFor a certain task t in the task setjIs detected.
103-3, acquiring a pre-estimation selection task vector of each companion robot according to the task response value matrix and the distance change matrix, wherein the pre-estimation selection task vector is used for indicating the pre-estimation selection task of each companion robot.
In the step, the companion robot c obtains a task response value matrix and a distance change matrix according to the stepsiAmong the tasks which are close to each other, the task with the largest task response value is taken as the companion robot ciThe estimation selection task comprises the following steps:
Figure BDA0002907145340000142
meanwhile, generating the estimated selection task vectors of all the companion robots:
Total_PT＝[P_T1 P_T2 … P_Tl]。
and 103-4, acquiring a prediction selection matrix of each task according to the task response value matrix and each pre-prediction selection task vector.
In the step, the selected condition of each task in the task set is counted according to the task response value matrix and each pre-estimated selection task vector obtained in the step. Let S _ TijRepresenting a task tjWhether or not by companion robot ciThe selection of the one or more of the components,
Figure BDA0002907145340000143
the indication is selected such that the user is presented with,
Figure BDA0002907145340000144
indicating that it is not selected. When j is equal to P _ TiWhen the temperature of the water is higher than the set temperature,
Figure BDA0002907145340000145
otherwise
Figure BDA0002907145340000146
The predicted selection of all tasks can be represented by the matrix Total _ S:
Figure BDA0002907145340000147
and 103-5, calculating the potential participant number of the candidate task according to the prediction selection matrix.
In this step, according to the prediction selection matrix obtained above, counting the number En of potential participants of the candidate task, and if k is selected (t):
Figure BDA0002907145340000151
wherein, EnkRepresents the number of potential participants of the candidate task select (t) selected by the current robot r.
In step 104, the current robot performs predictive evaluation on the discoverable peer robots, analyzes and selects the number En of potential participants of the candidate task select (t)select(t)If the number of potential participants EnSelect(t)And if the number of the robots Rn-Cn is larger than the number of the robots currently required by the candidate task, the fact that enough companion robots participate in the candidate task select (t) before the current robot r indicates that the current robot r needs to adjust the selection of the current robot r for the second time so as to avoid conflict.
Therefore, it is necessary to determine whether the number of potential participants of the candidate task is less than the number of machines currently required for the candidate task.
In step 105, if the number of potential participants of the candidate task is less than the number of currently required machines of the candidate task, indicating that there is no potential conflict, the original selection is maintained, i.e. the candidate task is determined to be the final selection task.
And if the number of potential participants of the candidate task is more than or equal to the number of the currently required robots of the candidate task, indicating that potential conflict exists, dividing the candidate task into a successful task set of the current robot, and determining the unsuccessful task with the largest response function value in the rest unsuccessful tasks of the unsuccessful task set as a final selection task.
And after the occurrence of the conflict is predicted, permanently deleting the candidate tasks from the unsuccessful task set observed by the current robot, and adding the candidate tasks into the successful task set of the current robot. The current robot does not calculate the task but selects other tasks when planning next time, and the memory mechanism ensures that the current robot does not repeatedly select the same task, thereby improving the selection efficiency.
After step 105, the following steps are also included:
proceeding to the final selection task and judging whether the final selection task is reached;
if yes, executing a final selection task; otherwise, the step 101 is continued.
In the above steps, after the current robot determines the final selection task, the current robot will advance to the final selection task, and then judge whether the final selection task is reached. If the task is reached, the final selection task is executed, and if the task is not reached, the step 101 is continuously executed to perform judgment, so that the most appropriate task is selected to be executed.
The method of the embodiment of the invention is a real-time planning process, and the environment needs to be reviewed again and task selection needs to be performed again every time the robot moves forward one step. The real-time selection method is beneficial to detecting the change of the environment at any time, so that the method can be well adapted to the dynamic uncertain environment, and the robot can participate in more proper tasks.
In summary, the individual task selection method for distributed multiple robots provided by the embodiment of the present invention considers the behavior selection of individual robots in self-organization, thereby ensuring that, when some individual robots fail, other individual selection tasks are not affected, so that the system has a certain robustness, and the selection strategy is timely adjusted when it is predicted that there is a problem that more robots advance to the same task, so that the robots can participate in more appropriate tasks, the overall intelligence level of the system is improved, more complex cooperative tasks can be completed, and effective application in a new-period multiple-robot application system is facilitated.
It should be noted that the above-mentioned embodiments of the method are described as a series of actions for simplicity of description, but those skilled in the art should understand that the present invention is not limited by the described sequence of actions. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no particular act is required to implement the invention.
Referring to fig. 2, based on the same inventive concept, an embodiment of the present invention provides a distributed individual task selection device for multiple robots, where the device mainly includes:
a task set obtaining unit 21, configured to obtain a task set including at least one task, and assign each task into a successful task set and an unsuccessful task set respectively;
a candidate task determining unit 22, configured to calculate a current response function value of each unsuccessful task included in the unsuccessful task set, and determine the unsuccessful task with the largest response function value as a candidate task;
the potential participant determining unit 23 is configured to calculate the number of potential participants of the candidate task according to the predicted selection condition of the candidate task by each of the remaining peer robots that do not reach the candidate task;
the first judging unit 24 is used for judging whether the number of potential participants of the candidate task is smaller than the number of machines currently required by the candidate task;
a final selection task determining unit 25, configured to determine that the candidate task is the final selection task if the number of potential participants of the candidate task is smaller than the number of machines currently required by the candidate task; and if the potential participant number of the candidate task is more than or equal to the current required robot number of the candidate task, dividing the candidate task into a successful task set, and determining the unsuccessful task with the largest response function value in the rest unsuccessful tasks in the unsuccessful task set as the final selection task.
It should be noted here that the task set obtaining unit 21, the candidate task determining unit 22, the potential participant determining unit 23, the first judging unit 24 and the final selected task determining unit 25 correspond to steps 101 to 105 in the above method embodiment, and the four units are the same as the examples and application scenarios realized by the corresponding steps, but are not limited to the disclosure in the above method embodiment.
It should be noted that the distributed multi-robot individual task selection device provided in this embodiment can be used for cluster robot selection in any field and executing tasks.
In an optional embodiment provided by the present application, the apparatus further comprises:
and the task information acquisition unit is used for acquiring the position information of each unsuccessful task included in the unsuccessful task set and the deterministic participant number of each unsuccessful task.
In an optional embodiment provided by the present application, the candidate task determining unit 22 is specifically configured to:
calculating a current first response function value for each unsuccessful task included in the set of unsuccessful tasks according to the following equation:
Figure BDA0002907145340000181
Figure BDA0002907145340000182
wherein, TjFor the current robot r to the unsuccessful task tjThe current first response function value Rq is the current position of the current robot r, RqjFor unsuccessful task tjCurrent position of (2), RnjFor unsuccessful task tjNumber of machines currently required, CnjFor unsuccessful task tjDeterministic number of participants, k1And k2Is a first tuning parameter associated with the multi-robot system, and alpha is a prevention denominatorA second adjustment parameter of zero is set to,
Figure BDA0002907145340000183
is a piecewise function of x, x being Rnj-Cnj。
In an optional embodiment provided by the present application, the potential participant determination unit 23 is specifically configured to:
acquiring a task response value matrix of each peer robot which does not reach the candidate task to the current task of each task in the task set, wherein the task response value matrix is used for indicating the task response value of each peer robot to each task in the task set;
acquiring a distance change matrix from each companion robot to each task according to the historical information of each companion robot, wherein the distance change matrix is used for indicating whether each companion robot approaches to each task or not;
acquiring a pre-estimated selection task vector of each companion robot according to the task response value matrix and the distance change matrix, wherein the pre-estimated selection task vector is used for indicating the pre-estimated selection task of each companion robot;
according to the task response value matrix and each pre-estimated selection task vector, obtaining a prediction selection matrix of each task;
and calculating the number of potential participants of the candidate task according to the prediction selection matrix.
In an optional embodiment provided by the present application, the apparatus further comprises:
the second judgment unit is used for judging whether the current robot reaches the final selection task;
the task execution unit is used for enabling the current robot to execute the final selection task if the current robot reaches the final selection task; and if the current robot does not reach the final selected task, continuing to execute the steps of acquiring a task set comprising at least one task and dividing each task into a successful task set and an unsuccessful task set according to the current robot.
In summary, the individual task selection device for distributed multiple robots provided by the embodiment of the present invention considers the behavior selection of individual robots in self-organization, thereby ensuring that, when some individual robots fail, other individual selection tasks are not affected, so that the system has a certain robustness, and the selection strategy is timely adjusted when it is predicted that there is a problem that more robots advance to the same task, so that the robots can participate in more appropriate tasks, the overall intelligence level of the system is improved, more complex cooperative tasks can be completed, and effective application in a new-period multiple-robot application system is facilitated.
It should be noted that the individual task selection device for distributed multiple robots provided in the embodiment of the present invention and the individual task selection method for distributed multiple robots described in the foregoing embodiment belong to the same technical concept, and the specific implementation process thereof may refer to the description of the method steps in the foregoing embodiment, and will not be described herein again.
It should be understood that the above distributed multi-robot individual task selection device includes only units that are logically divided according to the functions implemented by the device, and in practical applications, the above units may be stacked or split. Moreover, the functions implemented by the distributed multi-robot individual task selection device provided in this embodiment correspond to the distributed multi-robot individual task selection method provided in the above embodiment one to one, and for the more detailed processing flow implemented by the device, detailed description is already made in the above method embodiment, and detailed description is not given here.
Referring to fig. 3, based on the same technical concept, an embodiment of the present invention provides a distributed individual task selection system for multiple robots, where the system mainly includes:
a cluster robot 31 including at least two robots which communicate with each other through a communication interface;
a server 32, in communication with the clustered robots 31, for transmitting a task set comprising at least one task to each of the clustered robots 31;
each of the cluster robots 31 is configured to perform the following steps: acquiring a task set comprising at least one task, and respectively assigning each task into a successful task set and an unsuccessful task set; calculating the current response function value of each unsuccessful task included in the unsuccessful task set, and determining the unsuccessful task with the maximum response function value as a candidate task; calculating the potential participant number of the candidate task according to the predicted selection condition of the candidate task by other peer robots not reaching the candidate task; judging whether the potential number of participants of the candidate task is smaller than the current number of robots required by the candidate task; if yes, determining the candidate task as a final selection task; otherwise, the candidate tasks are classified into the successful task set, and the unsuccessful task with the largest response function value in the rest unsuccessful tasks in the unsuccessful task set is determined as the final selection task.
It should be noted that the distributed multi-robot individual task selection system provided in the embodiment of the present invention can be used for cluster robot selection in any field and performing tasks.
It should be noted that the clustered robots described in this embodiment include, but are not limited to, two robots. Each robot can be a mobile robot with a small computer as a core and with autonomous computing capability and autonomous navigation capability. In addition, a plurality of communication interfaces can be installed inside each robot, and each robot can communicate with the communication interfaces on other robots through one communication interface on the robot.
In particular, the plurality of communication interfaces may include a WiFi network interface and a 4G IoT network interface. The WiFi network interface can be used for connection communication among the multiple robots, and the 4G IoT network interface can be used for connection communication between the server side and any one of the multiple robots.
Referring to fig. 4, based on the same inventive concept, an embodiment of the present invention provides a computing device, which mainly includes a processor 41 and a memory 42, wherein the memory 42 stores execution instructions. The processor 41 reads the execution instructions in the memory 42 for executing the steps described in any of the embodiments of the distributed multi-robot individual task selection method described above. Alternatively, the processor 41 reads the execution instructions in the memory 42 for implementing the functions of the units in any embodiment of the individual task selection device of the distributed multi-robot.
Fig. 4 is a schematic structural diagram of a computing device according to an embodiment of the present invention, as shown in fig. 4, the computing device includes a processor 41, a memory 42, and a transceiver 43; wherein the processor 41, the memory 42 and the transceiver 43 are interconnected by a bus 44.
The memory 42 is used for storing programs; in particular, the program may include program code including computer operating instructions. The memory 52 may include a volatile memory (volatile memory), such as a random-access memory (RAM); the memory 42 may also include a non-volatile memory (non-volatile memory), such as a flash memory (flash memory), a Hard Disk Drive (HDD) or a solid-state drive (SSD); the memory 42 may also comprise a combination of the above-mentioned kinds of memories.
The bus 44 may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown in FIG. 4, but this does not indicate only one bus or one type of bus.
The processor 41 may be a Central Processing Unit (CPU), a Network Processor (NP), or a combination of CPU and NP. But also a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a Programmable Logic Device (PLD), or a combination thereof. The PLD may be a Complex Programmable Logic Device (CPLD), a Field Programmable Gate Array (FPGA), a General Array Logic (GAL), or any combination thereof.
In addition, an embodiment of the present invention further provides a computer-readable storage medium containing computer-executable instructions, where the computer-executable instructions are used to perform the steps described in any embodiment of the distributed multi-robot individual task selection method. Alternatively, the computer executable instructions are used to perform the functions of the various elements of the distributed multi-robot individual task selection device embodiment described above.
Optionally, in this embodiment, the storage medium may be located in any one of computer terminals in a computer terminal group in a computer network, or in any one of mobile terminals in a mobile terminal group.
In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (10)

1. A distributed individual task selection method for multiple robots is characterized by comprising the following steps:
acquiring a task set comprising at least one task, and respectively allocating each task into a successful task set and an unsuccessful task set;
calculating the current response function value of each unsuccessful task included in the unsuccessful task set, and determining the unsuccessful task with the maximum response function value as a candidate task;
calculating the potential participant number of the candidate task according to the predicted selection condition of other peer robots not reaching the candidate task to the candidate task;
judging whether the potential number of participants of the candidate task is smaller than the current number of required robots of the candidate task;
if yes, determining the candidate task as a final selection task; otherwise, the candidate task is classified into the successful task set, and the unsuccessful task with the largest response function value in the rest unsuccessful tasks in the unsuccessful task set is determined as the final selection task.
2. The method of claim 1, wherein after the obtaining a task set including at least one task and the assigning each task into a successful task set and an unsuccessful task set, further comprising:
and acquiring the position information of each unsuccessful task included in the unsuccessful task set and the deterministic participant number of each unsuccessful task.
3. The method of claim 2, wherein calculating a current response function value for each unsuccessful task included in the set of unsuccessful tasks comprises:
calculating a current response function value for each unsuccessful task included in the set of unsuccessful tasks according to the following equation:
Figure FDA0002907145330000021
Figure FDA0002907145330000022
wherein, TjFor unsuccessful task tjThe current response function value Rq is the current position of the robot r, RqjFor unsuccessful task tjPosition of (2), RnjFor unsuccessful task tjNumber of machines currently required, CnjFor unsuccessful task tjDeterministic number of participants, k1And k2Is a first tuning parameter associated with the multi-robot system, alpha is a second tuning parameter that prevents the denominator from being zero,
Figure FDA0002907145330000023
is a piecewise function of x, x being Rnj-Cnj。
4. The method of claim 1, wherein calculating the number of potential participants of the candidate task according to the predicted selection of the candidate task by each peer robot not arriving at the candidate task comprises:
acquiring a task response value matrix of each peer robot which does not reach the candidate task to the current task of each task included in the task set, wherein the task response value matrix is used for indicating the task response value of each peer robot to each task included in the task set;
acquiring a distance change matrix from each companion robot to each task according to the historical information of each companion robot, wherein the distance change matrix is used for indicating whether each companion robot approaches to each task or not;
acquiring a pre-estimation selection task vector of each companion robot according to the task response value matrix and the distance change matrix, wherein the pre-estimation selection task vector is used for indicating a pre-estimation selection task of each companion robot;
acquiring a prediction selection matrix of each task according to the task response value matrix and each prediction selection task vector;
and calculating the potential participant number of the candidate task according to the prediction selection matrix.
5. The method according to any one of claims 1 to 4, wherein after the determining that the candidate task is a final selection task or the determining that an unsuccessful task with a largest response function value among the rest unsuccessful tasks in the unsuccessful task set is a final selection task, the method further comprises:
advancing to the final selection task and judging whether the final selection task is reached;
if yes, executing the final selection task; otherwise, continuing to execute the steps of acquiring a task set comprising at least one task and respectively allocating each task into a successful task set and an unsuccessful task set.
6. A distributed multi-robot individual task selection apparatus, comprising:
the task set acquisition unit is used for acquiring a task set comprising at least one task and dividing each task into a successful task set and an unsuccessful task set respectively;
the candidate task determining unit is used for calculating the current response function value of each unsuccessful task included in the unsuccessful task set and determining the unsuccessful task with the maximum response function value as a candidate task;
the potential participant determining unit is used for calculating the number of potential participants of the candidate task according to the predicted selection condition of other peer robots which do not reach the candidate task on the candidate task;
the first judgment unit is used for judging whether the number of potential participants of the candidate task is smaller than the number of machines currently required by the candidate task;
the final selection task determining unit is used for determining the candidate task as a final selection task if the number of potential participants of the candidate task is less than the number of machines required by the candidate task currently; and if the number of potential participants of the candidate task is more than or equal to the number of machines required by the candidate task currently, dividing the candidate task into the successful task set, and determining the unsuccessful task with the largest response function value in the rest unsuccessful tasks of the unsuccessful task set as a final selection task.
7. The apparatus according to claim 6, wherein the potential participant determination unit is specifically configured to:
acquiring a task response value matrix of each peer robot which does not reach the candidate task to the current task of each task included in the task set, wherein the task response value matrix is used for indicating the task response value of each peer robot to each task included in the task set;
acquiring a distance change matrix from each companion robot to each task according to the historical information of each companion robot, wherein the distance change matrix is used for indicating whether each companion robot approaches to each task or not;
acquiring a pre-estimation selection task vector of each companion robot according to the task response value matrix and the distance change matrix, wherein the pre-estimation selection task vector is used for indicating a pre-estimation selection task of each companion robot;
acquiring a prediction selection matrix of each task according to the task response value matrix and each prediction selection task vector;
and calculating the potential participant number of the candidate task according to the prediction selection matrix.
8. A distributed multi-robot individual task selection system, comprising:
the cluster robot comprises at least two robots which communicate with each other through a communication interface;
the server end is communicated with the clustered robots and used for transmitting a task set containing at least one task to each robot in the clustered robots;
each of the cluster robots is configured to perform the steps of: acquiring a task set comprising at least one task, and respectively allocating each task into a successful task set and an unsuccessful task set; calculating the current response function value of each unsuccessful task included in the unsuccessful task set, and determining the unsuccessful task with the maximum response function value as a candidate task; calculating the potential participant number of the candidate task according to the predicted selection condition of other peer robots not reaching the candidate task to the candidate task; judging whether the potential number of participants of the candidate task is smaller than the current number of required robots of the candidate task; if yes, determining the candidate task as a final selection task; otherwise, the candidate task is classified into the successful task set, and the unsuccessful task with the largest response function value in the rest unsuccessful tasks in the unsuccessful task set is determined as the final selection task.
9. A computing device comprising a processor and a memory, wherein the memory has stored therein execution instructions, the processor reading the execution instructions in the memory for performing the steps in the distributed multi-robot individual task selection method according to any one of claims 1-5.
10. A computer-readable storage medium containing computer-executable instructions for performing the steps in the distributed multi-robot individual task selection method of any of claims 1-5.
CN202110079611.XA 2021-01-20 2021-01-20 Individual task selection method, device and system for distributed multi-robot Active CN112862270B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110079611.XA CN112862270B (en) 2021-01-20 2021-01-20 Individual task selection method, device and system for distributed multi-robot

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110079611.XA CN112862270B (en) 2021-01-20 2021-01-20 Individual task selection method, device and system for distributed multi-robot

Publications (2)

Publication Number Publication Date
CN112862270A true CN112862270A (en) 2021-05-28
CN112862270B CN112862270B (en) 2023-08-11

Family

ID=76008487

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110079611.XA Active CN112862270B (en) 2021-01-20 2021-01-20 Individual task selection method, device and system for distributed multi-robot

Country Status (1)

Country Link
CN (1) CN112862270B (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140249676A1 (en) * 2013-03-04 2014-09-04 Microsoft Corporation Adapting robot behavior based upon human-robot interaction
CN107895225A (en) * 2017-11-01 2018-04-10 北京邮电大学 A kind of cooperation type method for allocating tasks of multi-Agent Lothrus apterus
US20180349207A1 (en) * 2017-06-01 2018-12-06 International Business Machines Corporation Software bot conflict-resolution service agent
CN108985580A (en) * 2018-06-16 2018-12-11 齐齐哈尔大学 Multirobot disaster based on improved BP searches and rescues method for allocating tasks
CN110852486A (en) * 2019-10-16 2020-02-28 中国人民解放军国防科技大学 Task planning method for autonomous cooperation of unmanned aerial vehicle cluster
JP6687788B1 (en) * 2018-11-19 2020-04-28 炬星科技（深▲せん▼）有限公司Ｓｙｒｉｕｓ Ｒｏｂｏｔｉｃｓ Ｃｏ．， Ｌｔｄ． Decentralized robot scheduling determination method, device, system, electronic device, and storage medium
CN111786987A (en) * 2020-06-29 2020-10-16 杭州海康机器人技术有限公司 Task issuing method, device, system and equipment
CN112001125A (en) * 2020-08-27 2020-11-27 河北工业大学 Failure rate prediction method for industrial robot

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140249676A1 (en) * 2013-03-04 2014-09-04 Microsoft Corporation Adapting robot behavior based upon human-robot interaction
US20180349207A1 (en) * 2017-06-01 2018-12-06 International Business Machines Corporation Software bot conflict-resolution service agent
CN107895225A (en) * 2017-11-01 2018-04-10 北京邮电大学 A kind of cooperation type method for allocating tasks of multi-Agent Lothrus apterus
CN108985580A (en) * 2018-06-16 2018-12-11 齐齐哈尔大学 Multirobot disaster based on improved BP searches and rescues method for allocating tasks
JP6687788B1 (en) * 2018-11-19 2020-04-28 炬星科技（深▲せん▼）有限公司Ｓｙｒｉｕｓ Ｒｏｂｏｔｉｃｓ Ｃｏ．， Ｌｔｄ． Decentralized robot scheduling determination method, device, system, electronic device, and storage medium
CN110852486A (en) * 2019-10-16 2020-02-28 中国人民解放军国防科技大学 Task planning method for autonomous cooperation of unmanned aerial vehicle cluster
CN111786987A (en) * 2020-06-29 2020-10-16 杭州海康机器人技术有限公司 Task issuing method, device, system and equipment
CN112001125A (en) * 2020-08-27 2020-11-27 河北工业大学 Failure rate prediction method for industrial robot

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
JUNHUA DUAN 等: "Study and design of multi-robots pursuing based on improved ant colony labor division method", 2012 IEEE INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND AUTOMATION ENGINEERING (CSAE) *
XINYE CHEN 等: "A distributed method for dynamic multi-robot task allocation problems with critical time constraints", ROBOTICS AND AUTONOMOUS SYSTEMS *
刘小梅;田彦涛;杨茂;: "基于博弈论的多机器人任务分配算法", 吉林大学学报(信息科学版), no. 03 *
宋玉丽;张国有;曾建潮;薛颂东;: "面向多目标搜索的群机器人任务分配研究", 太原科技大学学报, no. 04 *
段俊花 ，朱怡安: "基于群智能的多智能体系统体系结构设计与应用", 西北工业大学学报 *

Also Published As

Publication number Publication date
CN112862270B (en) 2023-08-11

Similar Documents

Publication Publication Date Title
Zheng et al. A deep bayesian policy reuse approach against non-stationary agents
Arviv et al. Collaborative reinforcement learning for a two-robot job transfer flow-shop scheduling problem
US20190138901A1 (en) Techniques for designing artificial neural networks
WO2020042739A1 (en) Data preprocessing method and apparatus, computer device, and storage medium
CN111712840B (en) Annotation method, annotation device, storage medium, and identification system
US8385649B2 (en) Information processing apparatus and method for detecting object in image data
US11513851B2 (en) Job scheduler, job schedule control method, and storage medium
WO2021236273A1 (en) Joint learning from explicit and inferred labels
JP7099968B2 (en) Arithmetic logic unit
US20220108153A1 (en) Bayesian context aggregation for neural processes
CN111401545A (en) Neural network optimization device and neural network optimization method
EP3591584A1 (en) Probabilistic training for binary neural networks
CN115729796A (en) Abnormal operation analysis method based on artificial intelligence and big data application system
CN109190757B (en) Task processing method, device, equipment and computer readable storage medium
JP2022514886A (en) How to Train Neural Networks
JP2022507144A (en) Computer architecture for artificial image generation
CN112862270A (en) Individual task selection method, device and system for distributed multiple robots
KR102152081B1 (en) Valuation method based on deep-learning and apparatus thereof
CN111027679A (en) Abnormal data detection method and system
CN109416748B (en) SVM-based sample data updating method, classification system and storage device
US20220358572A1 (en) Device and method to provide data associated with shopping mall web page
CN112894811B (en) Distributed multi-robot intelligent control method and device based on group intelligent MAS
CN115660155A (en) Task execution method and device of multi-agent diversity strategy and agent
US11651282B2 (en) Learning method for learning action of agent using model-based reinforcement learning
Kneissler et al. Filtering sensory information with XCSF: improving learning robustness and robot arm control performance

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