CN116245257B - Multi-robot scheduling method and device - Google Patents

Multi-robot scheduling method and device Download PDF

Info

Publication number
CN116245257B
CN116245257B CN202310503813.1A CN202310503813A CN116245257B CN 116245257 B CN116245257 B CN 116245257B CN 202310503813 A CN202310503813 A CN 202310503813A CN 116245257 B CN116245257 B CN 116245257B
Authority
CN
China
Prior art keywords
task
robot
preset
information
small node
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.)
Active
Application number
CN202310503813.1A
Other languages
Chinese (zh)
Other versions
CN116245257A (en
Inventor
彭云龙
安丽
李伟
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.)
Individual
Original Assignee
Individual
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 Individual filed Critical Individual
Priority to CN202310503813.1A priority Critical patent/CN116245257B/en
Publication of CN116245257A publication Critical patent/CN116245257A/en
Application granted granted Critical
Publication of CN116245257B publication Critical patent/CN116245257B/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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • 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
    • G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
    • 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)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Educational Administration (AREA)
  • Manipulator (AREA)

Abstract

The application relates to the technical field of multi-robot scheduling, and discloses a multi-robot scheduling method and device. The method is used for scheduling a plurality of robots, and comprises the following steps: step S1, road network vector diagram data, robot information and target task information of a patrol area are obtained, step S2, task allocation is carried out on a corresponding robot in the robot information according to the road network vector diagram data and the target task information and by combining a preset allocated position inheritance rule, a preset auction mechanism and a preset task allocation standard, a task allocation result is obtained, step S3, the task allocation result is optimized through a tabu search algorithm, an optimized task allocation result is obtained, and step S4, based on the optimized task allocation result, the corresponding robot in the robot information starts to execute tasks; the application not only can reasonably distribute the task points, but also can reasonably arrange the task execution sequence, thereby improving the system execution efficiency.

Description

Multi-robot scheduling method and device
Technical Field
The application relates to the technical field of multi-robot scheduling, in particular to a multi-robot scheduling method and device.
Background
Along with the development of science and technology, the artificial intelligence technology is mature, and the construction of wisdom garden and wisdom commodity circulation has brought a series of facility for people. The traditional manual inspection, logistics distribution and other works have the problems of high labor intensity, low execution efficiency, overhigh labor cost and the like. Therefore, the inspection security work, logistics park distribution and the like are going to the intelligent inspection and intelligent logistics based on the artificial intelligence technology, and robots are used for replacing the work of inspection personnel or distribution personnel, so that the inspection and distribution work efficiency can be improved, higher economic benefits are achieved, and the convenience of inspection and distribution work is greatly improved.
However, when a plurality of inspection and distribution robots exist in the system, the inspection and distribution robots are reasonably distributed and scheduled in consideration of the difference of the capacity of each inspection and distribution robot for executing different tasks and the fault tolerance capacity, so that the system efficiency becomes a complex optimization problem.
Therefore, in order to solve the technical problems of reasonable distribution and scheduling among inspection and distribution robots with different capacities, a scheduling method of robots is needed.
Disclosure of Invention
The application aims to provide a multi-robot scheduling method and device, which are used for referring to a commodity auction mechanism, bidding is carried out on target points through different robots, reasonable task points are distributed, and meanwhile, the task execution sequence is reasonably distributed through a tabu search algorithm in a parallel computing mode, so that the system execution efficiency is improved.
In a first aspect, the present application provides a multi-robot scheduling method, including:
step S1, road network vector diagram data, robot information and target task information of a patrol area are obtained;
step S2, task allocation is carried out on the corresponding robots in the robot information according to the road network vector diagram data and the target task information by combining a preset allocated position inheritance rule, a preset auction mechanism and a preset task allocation standard, so as to obtain a task allocation result;
step S3, optimizing the task distribution result through a tabu search algorithm to obtain an optimized task distribution result;
and S4, starting to execute tasks by the corresponding robots in the robot information based on the optimized task allocation result.
The multi-robot scheduling method provided by the embodiment of the application can realize the scheduling of robots, bid on target points by different robots by referring to a commodity auction mechanism, reasonable task points are distributed, and meanwhile, the task execution sequence is reasonably distributed by a tabu search algorithm in a parallel computing mode, so that the system execution efficiency is improved.
Optionally, the target task information includes a plurality of small node tasks; the step S2 includes:
Step S21, setting the running path data and the movement corner data required by the robot to complete the task as task execution cost indexes;
step S22, calculating task execution cost values of the robots corresponding to the robot information for completing the tasks of the plurality of small nodes according to the road network vector diagram data, the target task information and the preset task allocation standard;
step S23, extracting the minimum value of task execution cost values corresponding to all small node tasks based on the preset auction mechanism and the allocated position inheritance rule, and allocating the small node tasks corresponding to the minimum value to the robots corresponding to the minimum value to obtain the task allocation result.
The multi-robot scheduling method provided by the application can realize the scheduling of robots, and by referring to a commodity auction mechanism, bidding is carried out on target points through different robots, so that reasonable task points are distributed.
Optionally, the step S22 includes:
step S221, calculating running path data and movement corner data of the corresponding robot in the robot information for completing the tasks of the plurality of small nodes according to the road network vector diagram data, the target task information and the preset task allocation standard;
Step S222, based on a preset weight coefficient, carrying out weighted operation on the running path data and the movement corner data to obtain a task execution cost value of the robot for completing the tasks of the plurality of small nodes.
Optionally, the step S3 includes:
step S31, judging whether the task number of each robot in the task allocation result is larger than a preset threshold value; if yes, go to step S32; if not, executing step S33;
step S32, rearranging task execution sequences of robots with the number of tasks being greater than the preset threshold value through the tabu search algorithm to obtain an optimal task arrangement sequence, and optimizing the task distribution result based on the optimal task arrangement sequence to obtain the optimized task distribution result;
and step S33, determining the task allocation result as the optimized task allocation result.
The multi-robot scheduling method provided by the application can realize the scheduling of robots, reasonably arrange the task execution sequence through a tabu search algorithm in a parallel computing mode, and improve the system execution efficiency.
Optionally, after the step S4, the method further includes:
when the task is incomplete due to the robot fault in the task execution process, the incomplete task is selected as the task to be distributed, and the step S2 is executed in a return mode.
In a second aspect, the present application also provides a multi-robot scheduling apparatus, including:
the acquisition module is used for acquiring road network vector diagram data, robot information and target task information of the inspection area;
the allocation module is used for allocating tasks to the corresponding robots in the robot information according to the road network vector diagram data and the target task information by combining a preset allocated position inheritance rule, a preset auction mechanism and a preset task allocation standard to obtain a task allocation result;
the optimization module is used for optimizing the task distribution result through a tabu search algorithm to obtain an optimized task distribution result;
and the execution module is used for starting to execute the task by the corresponding robot in the robot information based on the optimized task allocation result.
The multi-robot scheduling device provided by the application can realize the scheduling of robots, bid on target points by different robots by referring to a commodity auction mechanism, reasonable task points are distributed, and meanwhile, the task execution sequence is reasonably distributed by a tabu search algorithm in a parallel computing mode, so that the system execution efficiency is improved.
Optionally, the target task information includes a plurality of small node tasks; the distribution module comprises:
The setting sub-module is used for setting the running path data and the movement corner data required by the robot to finish the task as task execution cost indexes;
the calculation sub-module is used for calculating the task execution cost value of the corresponding robot in the robot information for completing the tasks of the plurality of small nodes according to the road network vector diagram data, the target task information and the preset task allocation standard;
and the allocation sub-module is used for extracting the minimum value in the task execution cost value corresponding to each small node task based on the preset auction mechanism and the allocated position inheritance rule, and allocating the small node task corresponding to the minimum value to the robot corresponding to the minimum value to obtain the task allocation result.
The multi-robot scheduling device provided by the application can realize the scheduling of robots, and bid on target points through different robots by referring to a commodity auction mechanism, so that reasonable task points are distributed.
Optionally, the computing submodule includes:
the calculation unit is used for calculating the running path data and the movement corner data of the corresponding robots in the robot information for completing the tasks of the plurality of small nodes according to the road network vector diagram data, the target task information and the preset task allocation standard;
And the weighting unit is used for carrying out weighting operation on the running path data and the movement corner data based on a preset weight coefficient to obtain a task execution cost value of the robot for completing the tasks of the plurality of small nodes.
Optionally, the optimizing module includes:
the judging sub-module is used for judging whether the task number of each robot in the task allocation result is larger than a preset threshold value; if yes, triggering the optimization submodule to execute the corresponding steps; if not, triggering the determination submodule to execute the corresponding step;
the optimizing sub-module is used for rearranging task execution sequences of robots with the number of tasks being larger than the preset threshold value through the tabu search algorithm to obtain an optimal task arrangement sequence, and optimizing the task distribution result based on the optimal task arrangement sequence to obtain the optimized task distribution result;
and the determining submodule is used for determining the task distribution result as the optimized task distribution result.
Optionally, the apparatus further comprises:
and the return module is used for selecting the unfinished task as the task to be distributed when the task is unfinished due to the robot fault in the task execution process, and triggering the distribution module to execute the corresponding steps.
As can be seen from the foregoing, according to the multi-robot scheduling method and apparatus provided by the present application, through step S1, road network vector diagram data, robot information and target task information of a patrol area are obtained, step S2, according to the road network vector diagram data and the target task information, task allocation is performed on a corresponding robot in the robot information in combination with a preset allocated position inheritance rule, a preset auction mechanism and a preset task allocation standard, so as to obtain a task allocation result, step S3, the task allocation result is optimized through a tabu search algorithm, so as to obtain an optimized task allocation result, and step S4, based on the optimized task allocation result, the corresponding robot in the robot information starts to execute a task, thereby realizing scheduling of the multi-robot, referring to a commodity auction mechanism, bidding is performed on a target point through different robots, reasonable task points are allocated, and simultaneously, task execution order is reasonably arranged through a tabu search algorithm in a parallel computing manner, so that system execution efficiency is improved.
Additional features and advantages of the application will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the embodiments of the application. The objectives and other advantages of the application may be realized and attained by the structure particularly pointed out in the written description and drawings.
Drawings
Fig. 1 is a flowchart of a multi-robot scheduling method according to an embodiment of the present application.
Fig. 2 is another flowchart of a multi-robot scheduling method according to an embodiment of the present application.
Fig. 3 is a schematic structural diagram of a multi-robot scheduling system according to an embodiment of the present application.
Fig. 4 is a schematic diagram of a task execution sequence of a robot according to an embodiment of the present application.
Fig. 5 is a schematic process diagram of an auction mechanism according to an embodiment of the present application.
Fig. 6 is a schematic diagram of domain solution set generation of a tabu search algorithm according to an embodiment of the present application.
Fig. 7 is a schematic structural diagram of a multi-robot scheduling device according to an embodiment of the present application.
Description of the reference numerals: 301. a robot; 302. road network and nodes; 303. a goods shelf; 501. a full connection road network; 502. a first round of auction; 503. a second round of auction; 504. a third round of auction; 601. generating a candidate solution; 701. an acquisition module; 702. a distribution module; 703. an optimization module; 704. and executing the module.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the application, as presented in the figures, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by a person skilled in the art without making any inventive effort, are intended to be within the scope of the present application.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.
Referring to fig. 1, fig. 1 is a flowchart of a multi-robot scheduling method according to an embodiment of the application. The multi-robot scheduling method is used for scheduling a plurality of robots and comprises the following steps:
step S101, road network vector diagram data, robot information and target task information of a patrol area are obtained;
the target task information includes a plurality of small node tasks and a target total task, and the plurality of small node tasks form the target total task.
The robot information includes information of a plurality of executable robots, and the robot information is composed of information of the plurality of executable robots.
Step S102, task allocation is carried out on the corresponding robots in the robot information according to the road network vector diagram data and the target task information by combining a preset allocated position inheritance rule, a preset auction mechanism and a preset task allocation standard, so as to obtain a task allocation result;
It should be noted that, the preset assigned position inheritance rule is that, when a new round of auction bidding is performed, if any robot assigns (bids) a small node task on a round, the robot needs to perform cost calculation (bidding) on the basis of assigning the small node task, that is, calculates the cost from the position where the robot completes the previous round of task to the position where the task completes the new round of auction.
The preset auction mechanism is in a form of bidding, and task allocation is carried out by taking minimum task execution cost from an executable robot in the system to each small node task as a reference.
The preset task allocation standard is that the total cost of executing all executable robot tasks in the system is minimum, and the task allocation standard is an execution target of the system.
The movement angle data is the accumulation of rotation angles used by the executable robot to complete the task.
In the embodiment of the invention, running path data and movement corner data required by a robot to complete a task are set as task execution cost indexes, task execution cost values of a plurality of small node tasks, which are completed by the corresponding robot, in the robot information are calculated according to road network vector diagram data, target task information and preset task allocation standards, minimum values in the task execution cost values corresponding to the small node tasks are extracted based on a preset auction mechanism and an allocated position inheritance rule, and the small node tasks corresponding to the minimum values are allocated to the robots corresponding to the minimum values, so that a task allocation result is obtained.
Step S103, optimizing a task distribution result through a tabu search algorithm to obtain an optimized task distribution result;
in the embodiment of the application, judging whether the task number of each robot in the task allocation result is larger than a preset threshold value; if yes, rearranging the tasks of the robot with the number of the tasks being larger than a preset threshold value through a tabu search algorithm to obtain an optimal task arrangement sequence, and optimizing a task distribution result based on the optimal task arrangement sequence to obtain an optimized task distribution result; if not, determining the task allocation result as the optimized task allocation result.
Step S104, based on the optimized task allocation result, the corresponding robot in the robot information starts to execute the task.
In the embodiment of the application, based on the optimized task allocation result, after the corresponding robot in the robot information starts to execute the task, when the task is incomplete due to the occurrence of the robot fault in the task execution process, the task which is not complete is selected as the task to be allocated, and the execution step S102 is returned.
It can be seen from the foregoing that, in the multi-robot scheduling method provided by the embodiment of the present application, road network vector diagram data, robot information and target task information of a patrol area are obtained through step S101, step S102, task allocation is performed on a robot corresponding to the robot information according to a preset allocated position inheritance rule, a preset auction mechanism and a preset task allocation standard, so as to obtain a task allocation result, step S103, the task allocation result is optimized through a tabu search algorithm, so as to obtain an optimized task allocation result, step S104, based on the optimized task allocation result, the corresponding robot in the robot information starts to execute tasks, thereby realizing scheduling of multiple robots, competing for the target points by different robots based on a commodity auction mechanism, allocating reasonable task points, and simultaneously reasonably distributing task execution sequences through a tabu search algorithm in a parallel computing manner, so as to improve system execution efficiency.
Referring to fig. 2, fig. 2 is another flowchart of a multi-robot scheduling method according to an embodiment of the application. The multi-robot scheduling method is used for scheduling a plurality of robots and comprises the following steps:
step S201, road network vector diagram data, robot information and target task information of a patrol area are obtained; the target task information comprises a plurality of small node tasks;
in the embodiment of the application, road network vector diagram data, robot information and target task information are acquired, wherein the robot information comprises an executable robot set, wherein ,r1 ，r 2 ，…，r n Representing an executable robot, the target task information including a small node task l m And target general task->, wherein ,l1 ，l 2 ，…，l m Representing small node tasks, wherein the target total task consists of a plurality of small node tasks.
In a specific implementation, the road network vector diagram data of the inspection area is a set formed by road network nodes, wherein the road network nodes roadNode are shown in the following formula:
；
wherein ,is the current road network node id with uniqueness,/->，/>，/>The coordinate values of x, y and z of the current road network node are respectively +.>Is a successor node taking the current road network node as a source node, and is +.>Is the course angle from the current road network node to the corresponding subsequent node.
Executable robot refers to a robot that can be normally connected to a server, and an executable robot setRefers to a collection of robots that can normally connect to a server.
Target general taskIs a small node task set expressed by a road network node id. Small node tasks (i.e., task points) are typically road network nodes.
Referring to fig. 3, fig. 3 is a schematic structural diagram of a multi-robot scheduling system provided by the embodiment of the application, wherein 301 is a robot, 302 is a road network and nodes, 303 is a shelf, the multi-robot scheduling system includes a central server, various robots 301 with obstacle avoidance and tracking functions, the central server is responsible for storing the road network, collecting task information, communicating and data interacting among the robots 301, and completing scheduling and other functions of the robots 301 based on current state information of the robots 301; the robot 301 can perform inspection, logistics distribution, and the like. The multi-robot scheduling system adopts a multi-robot scheduling method based on fusion of an auction mechanism and tabu search to complete tasks such as inspection or delivery.
Step S202, setting the running path data and the movement corner data required by the robot to complete the task as task execution cost indexes;
In the embodiment of the invention, the task execution cost can be indexes such as a robot running path, robot running time, robot movement corner and the like, and the robot task execution cost C is defined by taking the running path data and the movement corner data of the robot as the cost by considering the kinematic constraint of the robot.
Step S203, calculating task execution cost values of the corresponding robots in the robot information for completing the tasks of the plurality of small nodes according to the road network vector diagram data, the target task information and the preset task allocation standard;
in an alternative embodiment, step S203 includes:
step S2031, calculating running path data and movement corner data of a plurality of small node tasks completed by a corresponding robot in the robot information according to road network vector diagram data, target task information and preset task allocation standards;
step S2032, performing a weighted operation on the running path data and the movement corner data based on a preset weight coefficient, to obtain a task execution cost value for the robot to complete the tasks of the plurality of small nodes.
According to the method and the device, running path data and movement corner data of the robot for completing the tasks of the small nodes in the robot information are calculated according to road network vector diagram data, target task information and preset task allocation standards, and weighting operation is carried out on the running path data and the movement corner data based on preset weight coefficients, so that task execution cost values of the robot for completing the tasks of the small nodes are obtained.
In a specific implementation, the robots are respectively calculatedTask execution cost values (bids) to all inode tasks. The calculation formula of the running path data specifically comprises the following steps:
；
wherein ,for the path data of the robot, +.>Is a robot r i The position is located to the road network node +.>The shortest path is obtained based on an A-path searching algorithm (if a task point is assigned to the current robot in the previous round, the position of the shortest path is the position where the task point is completed in the previous round); it should be noted that, according to the assigned position inheritance rule, if the current robot has bid successfully in the last round of auction, it is necessary to perform cost calculation (bidding) again based on the execution of the last round of small node tasks, for example, assume that robot r 1 The previous round (or several previous rounds) of auction is assigned a small node task l 1 Then, in the current round of small node task I 2 In the auction of (a), robot r 1 Then task l is required at the small node 1 Bidding on the basis of (i) i.e. on the basis of the small node task/ 2 In the auction of (a), robot r 1 When bidding, calculating robot r 1 Task l at small node 1 Completing the task from the position to the small node task 2 Completing the task execution cost value of the position where the task is located if the robot r 1 No small node task/is assigned in the last (or previous) round of auctions 1 Task l is needed in small node if not needed 1 Bidding on the basis of (i) i.e. computing robot r 1 In original position to small node task 2 Completing the task execution cost value of the position where the task is located; />Is the robot at the current t moment +.>Is (are) the electric quantity of the battery>Is robot +.>The energy consumption of unit distance ensures that the robot can smoothly execute the task of the small node; if the cost is->Then indicate robot +.>Cannot complete the task of the small node->Bidding of the small node task is not participated; />Indicating that the robot cannot complete the small node task, else is bidding not to participate in the small node task.
Accordingly, for the current robot r i The position of the node is located to the road network node l j The travel distance of a shortest path, i.e. the pathThe calculation formula for calculating the movement corner data of the path is specifically as follows:
；
wherein ,for the movement angle data of the robot, +.>Is the heading angle of the current small node task of the robot on the path P, the range is +.> ，/>Is the heading angle of the robot for the next nodelet task on path P, the range is +.>The motion corner data is estimated by the accumulation of the rotation angles of the robot on all the subtask nodes of the path P.
The robot is obtained by weighting the running path data and the movement corner data of the robotTo the task pointTask execution cost value->The weighting formula is specifically as follows:
；
wherein ,is robot->To the task point->Auction bid (i.e. robot +.>To the task point->The task execution cost value of (2), K is a scaling factor,/->Is a weight coefficient, wherein ∈>。K、The setting can be preset according to the actual needs, for example, K=100,>but is not limited thereto.
Step S204, extracting the minimum value in the task execution cost value corresponding to each small node task based on a preset auction mechanism and a preset allocated position inheritance rule, and allocating the small node task corresponding to the minimum value to the robot corresponding to the minimum value to obtain a task allocation result;
in the embodiment of the application, task allocation is performed through a preset auction mechanism and a preset allocated position inheritance rule, the minimum value in the task execution cost value corresponding to the small node task is extracted, and the small node task is allocated to the robot corresponding to the minimum value, so that a task allocation result is obtained.
In a specific implementation, referring to fig. 4, fig. 4 is a schematic diagram of a task execution sequence of a robot according to an embodiment of the present application, where, The method is characterized in that a task execution sequence is adopted, and figures in square grids are small node tasks; selecting the minimum bid (the minimum value in the task execution cost value) of the current round +.>(robot to task Point->Minimum value in the task execution cost values) of the robot>And task Point->Task Point->Assigning to robots +.>Updating the robot according to the update task execution sequence formula>Task execution sequence->Target total task set->I.e. task of small node +.>Add robot->Task execution sequence->Little node task->From the target total task set->And (5) removing. The update task execution sequence formula is specifically:
；
；
referring to fig. 5, fig. 5 is a schematic process diagram of an auction mechanism according to an embodiment of the present application, in which 501 is a fully connected network, 502 is a first round of auction, 503 is a second round of auction, 504 is a third round of auction, r 1 、r 2 Is a robot, l 1 、l 2 、l 3 For a small node task, the dotted line segment corresponds to a number to represent the task execution cost from each robot to the current small node task, the arrow segment corresponds to a number to represent the minimum task execution cost from each robot to the current small node task, and the arrow represents the assignment of the current robot to execute the current small node task; repeating the above steps until the collection Empty (all small node tasks are allocated). It should be noted that when the robot is +_ in the course of making a new auction bid>The previous round of assignment (bidding) of the small node task +.>Robot->Is required to be at->On the basis of (a) a cost calculation (bid), i.e. +.>，/>Representation robot r i Task->Task l to small node j Is a task execution cost value.
Step S205, judging whether the number of tasks of each robot in the task allocation result is larger than a preset threshold value; if yes, go to step S206; if not, executing step S207;
in the embodiment of the invention, the robot in the task allocation result is obtainedCorresponding task execution sequence->Judging whether the task number of each robot in the task allocation result is greater than a preset threshold, for example, the preset threshold is set to be 1, and judging whether the task number of the robot is greater than 1, namely, the robot r i Task number->If yes, go to step S206; if not, step S207 is performed.
Step S206, rearranging the task execution sequence of the robot with the number of tasks larger than a preset threshold value through a tabu search algorithm to obtain an optimal task arrangement sequence, and optimizing a task distribution result based on the optimal task arrangement sequence to obtain an optimized task distribution result;
In the embodiment of the invention, a tabu search algorithm is applied to reorganize the task execution sequence of the robot with the number of tasks larger than a preset threshold (i.e. reorganize the sequence of the tasks in the task execution sequence), for example, when the preset threshold is set to 1, the task execution sequence of the robot with the number of tasks larger than 1 is reorganized to obtain an optimal task arrangement sequence, and based on the optimal task arrangement sequence, the task distribution result is optimized to obtain an optimized task distribution result.
In a specific implementation, a tabu search algorithm is utilized and is based on(the total cost of completing the task execution of all the task points by the robot is minimum) the task execution sequence is optimized by the target->Task rearrangement is performed to obtain a robot +.>New task execution sequence->(optimal task ordering)Column) based on the auction algorithm result +.>As a current robot->Is the initial solution of (1), i.e. the current solution->Current optimal solution->Setting the maximum iteration number Z of the algorithm, setting a tabu table H to be empty, and enabling the current iteration number Z to be=0;
the following iterative process is performed: judging whether the algorithm meets the termination condition z<Z, i.e. whether the current iteration number Z reaches the maximum iteration number Z, if so, then Ending the iteration, otherwise, continuing to execute the iteration process;
referring to fig. 6, fig. 6 is a schematic diagram illustrating generation of a domain solution set of a tabu search algorithm according to an embodiment of the present application, wherein 601 is a candidate solution generation, and numbers in squares are small node tasks. Each iteration process includes:
1) Constructing a current solution based on the rule N under the constraint of a tabu table HDomain solution set->I.e.. The rule N is +.>The constraint of the tabu list H means that the position exchange is not carried out on the exchange task point pairs in the tabu list H; if in the list of taboo HUnder the beam, lead to->And based on the scofflaw, the optimal state (optimal solution, namely optimal exchange task point) in the tabu table H is forbidden.
2) Based on all solutions S in the domain solution set S generated by the steps, S is a set of task execution sequences generated in the iterative process, S is a task execution sequence generated in the iterative process, and corresponding objective function values are calculatedSelecting the optimal solution x in the field solution set S, namely +.>，/>Representing arbitrary solution S within the field solution set S, let +.>Updating the tabu table H, generating a pair of switching task points corresponding to the optimal solution x (assuming +. >) Is added to the tabu table H as a constraint of the tabu table H. After updating, if->Then the optimal solution is updated, i.e. +.>Otherwise, the optimal solution is not updated. Said objective function->Refers to a robot r i And completing the journey of the task points according to the sequence of the task execution sequence s.
3) The current iteration number z=z+1.
Based on the tabu search algorithmFor optimization purposes is robot +>According to->The task sequence of (a) completes the path (cost) sum of all task points, wherein the path and the calculation cost are planned based on an A-based algorithm between the task points.
Step S207, determining that the task allocation result is the optimized task allocation result;
in the embodiment of the invention, if the number of tasks of the robot is smaller than or equal to the preset threshold, the preset threshold is assumed to be 1, that is, when the number of tasks of the robot is smaller than or equal to 1, the task allocation result is determined to be the optimal solution, and the task allocation result is determined to be the optimized task allocation result.
Step S208, based on the optimized task allocation result, the corresponding robot in the robot information starts to execute the task;
in an alternative embodiment, after step S208, the method further includes:
when the task is incomplete due to the robot fault in the task execution process, the task which is not complete is selected as the task to be distributed, and the execution step S202 is returned.
In the embodiment of the application, the corresponding robot is assigned based on the optimized task allocation resultStarting to execute the task when the task execution process robot +.>Failure results in a new task execution sequence +.>Incomplete, new task execution sequence +.>Residue in (a)The residual tasks are stored in the target total task set +.>Returning to step S202, if the system receives a new task point + ->Or task set->Storing the task in the target total task set +.>Step S202 is executed back.
Referring to fig. 7, fig. 7 is a schematic structural diagram of a multi-robot scheduling device according to an embodiment of the present application, configured to schedule a plurality of robots, where the multi-robot scheduling device includes:
the acquisition module 701 is configured to acquire road network vector diagram data, robot information and target task information of a patrol area;
the allocation module 702 is configured to perform task allocation on a corresponding robot in the robot information according to the road network vector diagram data and the target task information, and by combining a preset allocated position inheritance rule, a preset auction mechanism and a preset task allocation standard, to obtain a task allocation result;
the optimizing module 703 is configured to optimize the task allocation result through a tabu search algorithm, and obtain an optimized task allocation result;
And the execution module 704 is configured to start executing the task by the corresponding robot in the robot information based on the optimized task allocation result.
The multi-robot scheduling device provided by the embodiment of the application can realize the scheduling of robots, bid on target points by different robots by referring to a commodity auction mechanism, reasonable task points are distributed, and meanwhile, the task execution sequence is reasonably distributed by a tabu search algorithm in a parallel computing mode, so that the system execution efficiency is improved.
In some embodiments, the target task information includes a plurality of small node tasks; the allocation module 702 includes:
the setting sub-module is used for setting the running path data and the movement corner data required by the robot to finish the task as task execution cost indexes;
the calculation sub-module is used for calculating the task execution cost value of the corresponding robot in the robot information for completing the tasks of the plurality of small nodes according to the road network vector diagram data, the target task information and the preset task allocation standard;
and the allocation sub-module is used for extracting the minimum value in the task execution cost value corresponding to each small node task based on a preset auction mechanism and an allocated position inheritance rule, and allocating the small node task corresponding to the minimum value to the robot corresponding to the minimum value to obtain a task allocation result.
The multi-robot scheduling device provided by the embodiment of the application can realize the scheduling of robots, and by virtue of a commodity auction mechanism, target points are bid through different robots, so that reasonable task points are distributed.
In some embodiments, the computing submodule includes:
the calculation unit is used for calculating the running path data and the movement corner data of a plurality of small nodes of the corresponding robot in the robot information according to the road network vector diagram data, the target task information and the preset task allocation standard;
the weighting unit is used for carrying out weighting operation on the running path data and the movement corner data based on a preset weight coefficient to obtain a task execution cost value for the robot to finish a plurality of small node tasks.
In some embodiments, the optimization module 703 includes:
the judging sub-module is used for judging whether the number of tasks of each robot in the task allocation result is larger than a preset threshold value; if yes, triggering the optimization submodule to execute the corresponding steps; if not, triggering the determination submodule to execute the corresponding step;
the optimizing sub-module is used for rearranging the task execution sequences of the robots with the task numbers larger than a preset threshold value through a tabu search algorithm to obtain an optimal task arrangement sequence, optimizing the task distribution result based on the optimal task arrangement sequence, and obtaining an optimized task distribution result;
And the determining submodule is used for determining the task distribution result as the optimized task distribution result.
In some embodiments, the apparatus further comprises:
and the return module is used for selecting the unfinished task as the task to be distributed when the task is unfinished due to the robot fault in the task execution process, and triggering the distribution module 702 to execute the corresponding steps.
It can be seen from the above that the multi-robot scheduling device provided by the embodiment of the application can realize the scheduling of multiple robots, and by referring to the commodity auction mechanism, the target points are bid by different robots, reasonable task points are allocated, and meanwhile, by means of parallel calculation, the task execution sequence is reasonably arranged by a tabu search algorithm, so that the system execution efficiency is improved.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
Further, the units described as separate units may or may not be physically separate, and units displayed as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
Furthermore, functional modules in various embodiments of the present application may be integrated together to form a single portion, or each module may exist alone, or two or more modules may be integrated to form a single portion.
In this document, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.
The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and variations will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims (10)

1. A multi-robot scheduling method, comprising:
step S1, road network vector diagram data, robot information and target task information of a patrol area are obtained;
step S2, according to the road network vector diagram data and the target task information, combining a preset allocated position inheritance rule, a preset auction mechanism and a preset task allocation standard, and taking running path data and movement corner data required by a robot to finish a task as task execution cost indexes, performing task allocation on the corresponding robot in the robot information to obtain a task allocation result;
step S3, optimizing the task distribution result through a tabu search algorithm to obtain an optimized task distribution result;
step S4, based on the optimized task allocation result, the corresponding robot in the robot information starts to execute the task;
the preset assigned position inheritance rule is that in the process of performing a new round of auction bidding, if any robot is assigned with a small node task in one round or the previous rounds, the cost from the position where the robot finishes the task in the previous round or the previous rounds to the position where the task finishes the new round of auction is calculated; if any robot is not assigned with the small node task in the previous round or the previous rounds of auction, calculating the task execution cost value from the original position of the robot to the position where the new round of small node task is completed;
The calculation formula of the running path data specifically comprises the following steps:
；
the calculation formula of the movement rotation angle data specifically comprises the following steps:
；
the calculation formula of the task execution cost value is specifically as follows:
；
wherein ,for the path data of the robot, +.>Is the task of robot ri to the small node>Is a shortest path travel distance,/-)>Is the robot at the current t moment +.>Is (are) the electric quantity of the battery>Is robot +.>Energy consumption per unit distance, ensuring that the robot can smoothly execute the task of the small node +.>The method comprises the steps of carrying out a first treatment on the surface of the If the cost is->Then indicate robot +.>Cannot complete the task of the small node->Bidding of the small node task is not participated; />Indicating that the robot cannot complete the small node task, else is bidding not to participate in the small node task, ++>For the movement angle data of the robot, +.>Is the heading angle of the current small node task of the robot on the path P, the range is +.>，/>Is the heading angle of the robot for the next nodelet task on path P, the range is +.>，/>Is robot->To the task point->Auction bid of (2), the auction bid is robot +.>To the task point->Is a scaling factor,/->Is a weight coefficient, wherein ∈>。
2. The multi-robot scheduling method of claim 1, wherein the target task information comprises a plurality of small node tasks; the step S2 includes:
Step S21, setting the running path data and the movement corner data required by the robot to complete the task as task execution cost indexes;
step S22, calculating task execution cost values of the robots corresponding to the robot information for completing the tasks of the plurality of small nodes according to the road network vector diagram data, the target task information and the preset task allocation standard;
step S23, extracting the minimum value of task execution cost values corresponding to all small node tasks based on the preset auction mechanism and the allocated position inheritance rule, and allocating the small node tasks corresponding to the minimum value to the robots corresponding to the minimum value to obtain the task allocation result.
3. The multi-robot scheduling method according to claim 2, wherein the step S22 comprises:
step S221, calculating running path data and movement corner data of the corresponding robot in the robot information for completing the tasks of the plurality of small nodes according to the road network vector diagram data, the target task information and the preset task allocation standard;
step S222, based on a preset weight coefficient, carrying out weighted operation on the running path data and the movement corner data to obtain a task execution cost value of the robot for completing the tasks of the plurality of small nodes.
4. The multi-robot scheduling method according to claim 1, wherein the step S3 comprises:
step S31, judging whether the task number of each robot in the task allocation result is larger than a preset threshold value; if yes, go to step S32; if not, executing step S33;
step S32, rearranging task execution sequences of robots with the number of tasks being greater than the preset threshold value through the tabu search algorithm to obtain an optimal task arrangement sequence, and optimizing the task distribution result based on the optimal task arrangement sequence to obtain the optimized task distribution result;
and step S33, determining the task allocation result as the optimized task allocation result.
5. The multi-robot scheduling method according to claim 1, further comprising, after the step S4:
when the task is incomplete due to the robot fault in the task execution process, the incomplete task is selected as the task to be distributed, and the step S2 is executed in a return mode.
6. A multi-robot scheduling apparatus, comprising:
the acquisition module is used for acquiring road network vector diagram data, robot information and target task information of the inspection area;
The allocation module is used for allocating the tasks of the corresponding robots in the robot information by taking the running path data and the movement corner data required by the robots to finish the tasks as task execution cost indexes according to the road network vector diagram data and the target task information and combining a preset allocated position inheritance rule, a preset auction mechanism and a preset task allocation standard, so as to obtain a task allocation result;
the optimization module is used for optimizing the task distribution result through a tabu search algorithm to obtain an optimized task distribution result;
the execution module is used for starting to execute tasks for the corresponding robots in the robot information based on the optimized task allocation result;
the preset assigned position inheritance rule is that in the process of performing a new round of auction bidding, if any robot is assigned with a small node task in one round or the previous rounds, the cost from the position where the robot finishes the task in the previous round or the previous rounds to the position where the task finishes the new round of auction is calculated; if any robot is not assigned with the small node task in the previous round or the previous rounds of auction, calculating the task execution cost value from the original position of the robot to the position where the new round of small node task is completed;
The calculation formula of the running path data specifically comprises the following steps:
；
the calculation formula of the movement rotation angle data specifically comprises the following steps:
；
the calculation formula of the task execution cost value is specifically as follows:
；
wherein ,for the path data of the robot, +.>Is the task of robot ri to the small node>Is a shortest path travel distance,/-)>Is the robot at the current t moment +.>Is (are) the electric quantity of the battery>Is robot +.>Energy consumption per unit distance, ensuring that the robot can smoothly execute the task of the small node +.>The method comprises the steps of carrying out a first treatment on the surface of the If the cost is->Then indicate robot +.>Cannot complete the task of the small node->Bidding of the small node task is not participated; />Indicating that the robot cannot complete the small node task, else is bidding not to participate in the small node task, ++>For the movement angle data of the robot, +.>Is the heading angle of the current small node task of the robot on the path P, the range is +.>，/>Is the heading angle of the robot for the next nodelet task on path P, the range is +.>，/>Is robot->To the task point->Auction bid of (2), the auction bid is robot +.>To the task point->Is a scaling factor,/->Is a weight coefficient, wherein ∈>。
7. The multi-robot scheduling device of claim 6, wherein the target task information comprises a plurality of small node tasks; the distribution module comprises:
The setting sub-module is used for setting the running path data and the movement corner data required by the robot to finish the task as task execution cost indexes;
the calculation sub-module is used for calculating the task execution cost value of the corresponding robot in the robot information for completing the tasks of the plurality of small nodes according to the road network vector diagram data, the target task information and the preset task allocation standard;
and the allocation sub-module is used for extracting the minimum value in the task execution cost value corresponding to each small node task based on the preset auction mechanism and the allocated position inheritance rule, and allocating the small node task corresponding to the minimum value to the robot corresponding to the minimum value to obtain the task allocation result.
8. The multi-robot scheduling device of claim 7, wherein the computing sub-module comprises:
the calculation unit is used for calculating the running path data and the movement corner data of the corresponding robots in the robot information for completing the tasks of the plurality of small nodes according to the road network vector diagram data, the target task information and the preset task allocation standard;
and the weighting unit is used for carrying out weighting operation on the running path data and the movement corner data based on a preset weight coefficient to obtain a task execution cost value of the robot for completing the tasks of the plurality of small nodes.
9. The multi-robot scheduling device of claim 6, wherein the optimization module comprises:
the judging sub-module is used for judging whether the task number of each robot in the task allocation result is larger than a preset threshold value; if yes, triggering the optimization submodule to execute the corresponding steps; if not, triggering the determination submodule to execute the corresponding step;
the optimizing sub-module is used for rearranging task execution sequences of robots with the number of tasks being larger than the preset threshold value through the tabu search algorithm to obtain an optimal task arrangement sequence, and optimizing the task distribution result based on the optimal task arrangement sequence to obtain the optimized task distribution result;
and the determining submodule is used for determining the task distribution result as the optimized task distribution result.
10. The multi-robot scheduling device of claim 6, further comprising:
and the return module is used for selecting the unfinished task as the task to be distributed when the task is unfinished due to the robot fault in the task execution process, and triggering the distribution module to execute the corresponding steps.
CN202310503813.1A 2023-05-06 2023-05-06 Multi-robot scheduling method and device Active CN116245257B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310503813.1A CN116245257B (en) 2023-05-06 2023-05-06 Multi-robot scheduling method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310503813.1A CN116245257B (en) 2023-05-06 2023-05-06 Multi-robot scheduling method and device

Publications (2)

Publication Number Publication Date
CN116245257A CN116245257A (en) 2023-06-09
CN116245257B true CN116245257B (en) 2023-09-12

Family

ID=86624602

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310503813.1A Active CN116245257B (en) 2023-05-06 2023-05-06 Multi-robot scheduling method and device

Country Status (1)

Country Link
CN (1) CN116245257B (en)

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103279793A (en) * 2013-04-25 2013-09-04 北京航空航天大学 Task allocation method for formation of unmanned aerial vehicles in certain environment
CN104155999A (en) * 2014-07-31 2014-11-19 海南大学 Time-sensitive task dynamic allocation algorithm in battlefield environment for multiple unmanned aerial vehicles
CN106875090A (en) * 2017-01-09 2017-06-20 中南大学 A kind of multirobot distributed task scheduling towards dynamic task distributes forming method
CN109839933A (en) * 2019-02-20 2019-06-04 哈尔滨工程大学 A kind of multi-robot Task Allocation based on VDSOM algorithm
CN110264062A (en) * 2019-08-12 2019-09-20 南京邮电大学 Distributed more AGV dynamic task allocations and its paths planning method and system
CN111716356A (en) * 2020-06-18 2020-09-29 南京邮电大学 Method for cooperation of humanoid multi-robot
CN112591568A (en) * 2020-12-15 2021-04-02 海南大学 Device sharing method for fusion of data and information facing intention calculation and reasoning
CN112733421A (en) * 2020-12-01 2021-04-30 南京航空航天大学 Task planning method for cooperative fight against earth by unmanned aerial vehicle
CN112785132A (en) * 2021-01-14 2021-05-11 北京理工大学 Task allocation method for multi-robot mobile shelf for intelligent warehouse
CN115186878A (en) * 2022-06-22 2022-10-14 湖南大学无锡智能控制研究院 Multi-AGV online task allocation method and system
CN115796530A (en) * 2022-12-05 2023-03-14 广东工业大学 Robot logistics scheduling optimization method for simultaneously delivering and taking goods
CN115952942A (en) * 2023-03-13 2023-04-11 季华实验室 Multi-target inspection scheduling planning method and device, electronic equipment and storage medium

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022021119A1 (en) * 2020-07-29 2022-02-03 浙江大学 Method and system for fully autonomous waterborne transport scheduling between container terminals

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103279793A (en) * 2013-04-25 2013-09-04 北京航空航天大学 Task allocation method for formation of unmanned aerial vehicles in certain environment
CN104155999A (en) * 2014-07-31 2014-11-19 海南大学 Time-sensitive task dynamic allocation algorithm in battlefield environment for multiple unmanned aerial vehicles
CN106875090A (en) * 2017-01-09 2017-06-20 中南大学 A kind of multirobot distributed task scheduling towards dynamic task distributes forming method
CN109839933A (en) * 2019-02-20 2019-06-04 哈尔滨工程大学 A kind of multi-robot Task Allocation based on VDSOM algorithm
CN110264062A (en) * 2019-08-12 2019-09-20 南京邮电大学 Distributed more AGV dynamic task allocations and its paths planning method and system
CN111716356A (en) * 2020-06-18 2020-09-29 南京邮电大学 Method for cooperation of humanoid multi-robot
CN112733421A (en) * 2020-12-01 2021-04-30 南京航空航天大学 Task planning method for cooperative fight against earth by unmanned aerial vehicle
CN112591568A (en) * 2020-12-15 2021-04-02 海南大学 Device sharing method for fusion of data and information facing intention calculation and reasoning
CN112785132A (en) * 2021-01-14 2021-05-11 北京理工大学 Task allocation method for multi-robot mobile shelf for intelligent warehouse
CN115186878A (en) * 2022-06-22 2022-10-14 湖南大学无锡智能控制研究院 Multi-AGV online task allocation method and system
CN115796530A (en) * 2022-12-05 2023-03-14 广东工业大学 Robot logistics scheduling optimization method for simultaneously delivering and taking goods
CN115952942A (en) * 2023-03-13 2023-04-11 季华实验室 Multi-target inspection scheduling planning method and device, electronic equipment and storage medium

Also Published As

Publication number Publication date
CN116245257A (en) 2023-06-09

Similar Documents

Publication Publication Date Title
Liang et al. Mobility-aware charging scheduling for shared on-demand electric vehicle fleet using deep reinforcement learning
Liu et al. Distributed and energy-efficient mobile crowdsensing with charging stations by deep reinforcement learning
Arwa et al. Reinforcement learning techniques for optimal power control in grid-connected microgrids: A comprehensive review
Abdullah et al. Reinforcement learning based EV charging management systems–a review
CN110632907B (en) Scheduling optimization method and system for distributed assembly type replacement flow shop
Zhang et al. Intelligent electric vehicle charging recommendation based on multi-agent reinforcement learning
Shen et al. Mathematical modeling and multi-objective evolutionary algorithms applied to dynamic flexible job shop scheduling problems
CN107895225B (en) Multi-Agent conflict-free cooperative task allocation method
US20210319384A1 (en) Blockchain-Based Energy Interaction Apparatus, Energy Internet System and Interaction Method
Gao et al. A multiagent competitive bidding strategy in a pool-based electricity market with price-maker participants of WPPs and EV aggregators
CN113671987B (en) Multi-machine distributed time sequence task allocation method based on non-deadlock contract net algorithm
Bogyrbayeva et al. A reinforcement learning approach for rebalancing electric vehicle sharing systems
Kiaee Integration of electric vehicles in smart grid using deep reinforcement learning
Lei et al. Large-scale dynamic scheduling for flexible job-shop with random arrivals of new jobs by hierarchical reinforcement learning
Liu et al. An efficient ant colony system for multi-robot task allocation with large-scale cooperative tasks and precedence constraints
CN116245257B (en) Multi-robot scheduling method and device
Kang et al. Multi-Objective Teaching-Learning-Based Optimizer for a Multi-Weeding Robot Task Assignment Problem
Yu et al. Extended-cbba-based task allocation algorithm for on-orbit assembly spacecraft
Zhang et al. Dynamic scheduling for flexible job shop with insufficient transportation resources via graph neural network and deep reinforcement learning
Leite et al. Solving an energy resource management problem with a novel multi-objective evolutionary reinforcement learning method
CN109886574B (en) Multi-robot task allocation method based on improved threshold method
CN115952942B (en) Multi-target inspection scheduling planning method and device, electronic equipment and storage medium
Mutingi et al. Home Health Care staff scheduling: Effective grouping approaches
Wu et al. A deep reinforcement learning model for dynamic job-shop scheduling problem with uncertain processing time
Hao et al. Improved algorithms to plan missions for agile earth observation satellites

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