CN112070328A - Multi-water-surface unmanned search and rescue boat task allocation method with known environmental information part - Google Patents

Multi-water-surface unmanned search and rescue boat task allocation method with known environmental information part Download PDF

Info

Publication number
CN112070328A
CN112070328A CN201910499052.0A CN201910499052A CN112070328A CN 112070328 A CN112070328 A CN 112070328A CN 201910499052 A CN201910499052 A CN 201910499052A CN 112070328 A CN112070328 A CN 112070328A
Authority
CN
China
Prior art keywords
task
search
rescue
boat
auction
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
CN201910499052.0A
Other languages
Chinese (zh)
Other versions
CN112070328B (en
Inventor
黄海滨
庄宇飞
李亚南
梁景凯
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.)
Weihai Tianfan Intelligent Technology Co ltd
Harbin Institute of Technology Weihai
Original Assignee
Weihai Tianfan Intelligent Technology Co ltd
Harbin Institute of Technology Weihai
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 Weihai Tianfan Intelligent Technology Co ltd, Harbin Institute of Technology Weihai filed Critical Weihai Tianfan Intelligent Technology Co ltd
Priority to CN201910499052.0A priority Critical patent/CN112070328B/en
Publication of CN112070328A publication Critical patent/CN112070328A/en
Application granted granted Critical
Publication of CN112070328B publication Critical patent/CN112070328B/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
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • G06Q50/265—Personal security, identity or safety
    • 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
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

The invention relates to a task allocation method of a multi-water-surface unmanned search and rescue boat with partially known environmental information, aiming at solving the problems that under the condition that the sensing radius and the communication radius of the unmanned boat are limited and only the environmental information of a working area is partially mastered, develops an on-line task allocation technology of the unmanned search and rescue boat on the heterogeneous water surface, improves a sequential auction algorithm, for the task to be executed which is sensed on line by the search and rescue boat, firstly, the matching degree of the operation capacity of each boat and the task is measured to determine a candidate task sequence, then, the operation income of the associated unmanned ship is evaluated according to the task to be shot in the current round, the candidate execution ship sequence is updated, and finally, the method and the system have the advantages that the auction is executed with the minimized task energy consumption as the target, and the current round of tasks are executed by the medium-sized shooter.

Description

Multi-water-surface unmanned search and rescue boat task allocation method with known environmental information part
The technical field is as follows:
the invention belongs to the task allocation category of unmanned system clusters, and particularly relates to a task allocation method of an unmanned surface search and rescue boat cluster under the condition that a part of working environment is known.
Background art:
the actual maritime search and rescue task is often wide in operation water area range, complex in environment, long in operation time and tedious in process. Therefore, in recent years, with the increasing autonomy of unmanned systems, unmanned water search and rescue boats are gradually replacing manual work in maritime search and rescue tasks, and are widely applied. Compared with a single boat, the unmanned boat cluster has the advantages that a plurality of boats cooperatively execute tasks, the unmanned boat cluster has distribution characteristics in time, space, information, resources and functions, meanwhile, the efficiency and fault tolerance of execution of search and rescue tasks can be greatly improved, and the unmanned boat cluster is the mainstream development trend of future maritime search and rescue applications.
In the process of carrying out autonomous search and rescue on a plurality of targets by a plurality of unmanned boats, the plurality of individuals cannot be simply combined together in consideration of parallelism on individual behaviors and some emergency situations, and a reasonable task allocation mechanism needs to be introduced, so that the advantages of the unmanned boats are fully exerted, and the established targets are efficiently finished.
At present, a task allocation method of an unmanned system cluster mainly abstracts a problem into an optimization problem which can be processed by a computer, and solves the problem by adopting an intelligent optimization algorithm. Although the method is widely applied, for the optimization problem of large amount, the convergence rate of the algorithm is slow, and the probability of obtaining a reasonable optimization result is low. Therefore, the invention fully considers the difference of different search and rescue boat capabilities under the condition that the operation environment information is only partially known by using the basic idea of the sequential auction algorithm, and realizes the efficient and reasonable distribution of tasks by optimizing the total task loss.
The invention content is as follows:
in order to overcome the defects of the prior art, the invention provides a multi-water-surface unmanned search and rescue boat task allocation technology based on an auction algorithm. On the basis of a traditional sequential auction algorithm, the matching degree between the type of the task to be executed and the operation capacity of the search and rescue boat and the specific difference of the physical properties of each boat are comprehensively considered, so that the on-line operation of the task allocation process is realized, the finally obtained task allocation scheme is more reasonable, and the actual requirements are better met.
The invention adopts the following specific technical scheme:
step 1; initializing, loading known operation sea area part environment information:
wherein the number of tasks to be performed: set to m, marked as set T ═ T1,t2,...,tm}；
Type of task to be performed: by means of variables
Figure BDA0002089578810000011
Characterisation of the execution of the jth task tjThe variable is only related to the type of the executed task and is not related to the performance of the search and rescue boat executing the task;
the number of the search and rescue boats is as follows: set to n, and mark as set U ═ U1,u2,...,un}；
The operation capability of the search and rescue boat is as follows: the ith search and rescue boat uiN implements the jth task t, i 1,2jThe capability of j 1, 2.. multidot.m is set as
Figure BDA0002089578810000021
And order
Figure BDA0002089578810000022
When in use
Figure BDA0002089578810000023
Time, show search and rescue boat uiNot having to perform task tjThe ability of (c);
setting the initial task sub-sequence of each search and rescue boat as an empty set;
step 2: setting a sampling time interval delta T, collecting the information of the carried sensors by each search and rescue boat every delta T, and inserting the detected task to be executed into the task subsequence of the corresponding boat;
and step 3: executing internal auction, marking all tasks to be auctioned in the current round as t'1,t′2...t′pP is less than or equal to m and aims at each task to be photographed t'k1, 2.. p performs the following steps:
step 3-1: will and task t′kAnd marking all candidate search and rescue boat sets corresponding to p as 1,2
Figure BDA0002089578810000024
I.e. the principal wheel has qkTask t 'is detected by boat'kAnd the boats can communicate with each other, and the complementary set is defined as
Figure BDA0002089578810000025
Corresponding to no task t 'detected in the current wheel'kThe search and rescue boat set;
step 3-2: for arbitrary uks∈Uk,s＝1,2,...,qkCalculating all search and rescue boats uiE.u, 1 ═ 1,2ksInternal auction bid value of
Figure BDA0002089578810000026
Wherein the content of the first and second substances,
Figure BDA0002089578810000027
corresponding to completion of task t'kThe price to be paid for is that,
Figure BDA0002089578810000028
corresponding boat uksImplementation of task t'kThe capability of (a) to (b),
Figure BDA0002089578810000029
b is a constant parameter set by a user and represents the current unknown search and rescue boat uiAnd task t'kThe geometric distance between the two parts of the frame,
Figure BDA00020895788100000210
corresponding search and rescue boat uksAnd to-be-executed task t'kThe geometric distance between the two is larger than 0, and the user-defined smaller constant value parameter is obtained;
step 3-3: if it is
Figure BDA00020895788100000211
Then u is reservedksIn candidate search and rescue boat set UkThe medium is unchanged; otherwise, if C is equal to C, updating the candidate search and rescue boat set UkBy uiInstead of uks；
And 4, step 4: candidate search and rescue boat set U corresponding to p tasks to be executed in current round after updating1,U2,...,UpAnd taking the task corresponding to the set with the most contained elements as the auction task of the current round and marking the task as the auction task of the current round
Figure BDA00020895788100000212
(subscript f corresponds to the current auction as round f);
and 5: external auction and local round task to be auctioned
Figure BDA0002089578810000031
Marking the corresponding candidate search and rescue boat set as
Figure BDA0002089578810000032
Step 5-1: for all search and rescue boats
Figure BDA0002089578810000033
Calculate it and task
Figure BDA0002089578810000034
Geometric distance between them
Figure BDA0002089578810000035
Then determining the search and rescue boat for the current round of host auction
Figure BDA0002089578810000036
Figure BDA0002089578810000037
And at the same time, as the communication node of the auction round,
Figure BDA0002089578810000038
responsible statistics aggregation
Figure BDA0002089578810000039
The bidding conditions of all the search and rescue boats are shown in
Figure BDA00020895788100000310
Broadcasting the final auction result within the range;
step 5-2: the performance index used in the auction process is
Figure BDA00020895788100000311
Figure BDA00020895788100000312
Auction task allocation to search and rescue boat for representing h round
Figure BDA00020895788100000313
Otherwise
Figure BDA00020895788100000314
Variables of
Figure BDA00020895788100000315
Wherein, variable
Figure BDA00020895788100000316
Corresponding search and rescue boat
Figure BDA00020895788100000317
Run to task on h-th round
Figure BDA00020895788100000318
The energy consumed at the location;
step 5-3:
Figure BDA00020895788100000319
corresponding search and rescueThe boat is the final winner of the auction in the current round;
step 6: and updating the task sequence to be executed, and removing the completed tasks from the task sequence. And (5) if the task sequence to be executed is not empty, returning to the step (2), otherwise, ending the algorithm process.
The invention has the following beneficial technical effects:
most of traditional auction algorithms are only suitable for the task allocation problem of homogeneous cluster systems, and heterogeneous systems with incomplete prior knowledge, unmanned system communication and limited detection radius are often low in operation efficiency and unreasonable in allocation scheme. The method improves the bidding strategy and the specific auction mode of each round of auction based on the traditional sequential auction algorithm. Meanwhile, the difference of respective capabilities of the heterogeneous unmanned ship clusters is considered. By repeatedly allocating tasks on line in real time, the efficiency and the rationality of task allocation are improved.
Description of the drawings:
FIG. 1 is a flow chart of the present invention.
FIG. 2 is a schematic diagram of a task allocation scheme for an isomorphic scenario according to the present invention.
Fig. 3 is a schematic diagram of a task allocation scheme in a heterogeneous scenario according to the present invention.
Fig. 4 is a schematic diagram of a task allocation scheme under the condition that the communication radius and the perception radius of the search and rescue boat are limited.
The specific implementation mode is as follows:
in order to make the objects, technical solutions and advantages of the present invention more clear and clearer, the following detailed description of the embodiments of the present invention is made with reference to the accompanying drawings and examples. The specific embodiments described herein are merely illustrative of the invention and are not intended to be limiting. In the simulation process, the emphasis is placed on the verification of the performance of the task allocation algorithm, so that the mathematical model of the unmanned surface rescue boat adopts a linear second-order system. The simulation environment is as follows: intel 3.00GHz main frequency, 4.00GB memory PC, Windows 7 operating system, Matlab R2012a simulation platform.
The algorithm flow chart of the invention is shown in FIG. 1, and the detailed steps are as follows:
step 1: initialization: loading the known environment information of the part of the working sea area,
step 1-1: setting the number of tasks to be executed: set to m, marked as set T ═ T1,t2,...,tm}；
Step 1-2: type of task to be performed: by means of variables
Figure BDA0002089578810000041
Characterisation of the execution of the jth task tjThe variable is only related to the type of the executed task and is not related to the performance of the search and rescue boat executing the task;
step 1-3: the number of the search and rescue boats is as follows: set to n, and mark as set U ═ U1,u2,...,un}；
Step 1-4: the operation capability of the search and rescue boat is as follows: the ith search and rescue boat uiN implements the jth task t, i 1,2jThe capability of j 1, 2.. multidot.m is set as
Figure BDA0002089578810000042
And order
Figure BDA0002089578810000043
When in use
Figure BDA0002089578810000044
Time, show search and rescue boat uiNot having to perform task tjThe ability of (c); setting the initial task sub-sequence of each search and rescue boat as an empty set;
n is usually required to be less than or equal to m, and the working capacity of the search and rescue boat in the steps 1-4 is convenient for subsequent application
Figure BDA0002089578810000045
Normalization processing is carried out, and during specific implementation, the task of each round of auction at least needs to ensure that the working capacity value of one boat is not zero; meanwhile, the operation capacity of a certain search and rescue boat in the cluster can be set to be 1 aiming at the designated task, namely the task is designated to be implemented only by the search and rescue boat with the operation capacity of 1.
Step 2: setting a sampling time interval delta T, collecting the information of the carried sensors by each search and rescue boat every delta T, and inserting the detected task to be executed into the task subsequence of the corresponding boat; the selection of the sampling time value has great influence on the actual system performance; if the value of Δ T is too small, the performance of the navigation system carried by each boat and the communication bandwidth between each boat need to be correspondingly improved, and conversely, if the value of Δ T is too large, the detection missing phenomenon may occur, and the final distribution scheme is poor.
And step 3: internal auction, marking all tasks to be auctioned in the current round as t'1,t′2...t′pP is less than or equal to m and aims at each task to be photographed t'k1, 2.. p performs the following steps:
step 3-1: will be with task t'kAnd marking all candidate search and rescue boat sets corresponding to p as 1,2
Figure BDA0002089578810000046
I.e. the principal wheel has qkTask t 'is detected by boat'kAnd the boats can communicate with each other, and the complementary set is defined as
Figure BDA0002089578810000051
Corresponding to no task t 'detected in the current wheel'kThe search and rescue boat set;
for arbitrary uks∈Uk,s＝1,2,...,qkCalculating all search and rescue boats uiE.u, 1 ═ 1,2ksInternal auction bid value of
Figure BDA0002089578810000052
Wherein the content of the first and second substances,
Figure BDA0002089578810000053
corresponding to completion of task t'kThe price to be paid for is that,
Figure BDA0002089578810000054
corresponding boat uksImplementation of task t'kThe capability of (a) to (b),
Figure BDA0002089578810000055
b is a constant parameter set by a user and represents the current unknown search and rescue boat uiAnd task t'kThe geometric distance between the two parts of the frame,
Figure BDA0002089578810000056
corresponding search and rescue boat uksAnd to-be-executed task t'kThe geometric distance between the two is larger than 0, and the user-defined smaller constant value parameter is obtained;
if it is
Figure BDA0002089578810000057
I.e. uiAt uksOutside the communication radius. Therefore, uksSearch and rescue boat u can not be directly acquirediThe value of the parameter B can be determined only in a prediction mode at the current specific position. In practical application, the value of B is larger than the upper limit of the communication radius of all the operation boats.
Step 3-2 is used for judging the search and rescue boat uksIn particular, whether the task of the current round t 'should be participated'kOr give up to search for other tasks that are subsequently more appropriate to their job capabilities. Meanwhile, to avoid task t'kAll corresponding candidate search and rescue boats
Figure BDA0002089578810000058
All failures in the internal auction occur, and the value of the constant parameter C must be greater than zero. Further, to ensure
Figure BDA0002089578810000059
And there is a significant difference in the value and therefore will be set to a sufficiently small normal value parameter.
Step 3-3: if it is
Figure BDA00020895788100000510
Then u is reservedksIn candidate search and rescue boat set UkThe medium is unchanged; otherwise, if C is equal to C, then the updating is waitingSelect search for and rescue ship set UkBy uiInstead of uks；
And 4, step 4: candidate search and rescue boat set U corresponding to p tasks to be executed in current round after updating1,U2,...,UpAnd taking the task corresponding to the set with the most contained elements as the auction task of the current round and marking the task as the auction task of the current round
Figure BDA00020895788100000511
(subscript f corresponds to the current auction as round f);
and 5: external auction and local round task to be auctioned
Figure BDA00020895788100000512
Marking the corresponding candidate search and rescue boat set as
Figure BDA00020895788100000513
Step 5-1: for all search and rescue boats
Figure BDA00020895788100000514
Calculate it and task
Figure BDA00020895788100000515
Geometric distance between them
Figure BDA00020895788100000516
Then determining the search and rescue boat for the current round of host auction
Figure BDA00020895788100000517
Figure BDA00020895788100000518
And at the same time, as the communication node of the auction round,
Figure BDA00020895788100000519
responsible statistics aggregation
Figure BDA00020895788100000520
The bidding conditions of all the search and rescue boats are shown in
Figure BDA00020895788100000521
Broadcasting the final auction result within the range;
step 5-2: the performance index used in the auction process is
Figure BDA0002089578810000061
Figure BDA0002089578810000062
Auction task allocation to search and rescue boat for representing h round
Figure BDA0002089578810000063
Otherwise
Figure BDA0002089578810000064
Variables of
Figure BDA0002089578810000065
Wherein, variable
Figure BDA0002089578810000066
Corresponding search and rescue boat
Figure BDA0002089578810000067
Run to task on h-th round
Figure BDA0002089578810000068
The energy consumed at the location;
in practical application, the search and rescue boat usually moves at a constant speed, so that the performance index
Figure BDA0002089578810000069
Can be used for characterizing the search and rescue boat
Figure BDA00020895788100000610
All energy consumed by the auction through round f (including round f).
Step 5-3:
Figure BDA00020895788100000611
the corresponding search and rescue boat is the final winner of the auction of the current round;
step 6: and updating the task sequence to be executed, and removing the completed tasks from the task sequence. And (5) if the task sequence to be executed is not empty, returning to the step (2), otherwise, ending the algorithm process.
In the algorithm implementation process, the search and rescue boat stores the completed task information and the task information detected by the current wheel, and simultaneously stores the positions of all other search and rescue boats and the completed task information within the communication range of the search and rescue boat. And periodically updating the information by acquiring the information of the on-board sensor at each sampling moment. In order to reduce the communication burden, the algorithm does not adopt a real-time communication mode, and only when the following conditions occur, the corresponding search and rescue boat proposes a communication application: (1) detecting a new task; (2) detecting other member boats; (3) a task is successfully allocated.
The parameters of this embodiment are set as follows: the area of the operation area is 100m multiplied by 100 m; all the search and rescue boats run at a constant speed of unit speed (1 m/s); sampling time delta T is 10 s; constant parameter B is 60, 0.0001; operational capability of each boat
Figure BDA00020895788100000612
Take value in the value range [0,1]And randomly generating.
Fig. 2 and fig. 3 correspond to allocation schemes of the algorithm in isomorphic (all the search and rescue boats in the cluster have the same operation capability, and the limitation of communication and perception radius is not considered) and heterogeneous (all the search and rescue boats in the cluster have different operation capabilities, and the limitation of communication and perception radius is not considered) scenes, respectively. 40 tasks to be executed (shown by solid dots in the figure) are randomly distributed in the operation area, and 3 search and rescue boats are arranged to execute the tasks. The hollow circles correspond to the initial positions of the three search and rescue boats in random distribution respectively. The diamond shaped marks in fig. 3 correspond to a special type of task that can only be performed by a designated search and rescue boat. The operation target is that three search and rescue boats execute all tasks, and returns to the corresponding initial positions after the operation is finished. The method and the device fully consider the matching degree between the search and rescue boat and the task to be executed, and are suitable for isomorphic and heterogeneous cluster systems. And each search and rescue boat executes the tasks distributed in the area nearby as much as possible, thereby avoiding excessive resource consumption and collision among the boats. The provided distribution scheme is reasonable and efficient.
FIG. 4 is a schematic diagram of a task allocation scheme under the condition that the communication radius and the perception radius of the search and rescue boat are limited according to the invention; the 50 tasks (shown by solid dots) are randomly distributed in the operation water area, and two boats are adopted to execute the search and rescue tasks. It can be seen that, in the case that the operating environment information is only partially known, the present invention can still obtain a more satisfactory allocation scheme through online real-time task reallocation. Thereby verifying the utility of the present invention.
The invention has the following beneficial technical effects:
most of traditional auction algorithms are only suitable for the task allocation problem of homogeneous cluster systems, and heterogeneous systems with incomplete prior knowledge, unmanned system communication and limited detection radius are often low in operation efficiency and unreasonable in allocation scheme. The method improves the bidding strategy and the specific auction mode of each round of auction based on the traditional sequential auction algorithm. Meanwhile, the difference of respective capabilities of the heterogeneous unmanned ship clusters is considered. By repeatedly allocating tasks on line in real time, the efficiency and the rationality of task allocation are improved.

Claims (3)

1. A task allocation method for a multi-water-surface unmanned search and rescue boat with known environmental information part is characterized by comprising the following steps:
step 1: initializing, loading known environment information of the operation sea area part, including
Step 1-1: number of tasks to be executed: set to m, marked as set T ═ T1,t2,...,tm}；
Step 1-2: type of task to be performed: by using variable ectjCharacterizing executionJth task tjThe variable is only related to the type of the executed task and is not related to the performance of the search and rescue boat executing the task;
step 1-3: the number of the search and rescue boats is as follows: set to n, and mark as set U ═ U1,u2,...,un}；
Step 1-4: the operation capability of the search and rescue boat is as follows: search and rescue boat u No. iiN implements the jth task t, i 1,2jThe capability of j 1, 2.. multidot.m is set as
Figure FDA0002089578800000011
And order
Figure FDA0002089578800000012
When in use
Figure FDA0002089578800000013
Time, show search and rescue boat uiNot having to perform task tjThe ability of (c);
setting the initial task sub-sequence of each search and rescue boat as an empty set;
step 2: setting a sampling time interval delta T, collecting the information of the carried sensors by each search and rescue boat every delta T, and inserting the detected task to be executed into the task subsequence of the corresponding boat;
and step 3: executing internal auction, marking all tasks to be auctioned in the current round as t'1,t′2...t′pP is less than or equal to m and aims at each task to be photographed t'k1, 2.. p performs the following steps:
step 3-1: will be with task t'kAnd marking all candidate search and rescue boat sets corresponding to p as 1,2
Figure FDA0002089578800000014
I.e. the principal wheel has qkTask t 'is detected by boat'kAnd the boats can communicate with each other, and the complementary set is defined as
Figure FDA0002089578800000015
Corresponding to no task t 'detected in the current wheel'kThe search and rescue boat set;
step 3-2: for arbitrary uks∈Uk,s＝1,2,...,qkCalculating all search and rescue boats uiE.u, i 1,2, n is relative to UksInternal auction bid value of
Figure FDA0002089578800000016
Wherein the content of the first and second substances,
Figure FDA0002089578800000017
corresponding to completion of task t'kThe price to be paid for is that,
Figure FDA0002089578800000018
corresponding boat uksImplementation of task t'kThe capability of (a) to (b),
Figure FDA0002089578800000019
b is a user-defined constant parameter for representing the current unknown search and rescue boat uiAnd task t'kThe geometric distance between the two parts of the frame,
Figure FDA00020895788000000110
corresponding search and rescue boat uksAnd to-be-executed task t'kThe geometric distance between the two is larger than 0, and the user-defined smaller constant value parameter is obtained;
step 3-3: if it is
Figure FDA0002089578800000021
Then u is reservedksIn candidate search and rescue boat set UkThe medium is unchanged; otherwise, if C is equal to C, updating the candidate search and rescue boat set UkBy uiInstead of uks；
And 4, step 4: candidate search and rescue boat set U corresponding to p tasks to be executed in current round after updating1,U2,...,UpAnd taking the task corresponding to the set with the most contained elements as the auction task of the current round and marking the task as the auction task of the current round
Figure FDA0002089578800000022
(subscript f corresponds to the current auction as round f);
and 5: performing external auction and the task of waiting for auction in the current round
Figure FDA0002089578800000023
Marking the corresponding candidate search and rescue boat set as
Figure FDA0002089578800000024
Step 6: and (3) updating the task sequence to be executed, eliminating the completed tasks from the task sequence, if the task sequence to be executed is not empty, returning to the step (2), otherwise, ending the algorithm flow.
2. The method as claimed in claim 1, wherein the step 5 comprises the following steps:
step 5-1: for all search and rescue boats
Figure FDA0002089578800000025
Calculate it and task
Figure FDA0002089578800000026
Geometric distance between them
Figure FDA0002089578800000027
Then determining the search and rescue boat for the current round of host auction
Figure FDA0002089578800000028
Figure FDA0002089578800000029
And at the same time, as the communication node of the auction round,
Figure FDA00020895788000000210
responsible statistics aggregation
Figure FDA00020895788000000211
The bidding conditions of all the search and rescue boats are shown in
Figure FDA00020895788000000212
Broadcasting the final auction result within the range;
step 5-2: the performance index used in the auction process is
Figure FDA00020895788000000213
Figure FDA00020895788000000214
Auction task allocation to search and rescue boat for representing h round
Figure FDA00020895788000000215
Otherwise
Figure FDA00020895788000000216
Variables of
Figure FDA00020895788000000217
Wherein, variable
Figure FDA00020895788000000218
Corresponding search and rescue boat
Figure FDA00020895788000000219
Run to task on h-th round
Figure FDA00020895788000000220
Energy consumed at the locationAn amount;
step 5-3:
Figure FDA00020895788000000221
the corresponding search and rescue boat is the final winner of the auction of the current round.
3. The method for assigning tasks to unmanned multi-surface search and rescue craft according to claim 1, wherein n is set to be equal to or less than m in step 1, and for facilitating subsequent application, the operational capability of the search and rescue craft in steps 1-4 is set to be equal to or less than m
Figure FDA00020895788000000222
Normalization processing is performed.
CN201910499052.0A 2019-06-11 2019-06-11 Multi-water surface unmanned search and rescue boat task allocation method with partially known environmental information Active CN112070328B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910499052.0A CN112070328B (en) 2019-06-11 2019-06-11 Multi-water surface unmanned search and rescue boat task allocation method with partially known environmental information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910499052.0A CN112070328B (en) 2019-06-11 2019-06-11 Multi-water surface unmanned search and rescue boat task allocation method with partially known environmental information

Publications (2)

Publication Number Publication Date
CN112070328A true CN112070328A (en) 2020-12-11
CN112070328B CN112070328B (en) 2023-06-27

Family

ID=73658272

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910499052.0A Active CN112070328B (en) 2019-06-11 2019-06-11 Multi-water surface unmanned search and rescue boat task allocation method with partially known environmental information

Country Status (1)

Country Link
CN (1) CN112070328B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113190041A (en) * 2021-04-30 2021-07-30 哈尔滨工业大学 Unmanned aerial vehicle cluster online target distribution method based on constraint relaxation technology
CN113723805A (en) * 2021-08-30 2021-11-30 上海大学 Unmanned ship composite task allocation method and system
CN114004035A (en) * 2021-12-13 2022-02-01 哈尔滨工业大学（威海） Target tracking control method for unmanned surface vehicle

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108873894A (en) * 2018-06-11 2018-11-23 上海大学 A kind of target following cooperative control system and method based on more unmanned boats
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

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108873894A (en) * 2018-06-11 2018-11-23 上海大学 A kind of target following cooperative control system and method based on more unmanned boats
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

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
吴歇尔: "面向多无人机的协同任务预分配及重分配研究", 南昌航空大学, vol. 2018, no. 11 *
李杰;孙尧;: "基于DAMPSO算法的USVs集群攻击任务规划研究", 计算机工程与应用, vol. 49, no. 20, pages 1 - 4 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113190041A (en) * 2021-04-30 2021-07-30 哈尔滨工业大学 Unmanned aerial vehicle cluster online target distribution method based on constraint relaxation technology
CN113190041B (en) * 2021-04-30 2022-05-10 哈尔滨工业大学 Unmanned aerial vehicle cluster online target distribution method based on constraint relaxation technology
CN113723805A (en) * 2021-08-30 2021-11-30 上海大学 Unmanned ship composite task allocation method and system
CN113723805B (en) * 2021-08-30 2023-08-04 上海大学 Unmanned ship compound task allocation method and system
CN114004035A (en) * 2021-12-13 2022-02-01 哈尔滨工业大学（威海） Target tracking control method for unmanned surface vehicle

Also Published As

Publication number Publication date
CN112070328B (en) 2023-06-27

Similar Documents

Publication Publication Date Title
CN112070328A (en) Multi-water-surface unmanned search and rescue boat task allocation method with known environmental information part
CN108229647A (en) The generation method and device of neural network structure, electronic equipment, storage medium
CN109159113B (en) Robot operation method based on visual reasoning
Bateni et al. Predjoule: A timing-predictable energy optimization framework for deep neural networks
JPWO2020033767A5 (en)
CN110501169A (en) Diagnostic method, device and the electronic equipment of vehicle trouble
CN111064633A (en) Cloud-edge cooperative power information communication equipment automated testing resource allocation method
CN110181516A (en) A kind of paths planning method of spray robot, device, system and storage medium
CN115358831A (en) User bidding method and device based on multi-agent reinforcement learning algorithm under federal learning
CN115049821A (en) Three-dimensional environment target detection method based on multi-sensor fusion
CN109712160A (en) Improved lion group algorithm is combined to realize carrying out image threshold segmentation method based on generalized entropy
CN114520743A (en) Method and system for detecting network abnormal flow and storable medium
CN104077561B (en) Fingerprint automatic comparison method
CN112819152B (en) Neural network training method and device
CN117372877A (en) Star map identification method and device based on neural network and related medium
CN115334165B (en) Underwater multi-unmanned platform scheduling method and system based on deep reinforcement learning
Bode et al. Bounded: Neural boundary and edge detection in 3d point clouds via local neighborhood statistics
CN111860818B (en) SOM neural network algorithm processing method based on intelligent chip
CN111815658B (en) Image recognition method and device
Zhao et al. End-to-end autonomous driving based on the convolution neural network model
CN112150546A (en) Monocular vision pose estimation method based on auxiliary point geometric constraint
US20230186143A1 (en) Selecting training nodes for training machine-learning models in a distributed computing environment
JP2022509564A (en) Neural network active training system and image processing system
CN116935233B (en) Target detection method and device for laser radar
CN111079288B (en) Method and device for determining mathematical model of environmental information

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