CN112488506A - Extensible distributed architecture and self-organizing method of intelligent unmanned system cluster - Google Patents
Extensible distributed architecture and self-organizing method of intelligent unmanned system cluster Download PDFInfo
- Publication number
- CN112488506A CN112488506A CN202011374595.9A CN202011374595A CN112488506A CN 112488506 A CN112488506 A CN 112488506A CN 202011374595 A CN202011374595 A CN 202011374595A CN 112488506 A CN112488506 A CN 112488506A
- Authority
- CN
- China
- Prior art keywords
- leader
- global
- cluster
- unmanned system
- unmanned
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000007246 mechanism Effects 0.000 claims abstract description 64
- 230000008520 organization Effects 0.000 claims abstract description 7
- 230000000977 initiatory effect Effects 0.000 claims description 26
- 238000004891 communication Methods 0.000 claims description 19
- 239000003795 chemical substances by application Substances 0.000 claims description 18
- 230000004044 response Effects 0.000 claims description 18
- 238000013461 design Methods 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 6
- 238000009826 distribution Methods 0.000 claims description 5
- 239000000203 mixture Substances 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000008521 reorganization Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000013500 data storage Methods 0.000 claims description 2
- 239000003999 initiator Substances 0.000 claims description 2
- 230000008569 process Effects 0.000 claims description 2
- 230000003068 static effect Effects 0.000 abstract description 2
- 239000010410 layer Substances 0.000 description 39
- 230000015572 biosynthetic process Effects 0.000 description 12
- 238000005755 formation reaction Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 239000002131 composite material Substances 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000004927 fusion Effects 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 230000000717 retained effect Effects 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 206010033799 Paralysis Diseases 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000002346 layers by function Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
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
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an extensible distributed architecture and a self-organizing method of an intelligent unmanned system cluster, which mainly surround the realization of two characteristics of 'distributed' and 'extensible', are designed from two layers of a static organization architecture and a dynamic operation mechanism of the cluster system, and particularly comprise a flexible three-layer extensible distributed architecture of the unmanned system cluster and a dynamic self-organizing method of the unmanned system cluster. The invention establishes 13 operation mechanisms of 3 categories related to attendance of an unmanned system, role exemption of a cluster system and dynamic self-organization operation guarantee, provides comprehensive support for dynamic loss and increase of cluster members, realizes that the system members can enter and exit, dynamically updates global data and automatically reconstructs organization relationship, and the cluster system can still stably operate under the condition of member loss or increase.
Description
Technical Field
The invention relates to the technical field of intelligent unmanned systems, in particular to an extensible distributed architecture and a self-organizing method of an intelligent unmanned system cluster.
Background
With the increasingly complex scenes and tasks of unmanned systems, a single unmanned system cannot meet the requirements, and the complex tasks require that a plurality of unmanned systems are mutually matched to form a cluster to jointly complete the tasks. The key problem to be solved when constructing a cluster system is to organize elements such as an unmanned system cluster platform, functional modules, data and the like so that a cluster can exert greater efficiency, but the research on the problem is not sufficient.
The unmanned system cluster in the existing research usually adopts two architectures, namely a centralized architecture and a distributed architecture.
The centralized architecture means that a central node is formed by one or more host computers, data is stored in the central node in a centralized manner, all service units of the whole system are deployed on the central node in a centralized manner, and all functions of the system are processed in a centralized manner. Scholars such as Hujian chapters and the like adopt a centralized control architecture in research of water surface unmanned ship clusters and control methods thereof, a shore station is used as a central node, each unmanned ship can only receive instructions of the shore station, no information interaction exists among the ships, and the unmanned ships need to sail according to a pre-planned path. However, when the communication between the shore station and each boat is delayed or even disconnected, the unmanned boat cluster cannot normally execute tasks, and the method has no expandability of the number of cluster members, so that the loss or increase of unmanned boats occurs in the actual navigation task execution, the cluster cannot perform self-organization adjustment, the stability of the cluster cannot be ensured, and even the cluster is paralyzed.
The distributed architecture is decentralized, no node in the dominant position exists, each node has the capability of receiving a request from the outside and carrying out corresponding processing, and capacity expansion is not required to be carried out on a single center. For example, students who are smart and the like propose a three-layer layered structure of an intelligent unmanned cluster aiming at the intelligent unmanned cluster, the three-layer layered structure comprises a supporting layer, a functional layer and an application layer, the three-layer layered structure can adapt to various scenes, tasks are completed, and cluster members can communicate with each other and exchange information. However, the three-layer hierarchical structure does not support the completion of large-scale group multi-subject and multi-level task planning, and the influence of the increase and decrease of cluster members on the integrity of the cluster function is not mentioned. The university scholars such as the royal universe adopt a distributed architecture in the unmanned system cluster maritime combat application research, have the characteristics of no center and autonomous cooperation, the loss of members in the combat cannot damage the functional integrity of the cluster, and the cluster can continue to execute tasks. But does not support multitasking and presents difficulties with formation self-organization, formation splitting/reorganization, cluster collaborative navigation, etc.
It can be seen that the architecture design of the current cluster system faces the following problems:
(1) the task is single. The cluster system has single execution task, has overlarge difference with the actual task, and lacks of an unmanned system cluster architecture supporting multiple tasks.
(2) The cluster member size lacks scalability. For a cluster architecture design method, most of the existing disclosed architectures adopt a centralized architecture, expansion flexibility is lacked, and the stability of an unmanned system cluster cannot be guaranteed in the case of loss or increase of the unmanned system in the cluster.
(3) Clustering global optimization is difficult. By adopting a distributed architecture, the communication traffic of a cluster system is large, local conflict is easy to generate, most of the cluster systems can only adopt a given path, the flexibility is lacked, the combination with the actual situation is difficult, and the application in practice is very difficult.
Disclosure of Invention
The invention aims to solve the problems in the prior art, realize the task requirements of an unmanned system cluster, and provide an extensible distributed architecture and a self-organizing method of an intelligent unmanned system cluster, which have the advantages of good expandability in scale, multi-task support, high system operation efficiency and good global optimization.
The purpose of the invention is realized as follows:
an extensible distributed architecture and a self-organizing method of an intelligent unmanned system cluster comprise a flexible three-layer extensible distributed architecture of the unmanned system cluster and a dynamic self-organizing method of the unmanned system cluster.
The flexible three-layer extensible distributed architecture of the unmanned system cluster specifically comprises:
(1) basic structure and composition of clusters
A flexible three-layer extensible distributed architecture adapting to a 1+ C + X mode, wherein a global leader node comprising 1 cluster, C is a candidate leader node combination, X is a general unmanned system individual member set, and flexibility comprises two meanings: firstly, the member role is not invariable and can be dynamically adjusted according to the requirement; secondly, the three-layer structure is variable, the emergency can be dispatched beyond the layer, and the automatic reconstruction of three structures and control modes of three-layer structure two-stage control, two-layer/three-layer mixed structure and direct control/two-stage mixed control and two-layer structure direct control can be realized according to task needs and observation conditions;
(2) method for calculating number of various members in cluster
The number of C and X can be determined as follows: if the cluster scale of the unmanned system is N and C is determined by the reliability index of the system, if the system architecture supports that the cluster system is stable under the condition that the cluster members lose or increase x%, the number | C | of C is determined according to
Is determined in which
The system is designed in such a way that the global information of the cluster can still be kept when system members (including a leader and a candidate leader) are damaged, and the cluster can be reorganized to continue to execute the cooperative task through a dynamic reorganization and backup mechanism;
(3) unmanned system roles and responsibilities in a cluster
The design responsibility of the global leader is:
firstly, mastering global resources and taking charge of maintaining a global resource database (including unmanned system data volume, grouping, roles and the like);
secondly, the system is responsible for global task planning, distribution and coordination;
thirdly, considering factors such as observation range limitation, the leader mainly keeps a communication relationship with a candidate leader, but keeps a communication channel with a common unmanned system, and emergency can be directly communicated;
the candidate leader designs 3 roles, and besides the common operation function of the common unmanned system, the candidate leader also additionally takes 3 different types of responsibilities:
backup of a global leader can synchronously update a global resource database in real time;
a local leader is responsible for group task allocation and local conflict coordination;
the common unmanned system has an operation execution function;
(4) global database
The global database is mainly used for managing global parameters related to system composition and configuration, including unmanned system data volume, grouping, roles and capabilities;
the description and storage structure of the single unmanned system capability adopts the idea of data storage by using an entity-attribute value structure in a knowledge map collection, is designed into an unmanned system ID-capability attribute mode, and has very good expansibility;
the storage of other data can be realized by adopting a two-dimensional table structure of a conventional relational database;
the database is updated and maintained by the leader, and after updating, the database is sent to a candidate leader for backup in real time after one database transaction is completed to ensure data integrity.
The dynamic self-organization method of the unmanned system cluster specifically comprises the following steps:
by establishing 13 operation mechanisms of 3 categories related to attendance of the unmanned system, role exemption of the cluster system and dynamic self-organization operation guarantee, comprehensive support is provided for dynamic loss and increase of cluster members:
(1) a registration mechanism: the triggering condition is that a new unmanned system joins the cluster; the initiating agent is a newly joined unmanned system; the response main body is a global leader or a local leader; the operation content is that when the newly added unmanned system is registered to the global leader and cannot directly communicate with the global leader temporarily, the newly added unmanned system can be registered to the local leader, the local leader collects registration information to the global leader, and the registration information comprises identification information and capability description of the newly added unmanned system;
(2) the sign-in mechanism comprises: the triggering condition is the set check-in time; the initiating subject is an unmanned system which does not ask for leave; the response main body is a global leader or a local leader; the operation content is that a local leader signs in to a global leader within a set period, and the operation content is divided into two conditions, wherein the local leader signs in to a common unmanned system item under the condition of grouping, the local leader sends the sign-in information of other systems of the group when signing in to the global leader, and the local leader signs in to the global leader directly under the condition of no grouping;
(3) the roll call mechanism comprises: the triggering condition is overtime non-check-in or emergency; the initiating subject is a global leader or a local leader; the responder is an unmanned system that has not checked in; the global leader or the local leader roll the name of the unmanned system under the emergency condition (such as emergency task allocation and no sign-in within a specified time);
(4) please cancel the false mechanism: the triggering condition is leave asking or leave; the initiating agent is an unmanned system asking for leave; the response main body is a global leader or a local leader; the operation content is that the unmanned system is temporarily separated from the cluster system (such as off-line maintenance, task execution outside a communication range and the like) and asks for leave to a global leader or a local leader, and the leave is eliminated after the queue is returned;
(5) a name removal mechanism: the triggering condition is that the absence exceeds a limited number of times; the initiating subject is a global leader or a local leader; the response body is a global leader or other local leaders; the operation content is that the unmanned system with no name is called and overdue is not sold in the specified time, the name is removed, and the system is not considered in the subsequent task distribution;
(6) a leader generation mechanism: the triggering condition is cluster initialization or global leadership incapability; the initiating subject is a candidate leader; the response main body is other candidate leaders; when the operation content is the initialization of a cluster system, a global leader is manually specified, after the global leader loses the capability in the process of executing a task, a new global leader is generated from candidate leaders, and a generation mechanism comprehensively considers random generation, generation according to the calculation capability, generation according to the size of a group under jurisdiction, generation according to the geographic position (close to a cluster center), generation according to the safety of the executed task and the like;
(7) a leader handover mechanism: the triggering condition is according to task needs; the initiating agent is a global leader; the response main body is a candidate leader; the operation content is that the global leader can be dynamically changed according to task requirements (communication reliability, scheduling security and the like), and the global leader authority can be handed over according to a leader generation mechanism;
(8) a candidate leader generation mechanism: the triggering condition is that a new unmanned system is added into the cluster; the initiating agent is a global leader; the response main body is a candidate leader; the operation content is generated by two mechanisms, namely, grouping exists, a local leader automatically becomes a candidate leader, a large-scale group has priority, and no grouping exists, and the global leader randomly appoints the candidate leader;
(9) a candidate leader revocation mechanism: the triggering condition is that the name of the unmanned system is removed; the initiating agent is a global leader; the response main body is a candidate leader; the operation content is a revocation candidate leader, wherein the role of the operation content is changed in the global database, and the backup of the global database is stopped;
(10) an initial team organization mechanism: the triggering condition is cluster construction; the initiating agent is a global leader; the responding main body is a whole unmanned system; the operation content is that system initialization starts from a global leader, after the global leader is generated, other unmanned systems register to the global leader, the leader generates alternate leaders according to stability indexes, but does not group, and an initial structure is direct administration control of a two-layer structure;
(11) and (3) dynamic reconfiguration mechanism: the triggering condition is that cluster members change or task needs occur; the initiating agent is a global leader; the responding main body is a whole unmanned system; the operation content is that the cluster system can be grouped, can not be grouped, can also be partially grouped, can be partially directly administered, and the grouping condition can be dynamically adjusted according to the task requirement;
(12) a database backup mechanism: the triggering condition is that the cluster global resource database is updated; the initiating agent is a global leader; the response main body is a candidate leader or a related unmanned system; after the operation content is updated in a database, the global leader makes incremental backup for all candidate leaders in real time to ensure that the candidate leaders can take over the work of the global leader at any time, the newly generated candidate leaders directly copy the global database, and after the local leader is changed, relevant data is updated to an unmanned system governed by the local leader;
(13) the security authentication mechanism comprises: the triggering condition is communication; the initiating agent is a communication initiator; the answering body is the communication receiver; the operation content provides identity authentication information for mechanisms such as task allocation, role exemption, database backup, sign-in, roll call and the like, and encryption processing is carried out on communication information.
Compared with the prior art, the invention has the beneficial effects that:
the invention is designed into a flexible three-layer extensible distributed architecture suitable for a 1+ C + X mode, the member roles can be dynamically adjusted according to multi-task requirements, the system structure can be self-organized and adjusted, the emergency situation can be dispatched beyond the layer, and the automatic reconfiguration of three structures and control modes of three-layer structure two-stage control, two-layer/three-layer mixed structure and direct control/two-stage mixed control and two-layer structure direct control can be realized according to task requirements and observation conditions. Due to the limitation of the visual field and the observation capability, the cluster or subgroup often cannot master the global information, so that the decision is influenced, the step-by-step convergence and fusion of perception information can be supported through the 1+ C + X layered architecture mode, the three-layer structure can improve the working efficiency of the system, the two-layer structure design can deal with the emergency situation, meanwhile, the mixed structure can support the sharing and scheduling control of the information beyond the layers, and the influence of the limitation of the visual field and the observation capability on the behavior decision is reduced to the maximum extent.
The flexible three-tier scalable distributed architecture described above provides structural support for scalability (membership increase and decrease in x% amplitude) from four aspects:
(1) the number of unmanned systems can be dynamically expanded, C and X are not fixed and can be automatically calculated according to the technical index requirement of system stability;
(2) the unmanned system capability can be dynamically expanded, and flexible support is provided for the description of the capability in the aspect of database design;
(3) the roles of the unmanned systems can be dynamically and autonomously adjusted, and the roles of the single unmanned systems in the cluster are not invariable and are adjustable according to tasks;
(4) the hierarchical relation of the cluster system can be dynamically and autonomously adjusted, and the designed cluster structure is a flexible control structure which takes three layers as main and can be dispatched in an emergency situation by crossing layers.
And 3 major types of 13 operation mechanisms related to attendance of the unmanned system, role exemption of the cluster system and dynamic self-organization operation guarantee are established, so that comprehensive support is provided for dynamic loss and increase of cluster members, the system members can enter and exit, global data dynamic update and automatic reconstruction of organization relations are realized, and the cluster system can still stably operate under the condition of member loss or member increase.
Drawings
Fig. 1 is a flexible three-layer expandable distributed architecture diagram of an unmanned system cluster, wherein "1" is a global leader, C is a set of candidate leaders (also local leaders), and X is a set of other unmanned system members. The connecting arrows represent control relationships;
FIG. 2 is a dynamic self-organizing map of the unmanned system cluster of the present invention;
FIG. 3 is a main flow chart of the dynamic self-organizing operation of the unmanned system cluster of the present invention;
FIG. 4 is a diagram of initial formation member data information according to an embodiment of the present invention;
FIG. 5 is a diagram of formation member data information after 10% of members are lost in accordance with an embodiment of the present invention;
FIG. 6 is a diagram of formation member data information after 10% members are added in the embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
The invention aims to solve the problems in the prior art, realize the task requirements of an unmanned system cluster, and provide a design method of an expandable distributed architecture, which has good expandability in scale, supports multiple tasks, and has high system operation efficiency and good overall optimization. The method mainly comprises the implementation of two characteristics of distributed and expandable, and the design is carried out from two levels of a static organization architecture and a dynamic operation mechanism of a cluster system, and specifically comprises a flexible three-layer expandable distributed architecture of an unmanned system cluster and a dynamic self-organization method of the unmanned system cluster. The flexible three-layer extensible distributed architecture design of the unmanned system cluster mainly solves the problems of basic composition of the cluster system, role definition (responsibility distribution) of the unmanned system in the cluster system, distributed structure relation (including topology and control structure) and the like from a logic level, and mainly solves the problems of construction, storage and maintenance of a global resource database, structural support for cluster extensibility and the like from a physical level, wherein the flexibility means that the three-layer structure is variable and can be dispatched beyond the layer in emergency, the design of the three-layer structure mainly considers the system operation efficiency, the design of the two-layer structure mainly considers the emergency, and the design of a mixed structure is considered when the two conditions coexist. Due to the limitation of the visual field and the observation capability, the cluster or subgroup may not be able to master the global information, thereby influencing the decision, the flexible three-layer architecture mode can support the gradual aggregation and fusion of the perception information, and the hybrid structure can support the sharing and scheduling control of the cross-layer information, thereby reducing the influence of the observation capability on the behavior decision to the maximum extent. The dynamic self-organization method of the unmanned system cluster mainly ensures that system members can enter and exit, global data dynamic update and organization relation automatic reconstruction through a set of self-organization mechanism, and the cluster system can still stably operate under the condition of member supplement or reduction.
Example (b):
by adopting the distributed extensible architecture designed by the invention, when the unmanned system executes tasks, if cluster members change, the global information of the cluster system can be effectively reserved, the stability of the architecture of the system can be ensured through a cluster dynamic self-organizing mechanism, and the continuation of the tasks can be ensured.
In this embodiment, the following two member variation events are simulated in combination with the actual task scenario:
1. cluster systems lose 10% of their members;
2. the cluster system increases the membership by 10%.
In this embodiment, the initial cluster system consists of 20 unmanned systems, including 1 lead unmanned system, 2 candidate lead unmanned systems, and 17 normal unmanned systems. Which will be separately described below.
1. Cluster system loss of 10% membership
When 10% of members of the cluster system are lost, namely 2 unmanned systems are lost, if the 2 lost unmanned systems are 1 leading unmanned system and 1 candidate leading unmanned system, the roll call mechanism and the name removal mechanism are triggered to remove the name of the lost unmanned system. And then, executing a leader generation mechanism and a candidate leader generation mechanism, and selecting a new leader and a candidate leader. The new leaders and the candidate leaders are unmanned systems with high comprehensive evaluation values in the cluster system, and the number of the candidate leaders is 10% of the total number of cluster members.
2. Cluster system increased member by 10%
When 10% of members are added to the cluster system on the basis of 10% of lost members, namely 2 unmanned systems are added to the 18 unmanned systems, and the cluster system has 20 unmanned systems in total, a registration mechanism, a check-in mechanism and a point name mechanism are triggered to ensure that the newly added unmanned systems are effective. And then, executing a candidate lead generation mechanism to ensure that the candidate lead is an unmanned system with high comprehensive evaluation value in the cluster system, and the number of the candidate lead unmanned systems is 10% of the total number of the cluster members.
Example results and analysis.
The initial cluster system had a total of 20 unmanned systems, forming 3 groups of formations. The roles (global leader, candidate leader, and normal unmanned system, respectively, indicated by role attribute values 1,2, and 3), the state (unmanned system loss and unmanned system normal operation, respectively, indicated by state attribute values 0 and 1), the queue number (formation queue number to which the queue number belongs, indicated by queue number attribute value), the composite rating (composite rating value which the unmanned system has, indicated by composite rating attribute value), the check-in (unmanned system check-in failure and unmanned system check-in success, respectively, indicated by check-in attribute values 0 and 1), the IP address (IP address of the unmanned system, indicated by IP address attribute value), and the port number (communication port number between the unmanned systems, indicated by port number attribute value) shown by the global database information in fig. 4
1. Cluster system loss of 10% membership
When a cluster system loses 10% of members, namely 2 unmanned systems, in order to show a better cluster dynamic self-organization mechanism effect with fewer scenes, the 2 lost unmanned systems are the No. 1 leader and the No. 9 candidate leader. At this time, the global database information of the unmanned system in the cluster system changes as shown in fig. 5.
The lost unmanned system number is determined by the roll call mechanism. And then the lost unmanned system is named after the name is removed through a name removing mechanism. At which point the named unmanned system will no longer be considered in subsequent task assignments. As shown in fig. 5, in 20 unmanned systems, the state values of No. 1 and No. 9 are 0, that is, 2 unmanned systems that are lost. The angle value shows that the No. 1 unmanned system is a global leader, and the No. 9 unmanned system is a candidate leader. Then, executing a leader generation mechanism and a candidate leader generation mechanism, selecting a No. 4 unmanned system with the highest comprehensive evaluation value as a new global leader, and changing the role value into a leader; and selecting No. 2 and No. 10 with high comprehensive evaluation values as new candidate leaders, and changing the role values into the candidate leaders.
The results of the above embodiments show that the cluster system guarantees the number of leader unmanned systems and candidate leader unmanned systems of the cluster system through a cluster dynamic self-organizing mechanism under the condition of 10% unmanned system loss, and guarantees the stability of a flexible three-layer extensible distributed architecture system in a "1 + C + X" mode, so that global information is retained, and formation can continue to execute tasks stably.
2. Cluster system increased member by 10%
When 10% of members of the cluster system are added on the basis of 10% of members loss, that is, 2 unmanned systems are newly added, the global database information of the unmanned systems in the cluster system changes as shown in fig. 6.
And 2 unmanned systems are added to trigger a registration mechanism, a sign-in mechanism, a roll call mechanism and a candidate leader generation mechanism, as shown in fig. 6, the state values and the sign-in values of the 2 newly added unmanned systems are all 1, the newly added unmanned systems are successfully added to the formation to execute tasks, and the 21 unmanned system with the angle value and the queue length number value as a new candidate leader is used as a formation queue length.
The results of the above embodiments show that the cluster system, with 10% of unmanned systems added, ensures that the number of candidate leader unmanned systems accounts for 10% of the total number of cluster members through the cluster dynamic self-organization mechanism, and ensures that the flexible three-layer extensible distributed architecture system in the "1 + C + X" mode is stable, so that global information is retained, and the formation can continue to execute tasks stably.
The two embodiments are integrated to show that no matter 10% of unmanned systems are lost or 10% of unmanned systems are added in the cluster system, the stability of a flexible three-layer extensible distributed architecture system in a 1+ C + X mode can be ensured through a cluster dynamic self-organization mechanism; the global information is kept, and the formation can continue to stably execute the task.
Claims (3)
1. An extensible distributed architecture and a self-organizing method of an intelligent unmanned system cluster are characterized by comprising a flexible three-layer extensible distributed architecture of the unmanned system cluster and a dynamic self-organizing method of the unmanned system cluster.
2. The scalable distributed architecture and self-organizing method of an intelligent unmanned system cluster according to claim 1, wherein the flexible three-layer scalable distributed architecture of the unmanned system cluster specifically comprises:
(1) basic structure and composition of clusters
A flexible three-layer extensible distributed architecture adapting to a 1+ C + X mode, wherein a global leader node comprising 1 cluster, C is a candidate leader node combination, X is a general unmanned system individual member set, and flexibility comprises two meanings: firstly, the member role is not invariable and can be dynamically adjusted according to the requirement; secondly, the three-layer structure is variable, the emergency can be dispatched beyond the layer, and the automatic reconstruction of three structures and control modes of three-layer structure two-stage control, two-layer/three-layer mixed structure and direct control/two-stage mixed control and two-layer structure direct control can be realized according to task needs and observation conditions;
(2) method for calculating number of various members in cluster
The number of C and X can be determined as follows: the cluster size of the unmanned system is set as N, C is determined by the reliability index of the system, such as' the system architecture supports the cluster member in damageIf x% is lost or increased, the cluster system remains stable ", and the number | C | of C is determined according to
Is determined in which
The system is designed in such a way that the global information of the cluster can still be kept when system members (including a leader and a candidate leader) are damaged, and the cluster can be reorganized to continue to execute the cooperative task through a dynamic reorganization and backup mechanism;
(3) unmanned system roles and responsibilities in a cluster
The design responsibility of the global leader is:
firstly, mastering global resources and taking charge of maintaining a global resource database (including unmanned system data volume, grouping, roles and the like);
secondly, the system is responsible for global task planning, distribution and coordination;
thirdly, considering factors such as observation range limitation, the leader mainly keeps a communication relationship with a candidate leader, but keeps a communication channel with a common unmanned system, and emergency can be directly communicated;
the candidate leader designs 3 roles, and besides the common operation function of the common unmanned system, the candidate leader also additionally takes 3 different types of responsibilities:
backup of a global leader can synchronously update a global resource database in real time;
a local leader is responsible for group task allocation and local conflict coordination;
the common unmanned system has an operation execution function;
(4) global database
The global database is mainly used for managing global parameters related to system composition and configuration, including unmanned system data volume, grouping, roles and capabilities;
the description and storage structure of the single unmanned system capability adopts the idea of data storage by using an entity-attribute value structure in a knowledge map collection, is designed into an unmanned system ID-capability attribute mode, and has very good expansibility;
the storage of other data can be realized by adopting a two-dimensional table structure of a conventional relational database;
the database is updated and maintained by the leader, and after updating, the database is sent to a candidate leader for backup in real time after one database transaction is completed to ensure data integrity.
3. The scalable distributed architecture and self-organization method of an intelligent unmanned system cluster according to claim 1, wherein the dynamic self-organization method of the unmanned system cluster specifically comprises:
by establishing 13 operation mechanisms of 3 categories related to attendance of the unmanned system, role exemption of the cluster system and dynamic self-organization operation guarantee, comprehensive support is provided for dynamic loss and increase of cluster members:
(1) a registration mechanism: the triggering condition is that a new unmanned system joins the cluster; the initiating agent is a newly joined unmanned system; the response main body is a global leader or a local leader; the operation content is that when the newly added unmanned system is registered to the global leader and cannot directly communicate with the global leader temporarily, the newly added unmanned system can be registered to the local leader, the local leader collects registration information to the global leader, and the registration information comprises identification information and capability description of the newly added unmanned system;
(2) the sign-in mechanism comprises: the triggering condition is the set check-in time; the initiating subject is an unmanned system which does not ask for leave; the response main body is a global leader or a local leader; the operation content is that a local leader signs in to a global leader within a set period, and the operation content is divided into two conditions, wherein the local leader signs in to a common unmanned system item under the condition of grouping, the local leader sends the sign-in information of other systems of the group when signing in to the global leader, and the local leader signs in to the global leader directly under the condition of no grouping;
(3) the roll call mechanism comprises: the triggering condition is overtime non-check-in or emergency; the initiating subject is a global leader or a local leader; the responder is an unmanned system that has not checked in; the global leader or the local leader roll the name of the unmanned system under the emergency condition (such as emergency task allocation and no sign-in within a specified time);
(4) please cancel the false mechanism: the triggering condition is leave asking or leave; the initiating agent is an unmanned system asking for leave; the response main body is a global leader or a local leader; the operation content is that the unmanned system is temporarily separated from the cluster system (such as off-line maintenance, task execution outside a communication range and the like) and asks for leave to a global leader or a local leader, and the leave is eliminated after the queue is returned;
(5) a name removal mechanism: the triggering condition is that the absence exceeds a limited number of times; the initiating subject is a global leader or a local leader; the response body is a global leader or other local leaders; the operation content is that the unmanned system with no name is called and overdue is not sold in the specified time, the name is removed, and the system is not considered in the subsequent task distribution;
(6) a leader generation mechanism: the triggering condition is cluster initialization or global leadership incapability; the initiating subject is a candidate leader; the response main body is other candidate leaders; when the operation content is the initialization of a cluster system, a global leader is manually specified, after the global leader loses the capability in the process of executing a task, a new global leader is generated from candidate leaders, and a generation mechanism comprehensively considers random generation, generation according to the calculation capability, generation according to the size of a group under jurisdiction, generation according to the geographic position (close to a cluster center), generation according to the safety of the executed task and the like;
(7) a leader handover mechanism: the triggering condition is according to task needs; the initiating agent is a global leader; the response main body is a candidate leader; the operation content is that the global leader can be dynamically changed according to task requirements (communication reliability, scheduling security and the like), and the global leader authority can be handed over according to a leader generation mechanism;
(8) a candidate leader generation mechanism: the triggering condition is that a new unmanned system is added into the cluster; the initiating agent is a global leader; the response main body is a candidate leader; the operation content is generated by two mechanisms, namely, grouping exists, a local leader automatically becomes a candidate leader, a large-scale group has priority, and no grouping exists, and the global leader randomly appoints the candidate leader;
(9) a candidate leader revocation mechanism: the triggering condition is that the name of the unmanned system is removed; the initiating agent is a global leader; the response main body is a candidate leader; the operation content is a revocation candidate leader, wherein the role of the operation content is changed in the global database, and the backup of the global database is stopped;
(10) an initial team organization mechanism: the triggering condition is cluster construction; the initiating agent is a global leader; the responding main body is a whole unmanned system; the operation content is that system initialization starts from a global leader, after the global leader is generated, other unmanned systems register to the global leader, the leader generates alternate leaders according to stability indexes, but does not group, and an initial structure is direct administration control of a two-layer structure;
(11) and (3) dynamic reconfiguration mechanism: the triggering condition is that cluster members change or task needs occur; the initiating agent is a global leader; the responding main body is a whole unmanned system; the operation content is that the cluster system can be grouped, can not be grouped, can also be partially grouped, can be partially directly administered, and the grouping condition can be dynamically adjusted according to the task requirement;
(12) a database backup mechanism: the triggering condition is that the cluster global resource database is updated; the initiating agent is a global leader; the response main body is a candidate leader or a related unmanned system; after the operation content is updated in a database, the global leader makes incremental backup for all candidate leaders in real time to ensure that the candidate leaders can take over the work of the global leader at any time, the newly generated candidate leaders directly copy the global database, and after the local leader is changed, relevant data is updated to an unmanned system governed by the local leader;
(13) the security authentication mechanism comprises: the triggering condition is communication; the initiating agent is a communication initiator; the answering body is the communication receiver; the operation content provides identity authentication information for mechanisms such as task allocation, role exemption, database backup, sign-in, roll call and the like, and encryption processing is carried out on communication information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011374595.9A CN112488506B (en) | 2020-11-30 | 2020-11-30 | Extensible distributed architecture and self-organizing method of intelligent unmanned system cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011374595.9A CN112488506B (en) | 2020-11-30 | 2020-11-30 | Extensible distributed architecture and self-organizing method of intelligent unmanned system cluster |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112488506A true CN112488506A (en) | 2021-03-12 |
| CN112488506B CN112488506B (en) | 2024-07-12 |
Family
ID=74937668
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011374595.9A Active CN112488506B (en) | 2020-11-30 | 2020-11-30 | Extensible distributed architecture and self-organizing method of intelligent unmanned system cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112488506B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115811529A (en) * | 2022-11-18 | 2023-03-17 | 中国电子技术标准化研究院 | Unmanned cluster system architecture method |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030172145A1 (en) * | 2002-03-11 | 2003-09-11 | Nguyen John V. | System and method for designing, developing and implementing internet service provider architectures |
| CN1805379A (en) * | 2006-01-05 | 2006-07-19 | 西安交通大学 | Method of constructing cluster management and monitoring system in resilient system architecture |
| US20090083390A1 (en) * | 2007-09-24 | 2009-03-26 | The Research Foundation Of State University Of New York | Automatic clustering for self-organizing grids |
| US20150067135A1 (en) * | 2013-08-22 | 2015-03-05 | Inspur Electronic Information Industry Co., Ltd | Member-oriented hybrid cloud operating system architecture and communication method thereof |
| CN105045929A (en) * | 2015-08-31 | 2015-11-11 | 国家电网公司 | MPP architecture based distributed relational database |
| CN106027634A (en) * | 2016-05-16 | 2016-10-12 | 白杨 | Baiyang message port switch service |
| CN106953744A (en) * | 2017-02-27 | 2017-07-14 | 浙江工商大学 | A kind of SDN cluster controllers High Availabitity architecture design method |
| CN111093220A (en) * | 2019-11-14 | 2020-05-01 | 中国人民解放军军事科学院国防科技创新研究院 | Autonomous unmanned cluster dynamic management method and management platform |
| CN111830995A (en) * | 2020-08-06 | 2020-10-27 | 中国科学院自动化研究所 | Group intelligent cooperation method and system based on hybrid architecture |
| CN111858465A (en) * | 2020-06-29 | 2020-10-30 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Large-scale matrix QR decomposition parallel computing structure |
-
2020
- 2020-11-30 CN CN202011374595.9A patent/CN112488506B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030172145A1 (en) * | 2002-03-11 | 2003-09-11 | Nguyen John V. | System and method for designing, developing and implementing internet service provider architectures |
| CN1805379A (en) * | 2006-01-05 | 2006-07-19 | 西安交通大学 | Method of constructing cluster management and monitoring system in resilient system architecture |
| US20090083390A1 (en) * | 2007-09-24 | 2009-03-26 | The Research Foundation Of State University Of New York | Automatic clustering for self-organizing grids |
| US20150067135A1 (en) * | 2013-08-22 | 2015-03-05 | Inspur Electronic Information Industry Co., Ltd | Member-oriented hybrid cloud operating system architecture and communication method thereof |
| CN105045929A (en) * | 2015-08-31 | 2015-11-11 | 国家电网公司 | MPP architecture based distributed relational database |
| CN106027634A (en) * | 2016-05-16 | 2016-10-12 | 白杨 | Baiyang message port switch service |
| CN106953744A (en) * | 2017-02-27 | 2017-07-14 | 浙江工商大学 | A kind of SDN cluster controllers High Availabitity architecture design method |
| CN111093220A (en) * | 2019-11-14 | 2020-05-01 | 中国人民解放军军事科学院国防科技创新研究院 | Autonomous unmanned cluster dynamic management method and management platform |
| CN111858465A (en) * | 2020-06-29 | 2020-10-30 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Large-scale matrix QR decomposition parallel computing structure |
| CN111830995A (en) * | 2020-08-06 | 2020-10-27 | 中国科学院自动化研究所 | Group intelligent cooperation method and system based on hybrid architecture |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115811529A (en) * | 2022-11-18 | 2023-03-17 | 中国电子技术标准化研究院 | Unmanned cluster system architecture method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112488506B (en) | 2024-07-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106502266B (en) | Multi-machine collaborative operation command control system | |
| CN109917811A (en) | Unmanned aerial vehicle cluster cooperative obstacle avoidance-reconstruction processing method | |
| CN103365979A (en) | Long-distance double-center online processing method and system based on open database | |
| CN110288165A (en) | The unmanned monitoring and tracking method of multiple target based on quick clustering and optimization distribution | |
| CN113359838B (en) | Multi-UAV cooperative flight control system and method | |
| CN109951568B (en) | Aviation cluster mixed multi-layer alliance building method for improving contract network | |
| CN111209294A (en) | Multi-unmanned aerial vehicle-oriented distributed coverage search method based on area allocation | |
| CN115733754B (en) | Resource management system based on cloud primary center platform technology and elastic construction method thereof | |
| CN112801539A (en) | Flexible network architecture dynamic scheduling model of unmanned aerial vehicle cluster task | |
| Zhao et al. | An approach for weapon system-of-systems scheme generation based on a supernetwork granular analysis | |
| CN112488506A (en) | Extensible distributed architecture and self-organizing method of intelligent unmanned system cluster | |
| Duan et al. | Dynamic tasks scheduling model of UAV cluster based on flexible network architecture | |
| CN114518772B (en) | Unmanned aerial vehicle swarm self-organization method in rejection environment | |
| CN113284370A (en) | Intelligent management system of city sharing unmanned aerial vehicle based on block chain technology | |
| CN114337783A (en) | Space distributed edge computing architecture and service processing method | |
| Fu et al. | Toward energy-efficient UAV-assisted wireless networks using an artificial intelligence approach | |
| CN117726153A (en) | Real-time rescheduling method suitable for unmanned aerial vehicle cluster operation task | |
| CN110662272A (en) | Minimum-number pilot selection method based on swarm unmanned aerial vehicle | |
| CN116915313A (en) | Intelligent load balancing method and system for double-layer giant constellation | |
| CN116307336A (en) | Method and device for planning path of electricity exchanging and computer equipment | |
| CN106936706A (en) | Method for routing foundation and system and subdomain controller, master controller | |
| CN105282230B (en) | A kind of hard real time computer cluster dynamic scheduling system | |
| CN115525061A (en) | Multi-unmanned aerial vehicle cooperative control method based on graph theory | |
| US20210334649A1 (en) | Apparatus and method for distributed reinforcement learning | |
| Sousa et al. | Decentralized distribution of UAV fleets based on fuzzy clustering for demand-driven aerial services |
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 |