CN115150408A - Unmanned cluster distributed situation maintenance method based on information extraction - Google Patents
Unmanned cluster distributed situation maintenance method based on information extraction Download PDFInfo
- Publication number
- CN115150408A CN115150408A CN202210703875.2A CN202210703875A CN115150408A CN 115150408 A CN115150408 A CN 115150408A CN 202210703875 A CN202210703875 A CN 202210703875A CN 115150408 A CN115150408 A CN 115150408A
- Authority
- CN
- China
- Prior art keywords
- information
- local area
- local
- task execution
- global
- 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.)
- Pending
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000012423 maintenance Methods 0.000 title claims abstract description 15
- 239000010410 layer Substances 0.000 claims abstract description 46
- 239000002356 single layer Substances 0.000 claims abstract description 5
- 238000004891 communication Methods 0.000 claims description 26
- 238000004364 calculation method Methods 0.000 claims description 14
- 239000000126 substance Substances 0.000 claims 1
- 239000000284 extract Substances 0.000 abstract 2
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/104—Peer-to-peer [P2P] networks
- H04L67/1044—Group management mechanisms
Abstract
The invention relates to an unmanned cluster distributed situation maintenance method based on information extraction. The unmanned cluster comprises a plurality of agents, and each agent maintains state potential information of a global layer, a local layer and a single layer; all single-floor situation information in the local area is shared to the temporary information center node of the local area; the local temporary central node extracts information to form local layer attitude information and shares the local layer information to all local intelligent agents; after all local areas finish information extraction and sharing, sharing local layer information extracted by all local area temporary information center nodes to a global temporary information center node; and the global temporary central node extracts information to form global layer situation information, and shares the global situation information to all the intelligent agents of the unmanned cluster. The invention realizes unmanned cluster distributed situation maintenance based on information extraction, and can improve the execution efficiency of unmanned cluster tasks.
Description
Technical Field
The invention belongs to the field of distributed decision making, and particularly relates to an unmanned cluster distributed situation maintenance method based on information extraction.
Background
Since the 90 s of the last century, unmanned systems have gained remarkable development and have been widely used in many fields. However, under complex conditions, the load capacity and range of a single unmanned system are limited, and the requirements cannot be met. Therefore, a plurality of unmanned systems form an interrelated unmanned cluster, and accordingly, the unmanned cluster can cooperatively complete tasks. For an unmanned cluster system, a common task coordination mode is centralized control, and the principle of the coordination mode is simple and clear, so that the coordination mode is applied to various task scenes. However, the centralized cooperative control method is centralized control, and once a central node fails, the task execution performance is seriously reduced.
Another unmanned cluster cooperation mode is distributed cooperation, the cooperation mode is decentralized control, and each node only needs to make a decision according to information mastered by the node, so that the cooperation mode is strong in robustness and environment adaptability, and is paid more and more attention. However, the existing distributed collaboration mode still has a certain problem, which limits further application thereof, and is detailed as follows:
1) A large amount of data needs to be shared, and the task scene with poor communication cannot be adapted;
2) A hierarchical situation is not established, so that the decision is not hierarchical and cannot deal with complex scenes;
3) An information extraction mechanism is not established, and information used for decision making is original information.
In summary, how to solve the current problem of distributed collaboration is the key to further application of distributed collaboration.
Disclosure of Invention
Aiming at the defects of the prior art and the problem that the distributed situation of the unmanned cluster is difficult to maintain in a real complex scene, the invention provides the distributed situation maintenance method of the unmanned cluster based on information extraction, which can well maintain the situation in the distributed coordination of the unmanned cluster.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
an unmanned cluster distributed situation maintenance method based on information extraction, wherein the unmanned cluster comprises a plurality of agents; the method comprises the following steps:
(1) Each agent establishes its own single-layer situation information,
(2) The method comprises the steps that single-machine layer situation information of all agents in a local area is shared to a temporary information center node of the local area;
(3) The temporary information center node of the local area performs information extraction to form local layer state potential information of the corresponding local area, and the local layer information is shared to all intelligent agents in the corresponding local area;
(4) After all local layer information is shared in respective local areas, sharing the local layer information extracted by all local area temporary information center nodes to a global temporary information center node;
(5) And the global temporary information center node performs information extraction to form global layer situation information, and shares the global layer situation information to all intelligent agents in the unmanned cluster.
Further, each agent has communication capability and information sharing capability; the global layer situation information is the overall situation information of the unmanned cluster in the task execution process, and comprises information of task execution efficiency, task execution capacity of the unmanned cluster and resource redundancy; the local layer situation information is situation information of a local area and comprises task execution efficiency, sub-unmanned cluster task execution capacity, target moving trend and resource redundancy information in the local area; the single-machine layer situation information is original information of the intelligent agent and comprises positioning information, task execution capacity information and covered target information.
Further, the selection rule of the temporary information center node in the local area is as follows:
the intelligent agent located in the local area communication topology center can play a role, if the intelligent agent fails, the intelligent agent located in the local area communication topology sub-center can play a role, and the like.
Further, the information extraction modes in the step (3) and the step (5) comprise statistics, weighting and analysis prediction; in the step (3), the temporary information center node in the local area performs information extraction to form local layer attitude information of the corresponding local area; the task execution efficiency is calculated in the following mode:
in the above formula, tee l Indicating the performance of the task in the local area, U lnum Number of agents, te, representing the local area i Indicating the task execution capability of the ith agent, obj lnum Representing the number of coverage targets of the local area;
the calculation mode of the task execution capacity of the sub unmanned cluster is as follows:
in the above formula, te l Indicating the task execution capability of the local area, U lnum Number of agents, te, representing the local area i Representing the task execution capacity of the ith agent;
the calculation mode of the target moving trend is as follows:
Obj ltrend ＝f(Obj 0 ，Obj 1 ，...，Obj 9 )
in the above formula, obj ltrend Represents the moving trend of the target in the local area, obj 0 Representing the cluster centers, obj, of all the objects in the local area of the current time instant 1 Represents the clustering centers of all the targets in the local area at the first 1 moment, and so on, obj 9 Representing the clustering centers of all targets in the local area at the first 10 moments, wherein f represents a target trend prediction function;
the resource redundancy is calculated in the following manner:
in the above formula, U lre Representing the resource redundancy of the local area, U lnum Number of agents, U, representing the local area lenum Representing the number of agents performing the task in the local area;
in the step (5), the global temporary information center node performs information extraction to form global layer situation information; the task execution efficiency is calculated in the following mode:
in the above formula, tee represents the global task execution performance, L represents the number of local regions, tee l Representing the task execution efficiency of the ith local area;
the calculation mode of the task execution capacity of the unmanned cluster is as follows:
in the above formula, te denotes the task execution capability of the global unmanned cluster, te l The task execution capacity of the ith local area is represented;
the resource redundancy calculation mode is as follows:
in the above formula, U re Representing global resource redundancy, U lre The resource redundancy of the l local area is shown.
Further, all the local area temporary communication center nodes form a communication network, and the selection rule of the global temporary information center node is as follows:
the method comprises the steps that an intelligent agent located in a communication topology center of the communication network plays a role, if the intelligent agent fails, the intelligent agent located in a communication topology sub-center of the communication network plays a role, and the like.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention adopts an information extraction mechanism to extract a large amount of original information into a small amount of high-level information, can well limit the communication data volume and solves the problem of excessive bandwidth occupation in the current method.
2. The situation maintenance method can be suitable for distributed decision making, and each intelligent agent can make an autonomous decision based on the existing situation information.
3. The situation information of the invention is divided into a global layer, a local layer and a single machine layer, and the situation information can be used for unmanned cluster global decision, local decision and single machine decision, so that the execution efficiency of unmanned cluster tasks can be improved.
4. The invention adopts a temporary information center node mechanism, and can improve the damage resistance of the unmanned cluster.
Drawings
FIG. 1 is a flow chart of a method of an embodiment of the invention.
Fig. 2 is a schematic diagram of maintenance information of each layer situation in the embodiment of the present invention.
Fig. 3 is a schematic diagram of selecting a temporary information center in a local area according to an embodiment of the present invention.
Fig. 4 is a schematic diagram illustrating selection of a global temporary information center in an embodiment of the present invention.
Detailed Description
The conception, the technical advantages and the technical effects of the present invention will be clearly and completely described in connection with the embodiments, so that the objects, the features and the effects of the present invention can be fully understood. It should be noted that the specific embodiments described herein are only for explaining the present invention, and do not limit the present invention.
An unmanned cluster distributed situation maintenance method based on information extraction, referring to fig. 1, includes the following implementation steps:
step 1, the unmanned cluster includes a plurality of agents, as shown in fig. 2, each agent needs to maintain the situation information of a global layer, a local layer, and a single layer, which is described in detail as follows:
1a) Each agent has communication capacity and information sharing capacity;
1b) The global layer situation information is the overall situation information of the unmanned cluster in the task execution process and comprises information of task execution efficiency, task execution capacity of the unmanned cluster and resource redundancy;
1c) The local layer situation information is the situation information of the local area and comprises information of task execution efficiency, sub unmanned cluster task execution capacity, target moving trend and resource redundancy in the local area;
1d) The single-layer situation information is original information of the agent and comprises information such as positioning, task execution capacity and covered targets;
step 2, sharing all single machine layer situation information in the local area to the temporary information center node of the local area, wherein the specific area selection rule is as follows:
as shown in fig. 3, the local area temporary information center node is assumed to be served by the agent currently located in the area communication topology center, and if the agent fails, the agent currently located in the area communication topology center may serve as the agent, and so on.
And 3, extracting information by the local temporary central node to form local layer state potential information of the local part, and sharing the local layer information to all local intelligent agents, wherein the information extraction mode comprises the following steps: statistical, weighting, analytical prediction and other modes;
step 4, after all local areas finish the steps (2) to (3), sharing the local layer information extracted by all the local area temporary information center nodes to the global temporary information center node;
step 5, the global temporary central node performs information extraction to form global layer situation information, and shares the global situation information to all the agents of the unmanned cluster, and the global temporary central node selection rule and the information extraction mode are described as follows:
5a) The global temporary central node selection rule is as follows: as shown in fig. 4, the global temporary central node is served by the agent of the communication topology center formed by all the local temporary communication nodes, and if the agent fails, the agent of the communication topology center formed by all the local temporary communication nodes can be served by the agent of the communication topology center, and so on.
5b) The information extraction mode is as follows: statistics, weighting, analysis and prediction. The calculation method of the local area task execution efficiency is as follows:
in the above formula, tee l Indicating the performance of the task in the local area, U lnum Number of agents, te, representing the local area i Indicating the task execution capability of the ith agent, obj lnum Representing the number of coverage targets of the local area;
the calculation mode of the task execution capacity of the local area sub unmanned cluster is as follows:
in the above formula, te l Indicating the task execution capability of the local area, U lnum Number of agents, te, representing the local area i Representing the task execution capacity of the ith agent;
the calculation method of the local area target movement trend comprises the following steps:
Obj ltrend ＝f(Obj 0 ，Obj 1 ，...，Obj 9 )
in the above formula, obj ltrend Represents the moving trend of the target in the local area, obj 0 Representing the cluster centers, obj, of all the objects in the local area of the current time instant 1 Represents the clustering centers of all targets in the local area at the first 1 moment, and so on, and Obj 9 Cluster center representing all targets in local area at first 10 momentsF represents a target trend prediction function, and a long-time memory network (namely LSTM) is adopted in the example;
the calculation mode of the local area resource redundancy is as follows:
in the above formula, U lre Indicating the resource redundancy of the local area, U lnum Number of agents, U, representing the local area lenum Representing the number of agents performing the task in the local area;
the calculation method of the execution efficiency of the global layer task is as follows:
in the above formula, tee represents the global task execution performance, L represents the number of local regions, tee l Representing the task execution efficiency of the ith local area;
the computing mode of the task execution capacity of the global layer unmanned cluster is as follows:
in the above formula, te denotes the task execution capability of the global unmanned cluster, te l Indicating the task execution capacity of the ith local area;
the global layer resource redundancy calculation mode is as follows:
in the above formula, U re Representing global resource redundancy, U lre Indicating the resource redundancy of the ith local area.
Claims (5)
1. An unmanned cluster distributed situation maintenance method based on information extraction, wherein the unmanned cluster comprises a plurality of agents; the method is characterized by comprising the following steps:
(1) Each agent establishes its own single-layer situation information,
(2) The method comprises the steps that single-machine layer situation information of all agents in a local area is shared to a temporary information center node of the local area;
(3) The temporary information center node of the local area performs information extraction to form local layer state potential information of the corresponding local area, and the local layer information is shared to all intelligent agents in the corresponding local area;
(4) After all local layer information is shared in respective local areas, sharing the local layer information extracted by the temporary information central nodes of all local areas to a global temporary information central node;
(5) And the global temporary information center node performs information extraction to form global layer situation information, and shares the global layer situation information to all the intelligent agents in the unmanned cluster.
2. The unmanned cluster distributed situation maintenance method based on information extraction as claimed in claim 1, wherein each agent has communication capability and information sharing capability; the global layer situation information is the general situation information of the unmanned cluster in the task execution process, and comprises information of task execution efficiency, unmanned cluster task execution capacity and resource redundancy; the local layer situation information is situation information of a local area, and comprises task execution efficiency, sub unmanned cluster task execution capacity, target moving trend and resource redundancy information in the local area; the single-machine layer situation information is original information of the intelligent agent and comprises positioning information, task execution capacity information and covered target information.
3. The unmanned cluster distributed situation maintenance method based on information extraction as claimed in claim 2, wherein the selection rule of the temporary information center node in the local area is as follows:
the intelligent agent located in the local area communication topology center can play a role, if the intelligent agent fails, the intelligent agent located in the local area communication topology sub-center can play a role, and the like.
4. The unmanned cluster distributed situation maintenance method based on information extraction as claimed in claim 3, wherein the information extraction modes in step (3) and step (5) include statistics, weighting, analysis and prediction; in the step (3), the temporary information center node in the local area performs information extraction to form local layer state potential information of the corresponding local area; wherein the content of the first and second substances, the task execution efficiency is calculated in the following way:
in the above formula, tee l Indicating the performance of the task in the local area, U lnum Number of agents, te, representing the local area i Indicating the task execution capability of the ith agent, obj lnum Representing the number of coverage targets of the local area;
the calculation mode of the task execution capacity of the sub unmanned cluster is as follows:
in the above formula, te l Indicating the task execution capability of the local area, U lnum Number of agents, te, representing the local area i Representing the task execution capacity of the ith agent;
the calculation mode of the target moving trend is as follows:
Obj ltrend ＝f(Obj 0 ，Obj 1 ，...，Obj 9 )
in the above formula, obj ltrend Represents the moving trend of the target in the local area, obj 0 Representing the cluster centers, obj, of all the objects in the local area of the current time instant 1 Indicates the local area at the first 1 timeClustering centers of all targets, and so on, obj 9 Representing the clustering centers of all targets in the local area at the first 10 moments, wherein f represents a target trend prediction function;
the resource redundancy is calculated in the following manner:
in the above formula, U lre Indicating the resource redundancy of the local area, U lnum Number of agents, U, representing the local area lenum Representing the number of agents performing the task in the local area;
in the step (5), the global temporary information center node performs information extraction to form global layer situation information; the task execution efficiency is calculated in the following mode:
in the above formula, tee represents the global task execution performance, L represents the number of local areas, tee l Representing the task execution efficiency of the ith local area;
the calculation mode of the unmanned cluster task execution capacity is as follows:
in the above formula, te denotes the task execution capability of the global unmanned cluster, te l The task execution capacity of the ith local area is represented;
the resource redundancy calculation mode is as follows:
in the above formula, U re Representing global resource redundancy, U lre The resource redundancy of the l local area is shown.
5. The unmanned cluster distributed situation maintenance method based on information extraction as claimed in claim 4, wherein all local area temporary communication center nodes form a communication network, and the selection rule of the global temporary information center node is:
the method comprises the steps that an intelligent agent located in a communication topological center of the communication network serves as the intelligent agent, if the intelligent agent breaks down, the intelligent agent located in a communication topological secondary center of the communication network serves as the intelligent agent, and the like.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210703875.2A CN115150408A (en) | 2022-06-21 | 2022-06-21 | Unmanned cluster distributed situation maintenance method based on information extraction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210703875.2A CN115150408A (en) | 2022-06-21 | 2022-06-21 | Unmanned cluster distributed situation maintenance method based on information extraction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115150408A true CN115150408A (en) | 2022-10-04 |
Family
ID=83408841
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210703875.2A Pending CN115150408A (en) | 2022-06-21 | 2022-06-21 | Unmanned cluster distributed situation maintenance method based on information extraction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115150408A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090083390A1 (en) * | 2007-09-24 | 2009-03-26 | The Research Foundation Of State University Of New York | Automatic clustering for self-organizing grids |
| CN113286275A (en) * | 2021-04-23 | 2021-08-20 | 南京大学 | Unmanned aerial vehicle cluster efficient communication method based on multi-agent reinforcement learning |
| CN113299120A (en) * | 2021-05-25 | 2021-08-24 | 中国电子科技集团公司第二十八研究所 | Intelligent sensing system for air traffic situation supported by edge cloud in cooperation |
-
2022
- 2022-06-21 CN CN202210703875.2A patent/CN115150408A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090083390A1 (en) * | 2007-09-24 | 2009-03-26 | The Research Foundation Of State University Of New York | Automatic clustering for self-organizing grids |
| CN113286275A (en) * | 2021-04-23 | 2021-08-20 | 南京大学 | Unmanned aerial vehicle cluster efficient communication method based on multi-agent reinforcement learning |
| CN113299120A (en) * | 2021-05-25 | 2021-08-24 | 中国电子科技集团公司第二十八研究所 | Intelligent sensing system for air traffic situation supported by edge cloud in cooperation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109996039B (en) | Target tracking method and device based on edge calculation | |
| CN111342993B (en) | SDN-based air-space-ground controller deployment architecture and control method | |
| CN112488322A (en) | Federal learning model training method based on data feature perception aggregation | |
| Lv | Construction of marine ship automatic identification system data mining platform based on big data | |
| CN115209426A (en) | Dynamic deployment method of digital twin servers in edge internet of vehicles | |
| Chen et al. | 5G boosting smart cities development | |
| CN114710330A (en) | Anomaly detection method based on heterogeneous hierarchical federated learning | |
| CN115150408A (en) | Unmanned cluster distributed situation maintenance method based on information extraction | |
| Li et al. | CEBOW: A Cloud‐Edge‐Browser Online Web3D approach for visualizing large BIM scenes | |
| CN111770133B (en) | Multi-body intelligent-connection cloud control platform | |
| CN113360245A (en) | Internet of things equipment task downloading method based on mobile cloud computing deep reinforcement learning | |
| CN112218052A (en) | Comprehensive security centralized supervision platform and application thereof | |
| CN112965381A (en) | Method for establishing cooperative intelligent self-adaptive decision model | |
| CN116455768A (en) | Cloud edge end collaborative CNN reasoning method and system for global time delay optimization | |
| CN106210624A (en) | A kind of method of electrically-based industry multi-stage cascade Streaming Media intelligent scheduling | |
| CN113489942A (en) | Video monitoring system based on edge calculation | |
| US20210334649A1 (en) | Apparatus and method for distributed reinforcement learning | |
| Jiang et al. | Research on location planning of 5G base station based on DBSCAN clustering algorithm | |
| CN113223006A (en) | Lightweight target semantic segmentation method based on deep learning | |
| CN113179496A (en) | Video analysis framework based on MEC and indoor positioning system under framework | |
| Shuai et al. | Memtv: a research on multi-level edge computing model for traffic video processing | |
| CN117726153B (en) | Real-time re-planning method suitable for unmanned aerial vehicle cluster operation tasks | |
| CN112637900B (en) | Mobile ad hoc cloud terminal cluster construction and service method based on social perception | |
| Chen et al. | An energy-efficient task offloading decision in electric power IoT based on edge computing | |
| CN116744368B (en) | Intelligent collaborative heterogeneous air-ground unmanned system based on cloud side end architecture and implementation method |
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 |