A service-oriented intelligent ooda loop implementation method and system

Abstract

The application relates to the field of embedded intelligent computing, and provides a service-oriented intelligent OODA loop implementation method and system, which comprises the following steps: formally defining a service description format for intelligent services, the intelligent services including a perception type service, a cognition type service, a decision type service and an execution type service, and the intelligent services being used for constructing the OODA loop; applying the service description format to the intelligent services, the intelligent services being run based on an embedded intelligent basic computing platform; and according to the computing characteristics of the intelligent services, distributing the intelligent services to different computing resources of the embedded intelligent basic computing platform through a service resource mapping method. The technical problems of intelligent upgrading and service upgrading of the OODA loop in the prior art are solved, the perception, cognition, decision and execution processes in the OODA loop are encapsulated into services, a complete service description is provided for service combination and resource distribution, and the intelligent upgrading and service upgrading of the OODA loop are realized.

Description

Technical Field

[0001] This invention relates to the field of embedded intelligent computing, and in particular to a service-oriented intelligent OODA loop implementation method and system. Background Technology

[0002] The Observation-Orientation-Decision-Action (OODA) loop theory decomposes combat operations into four processes: perception, judgment, decision-making, and execution. Efficiently and rapidly completing the OODA loop has become a crucial winning mechanism in combat operations. With the rapid development of information technology, leveraging artificial intelligence to empower the OODA loop has become a mainstream development trend, and there is a consensus that "whoever can achieve intelligent OODA loops first will win future wars."

[0003] Intelligent technologies have enhanced the information processing capabilities of the OODA loop, making it possible to implement more functions by designing multiple intelligent algorithms in each process of the OODA loop, thereby improving adaptability to various combat scenarios. The increased number of algorithms presents challenges to flexible deployment and the establishment of communication networks between them. Introducing a service-oriented architecture can improve deployment efficiency and manage the communication network. Service-oriented OODA loop design has become the development trend of next-generation OODA loop architectures.

[0004] Realizing a service-oriented intelligent OODA loop faces two main challenges: first, how to achieve intelligent upgrades to the OODA loop, specifically including how to build an intelligent computing platform and how to map intelligent service domain computing resources; second, how to achieve service-oriented upgrades to the OODA loop, specifically including how to describe intelligent services and how to ensure secure communication between intelligent services. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a service-oriented intelligent OODA loop implementation method to solve the technical challenges of intelligent and service-oriented upgrades of OODA loops in the prior art. The method includes:

[0006] A formal definition of a service description format for intelligent services is provided. The intelligent services include perception services, cognition services, decision-making services, and execution services. The intelligent services are used to construct the OODA loop.

[0007] The service description format is applied to the intelligent service, which runs on an embedded intelligent basic computing platform;

[0008] Based on the computational characteristics of the intelligent service, the intelligent service is allocated to different computing resources of the embedded intelligent basic computing platform through a service resource mapping method.

[0009] Furthermore, the service description format of the intelligent service is: service i ={id i FS i RS i}

[0010] Furthermore, the communication application security governance method between the intelligent services ensures communication security.

[0011] Furthermore, the security governance methods for communication applications between intelligent service A and intelligent service B include:

[0012] Intelligent Service A registers with the Security Control Center. After verifying Intelligent Service A, the Security Control Center provides a list of corresponding services in the registration service information.

[0013] The security control center notifies intelligent service B that it can accept the request from intelligent service A and sets a security policy.

[0014] The security control center notifies intelligent service A of the service list and the security policy.

[0015] The intelligent service A communicates with the intelligent service B according to the service list and the security policy.

[0016] Furthermore, service resource mapping methods include static scheduling and dynamic scheduling.

[0017] Furthermore, the embedded intelligent basic computing platform includes a variety of chip resources.

[0018] Furthermore, this invention also provides a service-oriented intelligent OODA loop implementation system to solve the technical challenges of intelligent and service-oriented upgrades of OODA loops in the prior art. The system includes:

[0019] The service description format definition module is used to formally define the service description format for intelligent services, which include perception services, cognition services, decision-making services and execution services. The intelligent services are used to construct the OODA loop.

[0020] The intelligent service operation module is used to apply the service description format to the intelligent service, which runs on an embedded intelligent basic computing platform.

[0021] The computing resource allocation module is used to allocate the intelligent service to different computing resources of the embedded intelligent basic computing platform according to the computing characteristics of the intelligent service through a service resource mapping method.

[0022] Compared with existing technologies, the beneficial effects achieved by at least one of the above-mentioned technical solutions adopted in the embodiments of this specification include at least the following: This invention provides a service-oriented intelligent OODA loop implementation method, comprising: formally defining a service description format for intelligent services, wherein the intelligent services include perception services, cognition services, decision-making services, and execution services, and the intelligent services are used to construct the OODA loop; applying the service description format to the intelligent services, wherein the intelligent services run on an embedded intelligent basic computing platform; and allocating the intelligent services to different computing resources of the embedded intelligent basic computing platform according to the computing characteristics of the intelligent services through a service resource mapping method. The perception, cognition, decision-making, and execution processes in the OODA loop are encapsulated as services, and a complete service description provides guarantees for service composition and resource allocation, thereby realizing the intelligent and service-oriented upgrade of the OODA loop. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of a service-oriented intelligent OODA loop implementation method provided by an embodiment of the present invention;

[0025] Figure 2 This is a schematic diagram of the service-level authentication model provided in an embodiment of the present invention;

[0026] Figure 3 This is a system architecture diagram of a service-oriented intelligent OODA loop implementation provided by an embodiment of the present invention.

[0027] The attached diagram is labeled as follows: 300, System; 301, Service Description Format Definition Module; 302, Intelligent Service Operation Module; 303, Computing Resource Allocation Module. Detailed Implementation

[0028] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0029] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. This application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] This invention provides a service-oriented intelligent OODA loop implementation method, comprising: formally defining a service description format for intelligent services, wherein the intelligent services include perception services, cognition services, decision-making services, and execution services, and the intelligent services are used to construct the OODA loop; applying the service description format to the intelligent services, wherein the intelligent services run on an embedded intelligent basic computing platform; and allocating the intelligent services to different computing resources of the embedded intelligent basic computing platform according to the computational characteristics of the intelligent services through a service resource mapping method. By encapsulating the perception, cognition, decision-making, and execution processes in the OODA loop as services, and providing guarantees for service composition and resource allocation through complete service descriptions, the intelligent and service-oriented upgrade of the OODA loop is achieved.

[0031] The embodiments of the present invention specifically include the following steps:

[0032] Step S100: Formalize the service description format for intelligent services, wherein the intelligent services include perception services, cognition services, decision-making services and execution services, and the intelligent services are used to construct the OODA loop;

[0033] Step S200: Apply the service description format to the intelligent service, which runs on an embedded intelligent basic computing platform;

[0034] Step S300: Based on the computational characteristics of the intelligent service, the intelligent service is allocated to different computing resources of the embedded intelligent basic computing platform through a service resource mapping method.

[0035] Specifically, such as Figure 1As shown, this embodiment provides a service-oriented intelligent OODA loop implementation method. Based on OODA loop theory, it designs four types of services based on intelligent algorithms: perception, cognition, decision-making, and execution. Perception services are used for data acquisition, cognition services for data analysis, decision-making services for planning, and execution services for implementing the plans. Each service type includes multiple specific services; for example, cognition services include information fusion and target recognition. The method formally defines the service description format for intelligent services and establishes a service-oriented communication system. It implements service-oriented and security governance to ensure the security of OODA loop communication and designs an embedded intelligent basic computing platform to meet the high computing power requirements of intelligent services.

[0036] During the operational planning phase, intelligent algorithms are designed for each type of service. The types, names, input parameters, and outputs of these algorithms are described. The computational, storage, and network resource requirements for each algorithm are assessed and calculated. Offline allocation algorithms, such as mixed-integer programming, are used to allocate computing chips, memory, and bandwidth to the intelligent services, supporting the resource demands. During the execution of the operational mission, if unforeseen events render the offline allocation scheme infeasible, online allocation methods, such as genetic engineering, are used to quickly regenerate the allocation scheme.

[0037] Preferably, the service description format of the intelligent service is: service i ={id i FS i RS i}. Where id i The unique identifier for the task, FS i The basic description includes attributes such as service type, service name, input parameters, output format, and service description; RS i For resource-related descriptions, it can be represented as RS i ={R i S i D i}, R i S represents the task's requirements for various resources. i Indicates the type of resource used at runtime, D i Indicates the runtime of the specified task.

[0038] Furthermore, the unique identifier id for the computation task. i Used to uniquely identify computational tasks; Source requirement R i Represented as R i ={rc i ,rm i rB i}, that is, computation task t iThe requirements for computing, storage, and networking are respectively determined by rc i rm i rB i Indicates the computational requirement rc. i This can be expressed as the computing power required to execute a task, i.e., the number of computational operations. Because the amount of computation required during actual task execution depends on the actual load, the computational requirement rc... i It can be represented as a load-related function, i.e., rc i =f c i (·), where f c i The parameter (·) is a numerical value describing the load size. For example, if the load is a three-dimensional matrix, then f c i The parameters of (·) can be the length, width, and height of a three-dimensional matrix. Similarly, the storage requirement rm i This can be expressed as the amount of memory space required to execute a task, which also depends on the actual workload and can be expressed as rm. i =f mi (·). For network demand rB i Used to measure the communication volume between computational tasks, involving multiple computational tasks, it can be represented as rB. i ={rb i,j |j∈[1,n]}, where rb i,j The network communication resource requirements between computation task i and computation task j can be represented as rb. i，j =f b i,j (·), where f b i,j The parameter (·) is a value describing the communication overhead between computation task i and computation task j.

[0039] Furthermore, specifying case S i It can be represented as S i ={sC i sM i sB i}, that is, computation task t i The allocation of computing, storage, and network resources is determined by sC. i sM i sB i Indicated. Where sC i sM i sB i All can be further subdivided according to the actual situation, such as sC i It can be represented as sC i ={sc i cpu sci gpu sc i fpga sc i asic}, thus describing task t respectively i Specifications for CPU, GPU, FPGA, and ASIC. Specify runtime D. i The minimum runtime ds for the computation task is specified. i and longest runtime dl i D i ={ds i dl i}. Where ds i and dl i It can be derived theoretically or specified by the task issuer. If we consider ds... i If there are no requirements, then ds can be set. i =0, if for dl i If there are no requirements, then dl can be set. i =+∞, for each task D i The settings.

[0040] Preferably, this invention designs a zero-trust-based service authentication model for service-oriented architecture. Before communication, both parties must authenticate each other through a security control center. The involvement of the security control center enables security measurement of communication services and the planning of communication security policies to prevent malicious service intrusion. Figure 2 As shown, when service A communicates with service B, the authentication process is as follows:

[0041] (1) Service A registers with the security control center. After verifying Service A, the security control center provides a list of services in the registered service information.

[0042] (2) The controller notifies service B that it can accept service A's request and sets a security policy.

[0043] (3) The controller notifies Service A of the list of available services and the corresponding security policies.

[0044] (4) After receiving the message, Service A communicates with Service B.

[0045] Optional, such as Figure 1As shown, the basic computing platform in this invention includes various chips such as multi-core CPUs, NPUs, FPGAs, and GPUs. Different chips support different upper-layer applications based on their characteristics. Specifically, multi-core CPUs are used for calculations with many decisions and branches, such as heuristic optimization algorithms used in decision-making processes and control algorithms used in execution processes; NPUs are used to run neural networks with low precision requirements, such as target recognition in cognitive processes; FPGAs are used to run algorithms with higher precision requirements, such as game algorithms based on deep learning in decision-making processes; and GPUs are used to run traditional image enhancement algorithms, such as image enhancement and noise reduction algorithms in perception processes.

[0046] Optionally, the design incorporates a two-stage method for intelligent service allocation: offline and online allocation. The intelligent service allocation method is used to allocate appropriate physical resources such as chips, memory, and networks to the intelligent services. The offline allocation method in this invention is used in the operational mission planning phase and can employ optimal methods such as mixed-integer programming, characterized by its long execution time and high accuracy. The online allocation method is used in the operational mission execution phase and can employ heuristic algorithms such as genetic algorithms, characterized by its accuracy increasing with runtime.

[0047] Based on the same inventive concept, this invention also provides a service-oriented intelligent OODA loop implementation system, as described in the following embodiments. Since the principle of a service-oriented intelligent OODA loop implementation system in solving the problem is similar to that of a service-oriented intelligent OODA loop implementation method, the implementation of a service-oriented intelligent OODA loop implementation method can refer to the implementation of a service-oriented intelligent OODA loop implementation system; repeated details will not be elaborated further. As used below, the terms "unit" or "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the system described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0048] like Figure 3 The diagram shown is a schematic of a service-oriented intelligent OODA loop implementation system 300 according to an embodiment of the present invention. Specific steps include:

[0049] The service description format definition module 301 is used to formally define the service description format for intelligent services, which include perception services, cognition services, decision-making services and execution services. The intelligent services are used to construct the OODA loop.

[0050] The intelligent service operation module 302 is used to apply the service description format to the intelligent service, which runs on an embedded intelligent basic computing platform.

[0051] The computing resource allocation module 303 is used to allocate the intelligent service to different computing resources of the embedded intelligent basic computing platform according to the computing characteristics of the intelligent service through a service resource mapping method.

[0052] The embodiments of the present invention achieve the following technical effects:

[0053] This invention designs a service-oriented intelligent OODA loop implementation method and system, customizes an intelligent service description method for OODA loops, constructs a zero-trust service-level authentication system, integrates multiple intelligent chips to support diverse intelligent algorithms in OODA loops, and designs offline and online allocation methods to complete service and resource mapping at different stages. It efficiently realizes the service-oriented and intelligent upgrade of OODA loops for battlefield environments, and improves the degree of autonomy and adaptability in the face of complex battlefield environments.

[0054] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations can be made to the embodiments of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A service-oriented intelligent OODA loop implementation method, characterized in that, include: A formal definition of a service description format for intelligent services is provided. The intelligent services include perception services, cognition services, decision-making services, and execution services. The intelligent services are used to construct the OODA loop. The service description format is applied to the intelligent service, which runs on an embedded intelligent computing platform; the service description format is: service i ={id i FS i RS i }, where id i The unique identifier for the task, FS i Based on the description, RS i For resource-related descriptions; RS i Represented as RS i ={R i S i D i }, R i S represents the task's requirements for various resources. i Indicates the type of resource used at runtime, D i Indicates the runtime of the specified task; Resource Requirements R i Represented as R i ={rc i ,rm i rB i }, that is, computation task t i The requirements for computing, storage, and networking are respectively determined by rc i rm i rB i express; Specify resource usage type S i Represented as S i ={sC i sM i sB i }, that is, computation task t i The allocation of computing, storage, and network resources is determined by sC. i sM i sB i express; Specify runtime D i The minimum runtime ds for the computation task is specified. i and longest runtime dl i D i ={ds i dl i }; Based on the computational characteristics of the intelligent service, the intelligent service is allocated to different computing resources of the embedded intelligent basic computing platform through a service resource mapping method. The computational characteristics include the service type, service name, input parameters, and resource requirements of the intelligent algorithm defined in the service description format. The communication security governance method between the intelligent services ensures communication security. The security governance methods for communication between intelligent service A and intelligent service B include: Intelligent Service A registers with the Security Control Center. After verifying Intelligent Service A, the Security Control Center provides a corresponding service list in the registration service information. The security control center notifies intelligent service B to accept the request from intelligent service A and sets a security policy; The security control center notifies intelligent service A of the service list and the security policy. The intelligent service A communicates with the intelligent service B according to the service list and the security policy; Service resource mapping methods include static scheduling and dynamic scheduling.

2. The method for implementing a service-oriented intelligent OODA loop according to claim 1, characterized in that, Embedded intelligent basic computing platforms include a variety of chip resources.

3. A service-oriented intelligent OODA loop implementation system, characterized in that, The system includes: The service description format definition module is used to formally define the service description format for intelligent services, which include perception services, cognition services, decision-making services and execution services. The intelligent services are used to construct the OODA loop. The intelligent service operation module is used to apply the service description format to the intelligent service, which runs on an embedded intelligent basic computing platform; the service description format is: service i ={id i FS i RS i }, where id i The unique identifier for the task, FS i Based on the description, RS i For resource-related descriptions; RS i Represented as RS i ={R i S i D i }, R i S represents the task's requirements for various resources. i Indicates the type of resource used at runtime, D i Indicates the runtime of the specified task; Resource Requirements R i Represented as R i ={rc i ,rm i rB i }, that is, computation task t i The requirements for computing, storage, and networking are respectively determined by rc i rm i rB i express; Specify resource usage type S i Represented as S i ={sC i sM i sB i }, that is, computation task t i The allocation of computing, storage, and network resources is determined by sC. i sM i sB i express; Specify runtime D i The minimum runtime ds for the computation task is specified. i and longest runtime dl i D i ={ds i dl i The computing resource allocation module is used to allocate the intelligent service to different computing resources of the embedded intelligent basic computing platform according to the computing characteristics of the intelligent service through a service resource mapping method. The computing characteristics include the service type, service name, input parameters, and resource requirements of the intelligent algorithm defined in the service description format. The communication security governance method between the intelligent services ensures communication security. The security governance methods for communication between intelligent service A and intelligent service B include: Intelligent Service A registers with the Security Control Center. After verifying Intelligent Service A, the Security Control Center provides a corresponding service list in the registration service information. The security control center notifies intelligent service B to accept the request from intelligent service A and sets a security policy; The security control center notifies intelligent service A of the service list and the security policy. The intelligent service A communicates with the intelligent service B according to the service list and the security policy; Service resource mapping methods include static scheduling and dynamic scheduling.

Images