Cooperative task efficiency evaluation method, system and device

Abstract

The invention discloses a collaborative task efficiency evaluation method, system and device, and the method comprises the steps: carrying out the state description and analysis of a task, and constructing an evaluation index system; constructing a simulation model according to the task; and performing index screening and weight distribution based on the evaluation index system, and mapping to the simulation model. The system comprises a task description and evaluation system module, a simulation module and an efficiency evaluation module. The device comprises a memory and a processor used for executing the collaborative task efficiency evaluation method. By using the method, the key problems that the collaborative task efficiency of the heterogeneous unmanned platform is difficult to describe and measure, the design efficiency is low, the intelligent degree is low and the like can be solved. The method can be widely applied to the field of evaluation systems.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of evaluation system, in particular to a collaborative task performance evaluation method, system and device. BACKGROUND

[0002] The performance evaluation technology of a large number of heterogeneous unmanned platforms for collaborative task implementation refers to the technology for improving the completion efficiency and quality of tasks by evaluating the collaborative performance between a plurality of unmanned platforms of different types, different manufacturers or different versions, which are operated in collaboration. This technology involves multiple aspects, including collaborative design and optimization of communication, control, perception and decision-making between heterogeneous platforms, and evaluation and optimization of the effect of collaborative implementation. At the same time, this technology also needs to consider various factors in actual application scenarios, such as environment and resource constraints. The performance evaluation technology of a large number of heterogeneous unmanned platforms for collaborative task implementation can be applied in multiple fields, such as unmanned aerial vehicles, robots, etc.

[0003] There are a large number of heterogeneous unmanned platforms for collaborative tasks, and the capability characteristics of each unmanned platform are quite different. At present, there is a lack of effective unified task performance measurement model and evaluation system. SUMMARY

[0004] Therefore, in order to solve the problem that the existing technology lacks a unified and standardized evaluation index and method for the performance evaluation of a large number of heterogeneous unmanned platforms for collaborative task implementation, which leads to the difficulty in comparing the evaluation results between different researchers, the present application provides a collaborative task performance evaluation method, system and device. The method comprises the following steps:

[0005] State description and analysis of the task are performed to construct an evaluation index system;

[0006] A simulation model is constructed according to the task;

[0007] Index screening and weight distribution are performed based on the evaluation index system, and are mapped to the simulation model.

[0008] In some embodiments, the step of state description and analysis of the task to construct an evaluation index system specifically comprises:

[0009] The collaborative task is decomposed and a distribution model is constructed according to the target and characteristics of the collaborative task to obtain a collaborative task description model;

[0010] Based on the collaborative task description model, qualitative and quantitative analysis is performed to construct a task performance measurement model;

[0011] The core influencing factors of collaborative task performance are studied by using the analytic hierarchy process to construct an evaluation index system based on the task overall performance layer, local performance layer and platform performance layer.

[0012] In some embodiments, the cooperative task description model is specifically a conflict-free design of the task state of perception, communication and decision in the time and space dimensions by using a finite state machine; and the task effectiveness includes a task collateral effect, a duration effect and a satisfaction effect.

[0013] Through the preferred step of superimposing the self-coherent hologram and then performing back propagation, sample data can be increased, which is aimed at the case of less sample.

[0014] In some embodiments, it further includes:

[0015] The task overall effectiveness layer adopts an analysis method combining the overall function of the executed task and the characteristic attribute to construct a task evaluation index;

[0016] The local effectiveness layer adopts a factor analysis method to extract common influence factors to construct a local evaluation index;

[0017] The platform effectiveness layer adopts a data statistical analysis to analyze and classify the platform capability to construct a platform effectiveness evaluation index.

[0018] In some embodiments, the step of constructing a simulation model according to the task specifically includes:

[0019] Based on a preset engine, a platform simulation model is constructed by combining data analysis and machine learning;

[0020] Based on an adaptive robust estimation and control method, an intelligent behavior simulation model is constructed by combining artificial intelligence;

[0021] An information network simulation model is constructed by using a spatial distributed self-organizing ad hoc network method.

[0022] In some embodiments, the platform simulation model includes a platform situation awareness capability simulation model, a platform motion capability simulation model, a platform communication capability simulation model and a platform survival capability simulation model; the intelligent behavior simulation model includes an autonomous learning capability simulation model, an intelligent interaction capability simulation model and a dynamic cooperation capability simulation model; and the information network simulation model includes an information networking simulation model, a communication mapping simulation model and a distributed information integration simulation model.

[0023] In some embodiments, the step of performing index screening and weight distribution based on the evaluation index system and mapping to the simulation model specifically includes:

[0024] An index screening strategy is constructed based on a deep learning model by comprehensively considering expert experience;

[0025] A clustering method is used to merge the indexes to form core indexes;

[0026] A mapping relationship between a task performance result and the core indicators is constructed based on a neural network, and weights of each indicator are designed.

[0027] The application further provides a cooperative task performance evaluation system, which comprises:

[0028] A task description and evaluation system module is used for state description and analysis of the task, and an evaluation index system is constructed.

[0029] A simulation module is used for constructing a simulation model according to the task.

[0030] An performance evaluation module is used for index screening and weight distribution based on the evaluation index system, and is mapped to the simulation model.

[0031] The application further provides a cooperative task performance evaluation device, which comprises:

[0032] At least one processor;

[0033] At least one memory for storing at least one program;

[0034] When the at least one program is executed by the at least one processor, the at least one processor realizes the cooperative task performance evaluation method as described above.

[0035] Based on the above scheme, the application provides a cooperative task performance evaluation method, system and device, which can solve the key problems of difficult description and measurement of the cooperative task performance of the heterogeneous unmanned platform, low design efficiency and low intelligent degree, and form a virtual-real combined heterogeneous unmanned platform system design optimization intelligent auxiliary theory and method system. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 is a step flow chart of a cooperative task performance evaluation method of the application;

[0037] Figure 2 is a process schematic diagram of the step of state description and analysis of the task, construction of the evaluation index system and the like by the specific embodiment of the application;

[0038] Figure 3 is a process schematic diagram of the step of constructing a simulation model according to the task by the specific embodiment of the application;

[0039] Figure 4 is a process schematic diagram of the step of index screening and weight distribution based on the evaluation index system, and mapping to the simulation model by the specific embodiment of the application. DETAILED DESCRIPTION

[0040] With reference to the drawings, the technical solutions in the embodiments of the present application will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative effort are within the protection scope of the present application.

[0041] It should be noted that, for the convenience of description, only the parts related to the present application are shown in the drawings. The embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0042] It should be understood that the "system", "device", "unit" and / or "module" used in the present application is a method for distinguishing different components, elements, parts, sections or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0043] As shown in the present application and claims, unless the context clearly indicates otherwise, "one", "a", "an" and / or "the" do not refer to the singular, but also include the plural. Generally, the terms "comprise" and "include" only indicate that the steps and elements explicitly identified are included, and these steps and elements do not constitute an exclusive list, and the method or device can also include other steps or elements. The element defined by the statement "comprises a" does not exclude the presence of another element in the process, method, product or device comprising the element.

[0044] In the description of the embodiments of the present application, "a plurality of" means two or more than two. The following terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features.

[0045] In addition, flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or subsequent operations are not necessarily performed in sequence. On the contrary, each step can be processed in reverse order or simultaneously. At the same time, other operations can be added to these processes, or one or more steps of the operation can be removed from these processes.

[0046] Reference Figure 1 The flowchart of an optional example of the collaborative task effectiveness evaluation method proposed in the present application can be applied to a computer device. The evaluation method proposed in the embodiment can include but is not limited to the following steps:

[0047] S1, state description and analysis of the task, and construction of an evaluation index system;

[0048] Taking the cooperative operation of a UAV and an unmanned vehicle in a search and rescue task as an example, this step mainly includes:

[0049] S1.1, according to the target and characteristics of the cooperative task, decompose the cooperative task and build a distribution model to obtain a cooperative task description model;

[0050] Among them, according to the target and characteristics of the cooperative task, the stage decomposition and modeling method is adopted to complete the construction of the cooperative task decomposition and distribution model, and the vehicle-machine cooperative task is decomposed into an aerial search stage and a ground search and rescue stage; In view of the sudden situations such as internal and external interference, attack, communication network failure, etc., an online self-repair task allocation method based on event-triggered feedback mechanism is designed, so that when the UAV or unmanned vehicle is interfered or attacked, the other party can adjust the task allocation in real time to ensure the continuity and effectiveness of the task; For the cooperative task scheduling problem, a multi-level distributed task scheduling framework is designed to complete the construction of the heterogeneous unmanned platform cooperative task balanced scheduling model; The conflict-free design of perception, communication, decision and other task states in the time and space dimensions is completed by using finite state machine and other methods, and efficient task state description is completed.

[0051] S1.2, based on the cooperative task description model, qualitative and quantitative analysis is carried out to build a task efficiency measurement model;

[0052] Among them, based on the cooperative task description model, the qualitative and quantitative method is adopted to establish the task efficiency measurement model of the task dependent effect, duration effect and satisfaction effect; Through the establishment of the quantitative characteristic function of the openness of the heterogeneous unmanned platform in the architecture, the connectivity of data strategy and permission, the agility of data transmission integration, etc., the "man-machine" and "machine-machine" cooperative interoperability measurement model is established; Through the Monte Carlo method, the damage fatigue of the unmanned platform module, the intelligent behavior ability of anti-interference, the interception rate of system network information sharing, the cooperative redundancy of heterogeneous platform, etc. are studied to build the task efficiency resilience measurement model; Through the test and analysis of the intelligent behavior ability of the unmanned platform in the multi-domain environment such as autonomous navigation, situation awareness, cooperative decision-making in multiple task scenarios, the task efficiency autonomy measurement model is built; Through the study of the influence of platform model uncertainty, external interference, network information guarantee quality, "man-machine" and "machine-machine" interaction reliability on reliability, the characteristic weight is determined by using fuzzy rule, and the task efficiency reliability measurement model of heterogeneous unmanned platform is established.

[0053] S1.3, the core influencing factors of cooperative task efficiency are studied by using the analytic hierarchy process, and an evaluation index system based on the overall task efficiency layer, the local efficiency layer and the platform efficiency layer is constructed.

[0054] The core influencing factors of the cooperative task efficiency of the heterogeneous unmanned platform are studied by using the analytic hierarchy process, and a three-layer evaluation index system based on the task overall efficiency layer, the local efficiency layer and the platform efficiency layer is established. In the task overall efficiency layer, the analysis method combining the overall function and characteristic attributes of the heterogeneous unmanned platform executing the task is adopted to construct the task evaluation index system in multiple aspects such as task completion efficiency, interoperability, resilience, autonomy, reliability and task execution cost; in the local efficiency layer, the factor analysis method is adopted to extract common influencing factors to complete the refinement and decomposition of the evaluation index in the local efficiency layer in the cooperative task execution process; in the platform efficiency layer, the data statistical processing method is adopted to analyze and classify the core capabilities such as the hardware and software architecture, navigation capability, perception capability, control decision capability and communication capability of the heterogeneous platform to construct the evaluation index reflecting the platform efficiency layer.

[0055] The overall process and structure of step S1 are shown in Figure 2

[0056] S2, constructing a simulation model according to the task;

[0057] S2.1, constructing a platform simulation model based on a preset engine, combining data analysis and machine learning;

[0058] Based on the physical / rendering engine and the deep learning method, the simulation model construction of the situation awareness capability of the heterogeneous unmanned platform such as target detection and identification is realized by analyzing the environmental characteristics such as target attributes, perceptible range and terrain; the simulation model construction of the motion capability of the unmanned platform is realized by analyzing the motion attributes, control strategy and load attributes of the heterogeneous unmanned platform in the task completion process, adopting the multi-physical engine and componentized modeling technology; the simulation model construction of the communication capability of the unmanned platform is realized by adopting the method of machine learning based on data link, communication node and communication fault; the simulation model of the survival capability is established by analyzing the influence weight of the platform environmental adaptability, accident rate and concealment characteristics on the survival capability and constructing the probability density function thereof.

[0059] S2.2, constructing an intelligent behavior simulation model based on adaptive robust estimation and control method, combining artificial intelligence;

[0060] By analyzing the internal and external interference and sudden conditions of the complex multi-domain environment, the simulation model of the environmental adaptability is established by using the adaptive robust estimation and control method; the simulation model of the online autonomous learning capability of the unmanned platform based on the artificial intelligence algorithm is constructed by simulating the cooperative task scene; the simulation model of the intelligent interaction capability is realized by using the “human-machine” and “machine-machine” interactive interface technology, combining the communication protocol optimization and interactive data characteristics; the simulation model of the dynamic cooperation capability is constructed by building the “sensing-perception-decision-control” cooperation system based on the multi-agent distributed cooperative control method. ​

[0061] S2.3, using a spatially distributed self-organizing ad hoc network method, an information network simulation model is constructed.

[0062] A low-cost high-performance information networking simulation model is established by using a spatially distributed self-organizing networking method combined with a multi-channel multi-node data acquisition and publishing method. By analyzing the communication types, data transmission real-time performance, and communication synchronization requirements of heterogeneous unmanned platforms, a communication protocol mapping simulation model is constructed. Based on data access and integration middleware technology based on pattern integration and data replication, combined with a distributed data integration sharing framework, a distributed information integration simulation model is constructed.

[0063] The overall process and structure of step S2 are as shown in Figure 3 .

[0064] S3, based on the evaluation index system, index screening and weight allocation are performed, and are mapped to the simulation model.

[0065] S3.1, considering expert experience, an index screening strategy is constructed based on a deep learning model;

[0066] S3.2, a clustering method is used to merge the indexes to form core indexes;

[0067] For different collaborative task performance evaluation indexes, on the basis of comprehensive consideration of expert experience, a simplified and complete index screening strategy is constructed based on a deep learning model. To alleviate the influence of the increase in the number of unmanned platform nodes on the curse of dimensionality, a clustering method is used to merge the task performance evaluation indexes to form core index classes, thereby reducing the dimensionality of the performance evaluation index system. Using the ship engineering decomposition structure theory and evidence reasoning method, a decomposition method for the coupling indexes of the collaborative task is designed to realize high-fidelity decoupling of the collaborative task performance indexes.

[0068] S3.3, based on a neural network, a mapping relationship between the task performance results and the core indexes is constructed, and the weights of each index are designed.

[0069] Using a neural network method, a mapping model of the collaborative task performance results and the task overall performance layer indexes such as task completion performance, interoperability, resilience, autonomy, reliability, and task execution cost is established. Based on Bayesian theory and virtual-real combined data, an understandable overall performance layer index mapping relationship is established. Using the recursive prediction error method and virtual-real closed-loop optimization strategy, the weights of each performance index of the local performance layer and the platform performance layer are designed to realize efficient and reliable evaluation of the heterogeneous unmanned platform collaborative task.

[0070] The overall process and structure of step S3 are as shown in Figure 4 .

[0071] As described above, the method is accurate in analyzing new mechanisms such as autonomous decomposition of collaborative tasks, high-resilience online task allocation and efficient task state description in time and space dimensions, and builds a complete, accurate and conflict-free collaborative task description model in the modeling of a heterogeneous multi-domain unmanned platform system oriented to collaborative tasks. According to different dimensions of the efficiency measurement of the heterogeneous unmanned platform collaborative task, novel and efficient modeling methods such as Monte Carlo, fuzzy rules and statistical reasoning are used to realize comprehensive and accurate modeling of the efficiency measurement of the collaborative task in terms of task completion efficiency, interoperability, resilience, autonomy and reliability. A high-consistency analytic hierarchy process is used to build an evaluation index system with three hierarchical frameworks, i.e., a task overall efficiency layer, a local efficiency layer and a platform efficiency layer. Under the driving of the collaborative task, innovative methods such as artificial intelligence and human-computer interaction are used to deeply analyze key factors affecting various capability models from multiple levels such as platform capability, intelligent behavior capability and network information integration capability, and build high-fidelity unmanned platform capability simulation models, intelligent behavior simulation models and network information integration simulation models that adapt to complex environments.

[0072] A collaborative task efficiency evaluation system comprises:

[0073] A task description and evaluation system module is configured to describe and analyze the state of the task and build an evaluation index system.

[0074] A simulation module is configured to build a simulation model according to the task.

[0075] An efficiency evaluation module is configured to perform index screening and weight distribution based on the evaluation index system and map to the simulation model.

[0076] The content in the above method embodiments is applicable to the system embodiments, the system embodiments specifically realize the same functions as the above method embodiments, and achieve the same beneficial effects as the above method embodiments.

[0077] A collaborative task efficiency evaluation device comprises:

[0078] At least one processor;

[0079] At least one memory configured to store at least one program;

[0080] When the at least one program is executed by the at least one processor, the at least one processor implements the above collaborative task efficiency evaluation method.

[0081] The content in the above method embodiments is applicable to the device embodiments, the device embodiments specifically realize the same functions as the above method embodiments, and achieve the same beneficial effects as the above method embodiments.

[0082] A storage medium, wherein processor executable instructions are stored, the processor executable instructions, when executed by a processor, are used to implement a collaborative task performance evaluation method as described above.

[0083] The contents in the method embodiments are applicable to the storage medium embodiments, the storage medium embodiments specifically implement the same functions as the method embodiments, and achieve the same beneficial effects as the method embodiments.

[0084] The above is a specific description of the preferred embodiments of the present application, but the present application is not limited to the described embodiments, and those skilled in the art can make various equivalent modifications or replacements without departing from the spirit of the present application, and these equivalent modifications or replacements are all included in the scope defined by the claims of the present application.

Claims

1. A method for evaluating the effectiveness of collaborative tasks, characterized in that, Includes the following steps: Describe and analyze the status of the task, and construct an evaluation index system; Build a simulation model based on the task; Based on the evaluation index system, indexes are selected and weights are assigned, and then mapped to the simulation model.

2. The collaborative task performance evaluation method according to claim 1, characterized in that, The step of describing and analyzing the status of the task and constructing an evaluation index system specifically includes: Based on the goals and characteristics of collaborative tasks, the collaborative tasks are decomposed and an allocation model is constructed to obtain a collaborative task description model. Based on the collaborative task description model, qualitative and quantitative analyses are performed to construct a task performance measurement model. The Analytic Hierarchy Process (AHP) was used to study the core influencing factors of collaborative task effectiveness, and an evaluation index system based on the overall task effectiveness layer, local effectiveness layer, and platform effectiveness layer was constructed.

3. The collaborative task performance evaluation method according to claim 2, characterized in that, The collaborative task description model specifically employs a finite state machine to design conflict-free task states for perception, communication, and decision-making in the spatiotemporal dimension. The task effectiveness includes task-related effects, duration effects, and satisfaction effects.

4. The collaborative task performance evaluation method according to claim 2, characterized in that: The overall task effectiveness layer employs an analytical approach that combines the overall functionality and characteristic attributes of the task to construct task evaluation indicators. For the local performance layer, factor analysis is used to extract common influencing factors and construct local evaluation indicators. The platform performance layer uses data statistical analysis to analyze and classify the platform's capabilities, and constructs platform performance evaluation indicators.

5. The collaborative task performance evaluation method according to claim 1, characterized in that, The step of constructing a simulation model based on the task specifically includes: Based on a pre-defined engine, and combined with data analysis and machine learning, a platform simulation model is constructed. Based on adaptive robust estimation and control methods, combined with artificial intelligence, an intelligent behavior simulation model is constructed. An information network simulation model is constructed using a spatially distributed self-organizing and self-organizing network method.

6. The collaborative task performance evaluation method according to claim 5, characterized in that, The platform simulation model includes a platform situational awareness simulation model, a platform motion capability simulation model, a platform communication capability simulation model, and a platform survivability simulation model; the intelligent behavior simulation model includes a self-learning capability simulation model, an intelligent interaction capability simulation model, and a dynamic collaboration capability simulation model; the information network simulation model includes an information networking simulation model, a communication mapping simulation model, and a distributed information integration simulation model.

7. The collaborative task performance evaluation method according to claim 1, characterized in that, The step of selecting and assigning weights to indicators based on the evaluation indicator system and mapping them to the simulation model specifically includes: Taking into account expert experience, an indicator selection strategy was constructed based on a deep learning model; Clustering methods are used to merge the indicators to form core indicators; The mapping relationship between task performance results and the core indicators is constructed based on neural networks, and the weights of each indicator are designed.

8. A collaborative task performance evaluation system, characterized in that, include: The task description and evaluation system module is used to describe and analyze the status of tasks and build an evaluation index system. The simulation module is used to build simulation models based on the task. The performance evaluation module performs indicator selection and weight allocation based on the evaluation indicator system, and maps them to the simulation model.

9. A collaborative task performance evaluation device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a collaborative task performance evaluation method as described in any one of claims 1-7.

Images