A task allocation model construction method and system for heterogeneous intelligent agents

Abstract

The application provides a task allocation model construction method and system for a heterogeneous intelligent agent, and the method comprises the following steps: determining a target heterogeneous intelligent agent in real time, and matching a target performance parameter corresponding to the target heterogeneous intelligent agent in a preset database in real time; creating a corresponding professional term vocabulary according to the target performance parameter based on a preset rule in real time, and converting the professional term vocabulary into a plurality of distributed vectors corresponding thereto through a preset continuous bag-of-words model in real time; and training a task allocation model for the target heterogeneous intelligent agent according to a preset bidirectional long short-term memory network and the plurality of distributed vectors in real time, wherein each distributed vector has uniqueness. The task allocation model can objectively and accurately complete the task allocation of the heterogeneous intelligent agent, and the work efficiency is greatly improved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method and system for constructing a task allocation model for heterogeneous intelligent agents. Background Technology

[0002] With the advancement of technology and the rapid development of productivity, people have made significant progress in the field of artificial intelligence technology, and have developed intelligent agents such as drones and industrial robots, which have improved work efficiency.

[0003] In the existing field of intelligent agent technology, due to the large differences in the types of performance parameters among existing heterogeneous intelligent agents and the large number of performance parameters of a single intelligent agent, existing heterogeneous intelligent agents need to undergo targeted task allocation processing before executing tasks.

[0004] Furthermore, in the process of assigning tasks to heterogeneous intelligent agents, existing technologies mostly assign tasks based on the model of the heterogeneous intelligent agent. However, this method cannot accurately identify the specific performance parameters of the heterogeneous intelligent agent, which can easily lead to the mismatch between the assigned tasks and the actual heterogeneous intelligent agent, thereby reducing work efficiency. Summary of the Invention

[0005] Based on this, the purpose of this invention is to provide a method and system for constructing a task allocation model for heterogeneous intelligent agents, so as to solve the problem that in the process of task allocation in the prior art, the allocated tasks are easily mismatched with the actual heterogeneous intelligent agents, which reduces work efficiency.

[0006] The first aspect of the present invention proposes:

[0007] A method for constructing a task allocation model for heterogeneous intelligent agents, wherein the method includes:

[0008] The target heterogeneous intelligent agent is identified in real time, and the target performance parameters corresponding to the target heterogeneous intelligent agent are matched in real time in a preset database;

[0009] Based on preset rules, a corresponding terminology lexicon is created in real time according to the target performance parameters, and the terminology lexicon is converted into several corresponding distributed vectors in real time through a preset continuous bag-of-words model.

[0010] A task allocation model for the target heterogeneous intelligent agent is trained in real time based on a preset bidirectional short-time memory network and several distributed vectors, and each distributed vector is unique.

[0011] The beneficial effects of this invention are as follows: by identifying the required heterogeneous intelligent agents in real time, the corresponding processing objects can be clearly defined. Based on this, the corresponding target performance parameters can be matched again. Specifically, in order to facilitate the subsequent model construction, a corresponding professional terminology library can be created in real time according to the pre-set rules and the current target performance parameters, and a corresponding distributed vector can be generated synchronously. Based on this, the final model training can be performed, and the required task allocation model can be obtained. Thus, the task allocation model can be used to objectively and accurately complete the task allocation of each heterogeneous intelligent agent, avoiding the phenomenon of mismatch and greatly improving work efficiency.

[0012] Furthermore, the step of creating a corresponding terminology database in real time based on preset rules and the target performance parameters includes:

[0013] When the target performance parameters are acquired in real time, a full scan of the target performance parameters is performed to detect the types of parameters contained in the target performance parameters in real time.

[0014] For each parameter type, a corresponding target identifier is added, and a subset of parameters corresponding to each parameter type is matched from the target performance parameters based on each target identifier.

[0015] A corresponding terminology database is created in real time based on each parameter type and its corresponding parameter subset.

[0016] Furthermore, the step of creating a corresponding terminology lexicon in real time based on each parameter type and its corresponding parameter subset includes:

[0017] When each parameter type and its corresponding parameter subset are obtained in real time, several parameter types that are compatible with the parameter type are extracted in real time from the parameter subset.

[0018] The system detects the parameter keywords corresponding to each parameter type in real time and creates the professional terminology database based on the parameter type and the corresponding parameter keywords.

[0019] Furthermore, the step of creating the terminology database based on the parameter type and the corresponding parameter keywords includes:

[0020] When each parameter type and its corresponding parameter keywords are detected in real time, each parameter type is integrated to create the corresponding parameter string in real time.

[0021] The mapping relationship between each parameter type and its corresponding parameter keywords is constructed in real time, and each parameter keyword is mapped to the parameter string according to the mapping relationship to generate the corresponding parameter tree diagram in real time.

[0022] The parameter tree diagram is converted into a corresponding professional terminology lookup table in real time, and the professional terminology database is created in real time based on the professional terminology lookup table.

[0023] Furthermore, the step of training a task allocation model for the heterogeneous agent in real time based on a preset bidirectional short-time memory network and several of the distributed vectors includes:

[0024] When several distributed vectors are acquired in real time, a corresponding distributed array is created in real time based on the several distributed vectors.

[0025] The distributed array is input into a preset embedding layer so that the preset embedding layer outputs the corresponding embedding vector array in real time.

[0026] The task allocation model of the heterogeneous agent is trained based on the embedded vector array and the preset bidirectional short-time memory network.

[0027] Furthermore, the step of training the task allocation model of the heterogeneous agent based on the embedded vector array and the preset bidirectional short-time memory network includes:

[0028] When the embedding vector array is acquired in real time, the preset bidirectional short-time memory network is trained iteratively through the embedding vector array to output the corresponding initial allocation model in real time.

[0029] The initial allocation model is sequentially input into a preset aggregation layer and a preset fully connected layer, and then subjected to a second iteration of training to generate the corresponding task allocation model.

[0030] Furthermore, the step of sequentially inputting the initial allocation model into a preset aggregation layer and a preset fully connected layer, and performing a second iteration of training to generate the corresponding task allocation model includes:

[0031] When the initial allocation model is obtained in real time, the initial allocation model is aggregated through the preset aggregation layer, and the initial allocation model is globally pooled through the preset fully connected layer.

[0032] The task allocation model is output through a preset output layer.

[0033] The second aspect of the present invention proposes:

[0034] A system for constructing a task allocation model for heterogeneous intelligent agents, wherein the system comprises:

[0035] The matching module is used to identify the target heterogeneous intelligent agent in real time and match the target performance parameters corresponding to the target heterogeneous intelligent agent in a preset database in real time.

[0036] A creation module is used to create a corresponding terminology lexicon in real time based on preset rules and the target performance parameters, and to convert the terminology lexicon into several corresponding distributed vectors in real time through a preset continuous bag-of-words model.

[0037] The training module is used to train a task allocation model for the target heterogeneous intelligent agent in real time based on a preset bidirectional short-time memory network and several distributed vectors, wherein each distributed vector is unique.

[0038] Furthermore, the creation module is specifically used for:

[0039] When the target performance parameters are acquired in real time, a full scan of the target performance parameters is performed to detect the types of parameters contained in the target performance parameters in real time.

[0040] For each parameter type, a corresponding target identifier is added, and a subset of parameters corresponding to each parameter type is matched from the target performance parameters based on each target identifier.

[0041] A corresponding terminology database is created in real time based on each parameter type and its corresponding parameter subset.

[0042] Furthermore, the creation module is specifically used for:

[0043] When each parameter type and its corresponding parameter subset are obtained in real time, several parameter types that are compatible with the parameter type are extracted in real time from the parameter subset.

[0044] The system detects the parameter keywords corresponding to each parameter type in real time and creates the professional terminology database based on the parameter type and the corresponding parameter keywords.

[0045] Furthermore, the creation module is specifically used for:

[0046] When each parameter type and its corresponding parameter keywords are detected in real time, each parameter type is integrated to create the corresponding parameter string in real time.

[0047] The mapping relationship between each parameter type and its corresponding parameter keywords is constructed in real time, and each parameter keyword is mapped to the parameter string according to the mapping relationship to generate the corresponding parameter tree diagram in real time.

[0048] The parameter tree diagram is converted into a corresponding professional terminology lookup table in real time, and the professional terminology database is created in real time based on the professional terminology lookup table.

[0049] Furthermore, the training module is specifically used for:

[0050] When several distributed vectors are acquired in real time, a corresponding distributed array is created in real time based on the several distributed vectors.

[0051] The distributed array is input into a preset embedding layer so that the preset embedding layer outputs the corresponding embedding vector array in real time.

[0052] The task allocation model of the heterogeneous agent is trained based on the embedded vector array and the preset bidirectional short-time memory network.

[0053] Furthermore, the training module is specifically used for:

[0054] When the embedding vector array is acquired in real time, the preset bidirectional short-time memory network is trained iteratively through the embedding vector array to output the corresponding initial allocation model in real time.

[0055] The initial allocation model is sequentially input into a preset aggregation layer and a preset fully connected layer, and then subjected to a second iteration of training to generate the corresponding task allocation model.

[0056] Furthermore, the training module is specifically used for:

[0057] When the initial allocation model is obtained in real time, the initial allocation model is aggregated through the preset aggregation layer, and the initial allocation model is globally pooled through the preset fully connected layer.

[0058] The task allocation model is output through a preset output layer.

[0059] The third aspect of the present invention proposes:

[0060] A computer includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the task allocation model construction method for heterogeneous intelligent agents as described above.

[0061] The fourth aspect of the present invention proposes:

[0062] A readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the task allocation model construction method for heterogeneous intelligent agents as described above.

[0063] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0064] Figure 1 A flowchart of a task allocation model construction method for heterogeneous intelligent agents provided in the first embodiment of the present invention;

[0065] Figure 2 The structural block diagram of the task allocation model construction system for heterogeneous intelligent agents provided in the third embodiment of the present invention is shown.

[0066] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0067] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0068] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0069] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0070] Please see Figure 1The figure shows a task allocation model construction method for heterogeneous intelligent agents provided in the first embodiment of the present invention. The task allocation model construction method for heterogeneous intelligent agents provided in this embodiment can objectively and accurately construct the required task allocation model, and can allocate suitable tasks to each heterogeneous intelligent agent through the task allocation model, thereby greatly improving work efficiency.

[0071] Specifically, this embodiment provides:

[0072] A method for constructing a task allocation model for heterogeneous intelligent agents includes the following steps:

[0073] Step S10: Identify the target heterogeneous intelligent agent in real time and match the target performance parameters corresponding to the target heterogeneous intelligent agent in a preset database in real time;

[0074] Step S20: Based on preset rules, a corresponding terminology lexicon is created in real time according to the target performance parameters, and the terminology lexicon is converted into several corresponding distributed vectors in real time through a preset continuous bag-of-words model.

[0075] Step S30: Based on a preset bidirectional short-time memory network and several distributed vectors, a task allocation model for the target heterogeneous intelligent agent is trained in real time, and each distributed vector is unique.

[0076] Specifically, in this embodiment, it should first be noted that in order to objectively and effectively train the required task allocation model, accurate training data needs to be obtained in real time. It should be noted that the performance parameters of existing heterogeneous intelligent agents directly determine the target tasks they can handle; that is, the tasks that existing heterogeneous intelligent agents can handle are related to their performance parameters. Based on this, in order to objectively and accurately complete task allocation, this invention first needs to identify the corresponding processing object in practical applications. Specifically, this invention will first determine a target heterogeneous intelligent agent for processing. This target heterogeneous intelligent agent can be equipment such as drones and industrial robots. Based on this, this invention can first match the target performance parameters corresponding to the current target heterogeneous intelligent agent in the existing database, that is, the data processing capability of the current target heterogeneous intelligent agent, to facilitate subsequent processing.

[0077] Furthermore, after obtaining the required target performance parameters in real time through the above steps, the present invention can, based on the current target performance parameters, immediately generate a corresponding terminology lexicon according to pre-set rules. Based on this, in order to obtain data for subsequent training, the current terminology lexicon is further transformed in real time using the existing continuous bag-of-words model, and several distributed vectors are generated again. Specifically, these distributed vectors can be directly used as subsequent training data. Thus, the present invention can ultimately train a task allocation model for the aforementioned heterogeneous agents in real time based on a pre-set bidirectional short-term memory network and the current distributed vectors. In practical applications, this task allocation model can automatically complete the task allocation for each heterogeneous agent, thereby avoiding the phenomenon of mismatch between the allocated tasks and the processing capabilities of the heterogeneous agents, and significantly improving work efficiency.

[0078] Second Embodiment

[0079] Furthermore, the step of creating a corresponding terminology database in real time based on preset rules and the target performance parameters includes:

[0080] When the target performance parameters are acquired in real time, a full scan of the target performance parameters is performed to detect the types of parameters contained in the target performance parameters in real time.

[0081] For each parameter type, a corresponding target identifier is added, and a subset of parameters corresponding to each parameter type is matched from the target performance parameters based on each target identifier.

[0082] A corresponding terminology database is created in real time based on each parameter type and its corresponding parameter subset.

[0083] Furthermore, the step of creating a corresponding terminology lexicon in real time based on each parameter type and its corresponding parameter subset includes:

[0084] When each parameter type and its corresponding parameter subset are obtained in real time, several parameter types that are compatible with the parameter type are extracted in real time from the parameter subset.

[0085] The system detects the parameter keywords corresponding to each parameter type in real time and creates the professional terminology database based on the parameter type and the corresponding parameter keywords.

[0086] Furthermore, the step of creating the terminology database based on the parameter type and the corresponding parameter keywords includes:

[0087] When each parameter type and its corresponding parameter keywords are detected in real time, each parameter type is integrated to create the corresponding parameter string in real time.

[0088] The mapping relationship between each parameter type and its corresponding parameter keywords is constructed in real time, and each parameter keyword is mapped to the parameter string according to the mapping relationship to generate the corresponding parameter tree diagram in real time.

[0089] The parameter tree diagram is converted into a corresponding professional terminology lookup table in real time, and the professional terminology database is created in real time based on the professional terminology lookup table.

[0090] Furthermore, the step of training a task allocation model for the heterogeneous agent in real time based on a preset bidirectional short-time memory network and several of the distributed vectors includes:

[0091] When several distributed vectors are acquired in real time, a corresponding distributed array is created in real time based on the several distributed vectors.

[0092] The distributed array is input into a preset embedding layer so that the preset embedding layer outputs the corresponding embedding vector array in real time.

[0093] The task allocation model of the heterogeneous agent is trained based on the embedded vector array and the preset bidirectional short-time memory network.

[0094] Furthermore, the step of training the task allocation model of the heterogeneous agent based on the embedded vector array and the preset bidirectional short-time memory network includes:

[0095] When the embedding vector array is acquired in real time, the preset bidirectional short-time memory network is trained iteratively through the embedding vector array to output the corresponding initial allocation model in real time.

[0096] The initial allocation model is sequentially input into a preset aggregation layer and a preset fully connected layer, and then subjected to a second iteration of training to generate the corresponding task allocation model.

[0097] Furthermore, the step of sequentially inputting the initial allocation model into a preset aggregation layer and a preset fully connected layer, and performing a second iteration of training to generate the corresponding task allocation model includes:

[0098] When the initial allocation model is obtained in real time, the initial allocation model is aggregated through the preset aggregation layer, and the initial allocation model is globally pooled through the preset fully connected layer.

[0099] The task allocation model is output through a preset output layer.

[0100] Furthermore, in this embodiment, it should be noted that after obtaining the required target performance parameters in real time through the above steps, it is necessary to parse the current target performance parameters in real time to generate a professional terminology database for subsequent training efficiency improvement. Specifically, in order to comprehensively obtain all the information contained within the current target performance parameters, this invention will immediately perform a full scan of the current target performance parameters. At the same time, it can detect several parameter types contained within the current target performance parameters in real time. Specifically, for ease of understanding, these parameter types can be mechanical parameters and software parameters, etc. Based on this, it can match the parameter subset corresponding to each parameter type in real time within the target performance parameters. Specifically, for ease of implementation, this invention will parse the current parameter subset again and extract several parameter types that are compatible with the current parameter type from the parameter subset. For ease of understanding, these parameter types can be mechanical parameters and dynamic parameters, etc. Based on this, this invention will again use the current parameter types as the processing object. Furthermore, it can detect the parameter keywords corresponding to each parameter type in real time. These parameter keywords represent the information contained in the current parameter type. Based on this, to facilitate the integration of various data, this invention first integrates each parameter type and creates a corresponding parameter string in real time based on the character information contained within each parameter type. Simultaneously, to facilitate subsequent integration, this invention also constructs a mapping relationship between each parameter type and its corresponding parameter keywords in real time. Based on this, this invention directly maps each parameter keyword to the aforementioned parameter string according to the mapping relationship, thereby forming a corresponding parameter tree diagram in real time. Specifically, this parameter tree diagram clearly reflects the mapping relationship between parameter types and various parameter categories, thus clearly obtaining the distribution of professional terms corresponding to each parameter type. Subsequently, the current parameter tree diagram can be converted into a corresponding professional term reference table in real time, and a corresponding professional term database can be created in real time based on this professional term reference table for subsequent processing.

[0101] Furthermore, after obtaining the required distributed vectors in real time through the above steps, these distributed vectors can be arrayed to form a corresponding distributed array. Specifically, this invention inputs the current distributed array into a pre-set embedding layer in real time, and the pre-set embedding layer can convert the current distributed array into an embedding vector array for subsequent training in real time. Based on this, in order to quickly and effectively complete the subsequent model training, this invention immediately performs an iterative training on the current pre-set bidirectional short-term memory network using the current embedding vector array, and can train the required initial allocation model accordingly. The current initial allocation model is then input into the pre-set aggregation layer and fully connected layer in sequence. During this process, the current aggregation layer performs aggregation processing on the current initial allocation model, and the current fully connected layer performs global pooling processing on the current initial allocation model for a corresponding second iterative training, thereby finally training the required task allocation model and completing task allocation through this task allocation model, thus improving work efficiency.

[0102] Please see Figure 2 The third embodiment of the present invention provides:

[0103] A system for constructing a task allocation model for heterogeneous intelligent agents, wherein the system comprises:

[0104] The matching module is used to identify the target heterogeneous intelligent agent in real time and match the target performance parameters corresponding to the target heterogeneous intelligent agent in a preset database in real time.

[0105] A creation module is used to create a corresponding terminology lexicon in real time based on preset rules and the target performance parameters, and to convert the terminology lexicon into several corresponding distributed vectors in real time through a preset continuous bag-of-words model.

[0106] The training module is used to train a task allocation model for the target heterogeneous intelligent agent in real time based on a preset bidirectional short-time memory network and several distributed vectors, wherein each distributed vector is unique.

[0107] Furthermore, the creation module is specifically used for:

[0108] When the target performance parameters are acquired in real time, a full scan of the target performance parameters is performed to detect the types of parameters contained in the target performance parameters in real time.

[0109] For each parameter type, a corresponding target identifier is added, and a subset of parameters corresponding to each parameter type is matched from the target performance parameters based on each target identifier.

[0110] A corresponding terminology database is created in real time based on each parameter type and its corresponding parameter subset.

[0111] Furthermore, the creation module is specifically used for:

[0112] When each parameter type and its corresponding parameter subset are obtained in real time, several parameter types that are compatible with the parameter type are extracted in real time from the parameter subset.

[0113] The system detects the parameter keywords corresponding to each parameter type in real time and creates the professional terminology database based on the parameter type and the corresponding parameter keywords.

[0114] Furthermore, the creation module is specifically used for:

[0115] When each parameter type and its corresponding parameter keywords are detected in real time, each parameter type is integrated to create the corresponding parameter string in real time.

[0116] The mapping relationship between each parameter type and its corresponding parameter keywords is constructed in real time, and each parameter keyword is mapped to the parameter string according to the mapping relationship to generate the corresponding parameter tree diagram in real time.

[0117] The parameter tree diagram is converted into a corresponding professional terminology lookup table in real time, and the professional terminology database is created in real time based on the professional terminology lookup table.

[0118] Furthermore, the training module is specifically used for:

[0119] When several distributed vectors are acquired in real time, a corresponding distributed array is created in real time based on the several distributed vectors.

[0120] The distributed array is input into a preset embedding layer so that the preset embedding layer outputs the corresponding embedding vector array in real time.

[0121] The task allocation model of the heterogeneous agent is trained based on the embedded vector array and the preset bidirectional short-time memory network.

[0122] Furthermore, the training module is specifically used for:

[0123] When the embedding vector array is acquired in real time, the preset bidirectional short-time memory network is trained iteratively through the embedding vector array to output the corresponding initial allocation model in real time.

[0124] The initial allocation model is sequentially input into a preset aggregation layer and a preset fully connected layer, and then subjected to a second iteration of training to generate the corresponding task allocation model.

[0125] Furthermore, the training module is specifically used for:

[0126] When the initial allocation model is obtained in real time, the initial allocation model is aggregated through the preset aggregation layer, and the initial allocation model is globally pooled through the preset fully connected layer.

[0127] The task allocation model is output through a preset output layer.

[0128] The fourth embodiment of the present invention provides a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the task allocation model construction method for heterogeneous intelligent agents as described above.

[0129] The fifth embodiment of the present invention provides a readable storage medium storing a computer program thereon, wherein the program, when executed by a processor, implements the task allocation model construction method for heterogeneous intelligent agents as described above.

[0130] In summary, the task allocation model construction method and system for heterogeneous intelligent agents provided in the above embodiments of the present invention can objectively and accurately train the required task allocation model, and can complete the task allocation for each heterogeneous intelligent agent through the task allocation model, thereby improving work efficiency.

[0131] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0132] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0133] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0134] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0135] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0136] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for constructing a task allocation model for heterogeneous intelligent agents, characterized in that, The method includes: The target heterogeneous intelligent agent is identified in real time, and the target performance parameters corresponding to the target heterogeneous intelligent agent are matched in real time in a preset database; Based on preset rules, a corresponding terminology lexicon is created in real time according to the target performance parameters, and the terminology lexicon is converted into several corresponding distributed vectors in real time through a preset continuous bag-of-words model. A task allocation model for the target heterogeneous intelligent agent is trained in real time based on a preset bidirectional short-time memory network and several distributed vectors, and each distributed vector is unique. The step of creating a corresponding terminology database in real time based on preset rules and the target performance parameters includes: When the target performance parameters are acquired in real time, a full scan of the target performance parameters is performed to detect the types of parameters contained in the target performance parameters in real time. For each parameter type, a corresponding target identifier is added, and a subset of parameters corresponding to each parameter type is matched from the target performance parameters based on each target identifier. A corresponding terminology database is created in real time based on each parameter type and its corresponding parameter subset. The step of creating a corresponding terminology database in real time based on each parameter type and its corresponding parameter subset includes: When each parameter type and its corresponding parameter subset are obtained in real time, several parameter types that are compatible with the parameter type are extracted in real time from the parameter subset. The system detects the parameter keywords corresponding to each parameter type in real time and creates the terminology database based on the parameter type and the corresponding parameter keywords. The step of creating the terminology database based on the parameter type and the corresponding parameter keywords includes: When each parameter type and its corresponding parameter keywords are detected in real time, each parameter type is integrated to create the corresponding parameter string in real time. The mapping relationship between each parameter type and its corresponding parameter keywords is constructed in real time, and each parameter keyword is mapped to the parameter string according to the mapping relationship to generate the corresponding parameter tree diagram in real time. The parameter tree diagram is converted into a corresponding professional terminology lookup table in real time, and the professional terminology database is created in real time based on the professional terminology lookup table. The step of training a task allocation model for the heterogeneous agent in real time based on a preset bidirectional short-time memory network and several distributed vectors includes: When several distributed vectors are acquired in real time, a corresponding distributed array is created in real time based on the several distributed vectors. The distributed array is input into a preset embedding layer so that the preset embedding layer outputs the corresponding embedding vector array in real time. The task allocation model of the heterogeneous agent is trained according to the embedded vector array and the preset bidirectional short-time memory network. The step of training the task allocation model of the heterogeneous agent based on the embedded vector array and the preset bidirectional short-time memory network includes: When the embedding vector array is acquired in real time, the preset bidirectional short-time memory network is trained iteratively through the embedding vector array to output the corresponding initial allocation model in real time. The initial allocation model is sequentially input into a preset aggregation layer and a preset fully connected layer, and then subjected to a second iteration of training to generate the corresponding task allocation model. The step of sequentially inputting the initial allocation model into a preset aggregation layer and a preset fully connected layer, and performing a second iteration of training to generate the corresponding task allocation model includes: When the initial allocation model is obtained in real time, the initial allocation model is aggregated through the preset aggregation layer, and the initial allocation model is globally pooled through the preset fully connected layer. The task allocation model is output through a preset output layer.

2. A task allocation model construction system for heterogeneous intelligent agents, characterized in that, The system for implementing the task allocation model construction method for heterogeneous agents as described in claim 1 includes: The matching module is used to identify the target heterogeneous intelligent agent in real time and match the target performance parameters corresponding to the target heterogeneous intelligent agent in a preset database in real time. A creation module is used to create a corresponding terminology lexicon in real time based on preset rules and the target performance parameters, and to convert the terminology lexicon into several corresponding distributed vectors in real time through a preset continuous bag-of-words model. The training module is used to train a task allocation model for the target heterogeneous intelligent agent in real time based on a preset bidirectional short-time memory network and several distributed vectors, wherein each distributed vector is unique.

3. A computer, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the task allocation model construction method for heterogeneous intelligent agents as described in claim 1.

4. A readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the task allocation model construction method for heterogeneous agents as described in claim 1.

Images