A multi-agent cooperative control system based on dynamic event triggering

Abstract

The application discloses a multi-agent cooperative control system based on dynamic event triggering, comprising a collection unit, a cooperative control unit, a dynamic event triggering unit, a communication unit and an execution unit. The collection unit is used for sampling a consistent error multi-agent system. The cooperative control unit is used for data processing of the sampling result and issuing a cooperative control instruction according to the data processing result. The dynamic event triggering unit is used for dynamic event triggering judgment according to the cooperative control instruction and issuing a triggering instruction. The communication unit is used for transmitting the data processing result and the triggering instruction. The execution unit is used for executing the data processing result according to the triggering instruction and completing multi-agent cooperative control. The application can greatly reduce the probability of small probability events that cannot be coped with when the multi-agent system jointly acts, realize the communication and contact of data and information among the agent systems, can mutually refer to the decision, and greatly improve the performance of multi-agent linkage.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of agent control, in particular to a multi-agent cooperative control system based on dynamic event triggering. BACKGROUND

[0002] The automatic and intelligent control system can enhance the management ability of human beings, is widely used in various fields of production and life, and at present, the agent is an important concept in the field of artificial intelligence, any independent and interactive entity with the environment can be abstracted as an agent, that is, a computing entity residing in a certain environment, continuously and autonomously playing a role, having the characteristics of residence, reactivity, sociality and initiative.

[0003] The traditional computing system is closed, and needs to meet the consistency requirement, however, the social mechanism is open and cannot meet the consistency condition, and under this mechanism, part of the individuals need to reach an acceptable solution through a certain negotiation mechanism in the contradictory situation. Minsky calls such individuals in the computing society as agents. The organic combination of these individuals constitutes the computing society, that is, the multi-agent system.

[0004] The limited theory of Simon is another important theoretical basis for the formation of the multi-agent system, Simon believes that a large structure can make up for the limited working capacity of individuals by organizing many individuals; each individual is responsible for a specific task, which can make up for the limited ability of individuals to learn new tasks; the organized flow of information between social institutions can make up for the limited knowledge of individuals; precise social institutions and clear individual tasks can make up for the limited ability of individuals to process and apply information.

[0005] Therefore, although the agent system can perform its duties and provide rich security guarantees, there are still many problems, an agent system is often developed for a certain high-probability problem, which has economic benefits, but under the condition of the joint application of the multi-agent system, there may be countless small probability events that the system cannot cope with, at this time, the existing security system cannot independently judge the situation and take further measures, there is no data and information communication between the systems, and the control decisions cannot be mutually referred. SUMMARY

[0006] In view of the above problems, the present application provides a multi-agent cooperative control system based on dynamic event triggering to solve the technical problems existing in the prior art, which can greatly reduce the probability of small probability events that cannot be coped with when the multi-agent system works together, realize the communication of data and information between the agent systems, and can mutually refer to the decisions, greatly improving the performance of the multi-agent linkage.

[0007] In order to achieve the above object, the application provides a multi-agent cooperative control system based on dynamic event triggering, comprising a collection unit, a cooperative control unit, a dynamic event triggering unit, a communication unit and an execution unit.

[0008] The collection unit is used for sampling the multi-agent system with consistent error to obtain a sampling result.

[0009] The cooperative control unit is used for data processing of the sampling result and issuing a cooperative control instruction according to the data processing result.

[0010] The dynamic event triggering unit is used for dynamic event triggering judgment according to the cooperative control instruction and issuing a triggering instruction.

[0011] The communication unit is used for transmitting the data processing result and the triggering instruction.

[0012] The execution unit is used for executing the data processing result according to the triggering instruction to complete the multi-agent cooperative control.

[0013] Preferably, the cooperative control unit comprises a data processing module, a data storage module and a cooperative control module.

[0014] The data processing module is used for digital-analog conversion of the sampling result to obtain a conversion result.

[0015] The data storage module is used for storing the conversion result.

[0016] The cooperative control module is used for multi-agent state coordination according to the stored conversion result and issuing a cooperative control instruction based on the coordination result.

[0017] Preferably, the multi-agent state coordination comprises formulating a decision according to the stored conversion result and transmitting a cooperative control instruction to a subordinate unit according to the decision result.

[0018] Preferably, the dynamic event triggering unit comprises a condition constraint module and a triggering module.

[0019] The condition constraint module is used for setting a dynamic event triggering condition and performing dynamic event triggering judgment on the conversion result according to the cooperative control instruction.

[0020] The triggering module is used for issuing a triggering instruction according to the judgment result.

[0021] Preferably, the communication unit comprises a first communication module and a second communication module.

[0022] The first communication module is used for transmitting the data processing result.

[0023] The second communication module is used to transmit the trigger command.

[0024] Preferably, the first communication module and the second communication module communicate wirelessly.

[0025] Preferably, the execution unit includes a smart terminal and an execution module;

[0026] The intelligent terminal includes several intelligent systems; each intelligent system includes several intelligent agents;

[0027] The execution module is used to execute a differential game guidance law on the data processing result according to the trigger instruction, so as to complete multi-agent cooperative control.

[0028] Preferably, the types of intelligent systems include status monitoring systems, human-computer interaction systems, recorders, radar systems, remote communication systems, artificial intelligence systems, machine training systems, databases, and wireless search systems.

[0029] The present invention discloses the following technical effects:

[0030] This invention can significantly reduce the probability of low-probability events that cannot be handled when multiple intelligent agent systems work together. This system can autonomously make judgments on the situation and take further measures, realize data and information communication between various intelligent agent systems, realize mutual reference in making control decisions, and greatly improve the performance of multi-agent linkage. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram of the system of the present invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0035] Referring to Figure 1 The embodiment shown provides a multi-agent cooperative control system based on dynamic event triggering, comprising a collection unit, a cooperative control unit, a dynamic event triggering unit, a communication unit and an execution unit.

[0036] The collection unit is used for sampling the multi-agent system with consistent error; the cooperative control unit is used for data processing on the sampling results and issuing cooperative control instructions according to the data processing results; the dynamic event triggering unit is used for dynamic event triggering judgment according to the cooperative control instructions and issuing triggering instructions; the communication unit is used for transmitting the data processing results and the triggering instructions; and the execution unit is used for executing the data processing results according to the triggering instructions to complete the multi-agent cooperative control.

[0037] The cooperative control unit comprises a data processing module, a data storage module and a cooperative control module; the data processing module is used for analog-digital conversion on the sampling results to obtain conversion results; the data storage module is used for storing the conversion results; and the cooperative control module is used for multi-agent state coordination according to the stored conversion results and issuing cooperative control instructions based on the coordination results. The multi-agent state coordination comprises formulating decisions according to the stored conversion results and transmitting cooperative control instructions to lower-level units according to the decision results.

[0038] The decision formulation specifically comprises:

[0039] The cooperative control module is also connected with a plurality of subordinate intelligent agent systems to form a superior system to control a plurality of subordinate systems. The subordinate systems are the subsystems of the whole system or external systems. The subsystems refer to the internal components of the system, which can be deeply integrated with the superior system or other subsystems in terms of software and hardware resources. The external systems refer to the external systems that have the same status as the subsystems, accept the control of the central system or assist, but do not belong to the system. In the embodiment, the superior system can be provided with a central memory as an external memory to save the data of the monitored object, including the running data of the device, the running data of the subordinate system, the data and information content transmitted between the network nodes or the information processed in the later stage; the central controller includes a CPU, a memory, a communication module and other basic components, can make decisions according to the data transmitted by the acquisition unit, and directly transmit control commands to the execution system of the subordinate system; or transmit decision reference data to the controller of the subordinate system, which can also be completed by the subordinate system responsible for providing decision reference data, so that the controller of the subordinate system can correct the calculation results when making decisions; the sharing mode of the central controller or the central memory between multiple systems can be that a physical space is divided into multiple smaller units to allocate different subordinate systems, or multiple unit modules are combined together but serve different subordinate systems, or the same physical space or unit module is dynamically allocated at different times, and different modes can be mixed.

[0040] The dynamic event triggering unit includes a condition constraint module and a triggering module; the condition constraint module is used for setting a dynamic event triggering condition and making a dynamic event triggering judgment on the conversion result according to the cooperative control instruction; and the triggering module is used for issuing a triggering instruction according to the judgment result.

[0041] The dynamic triggering mechanism specifically includes:

[0042] The state of the consistent error system is sampled every interval random seconds to obtain the state after periodic sampling, and an event triggering condition is defined. When the difference between the current periodic sampling state of the system and the last triggering state violates the defined triggering condition, a new triggering time is generated, and the feedback guidance law is updated. By using the universal approximation characteristics of the neural network, a cost function and its terminal constraint for evaluating the approximate optimal network are designed, and then an approximate event-triggered differential game cooperative guidance law is obtained. According to the gradient descent method, an updating law is obtained to complete the updating step of the guidance law; and an approximate optimal control strategy is obtained based on the system event-triggered constraint condition.

[0043] The communication unit includes a first communication module and a second communication module; the first communication module is used for transmitting data processing results; and the second communication module is used for transmitting triggering instructions. The first communication module and the second communication module communicate in a wireless connection mode.

[0044] The execution unit comprises an intelligent terminal and an execution module; the intelligent terminal comprises several intelligent systems; each intelligent system comprises several intelligent agents; the execution module is used for executing differential countermeasure guidance law on data processing results according to trigger instructions, and completing multi-agent collaborative control. The types of intelligent systems include a state monitoring system, a man-machine interaction system, a recorder, a radar system, a remote communication system, an artificial intelligence system, a machine training system, a database and a wireless search system.

[0045] The man-machine interaction system in the embodiment comprises a switch, a keyboard, an electronic screen, a voice system, a brain-computer system or a combination thereof, and can be used for managing and using the system, or can be used for acquiring data of life characteristics, behavior characteristics or psychological characteristics of a person, or has the above functions; and different users can contact each other through a data center, or the system has a Bluetooth or other type of wireless search system, allowing users to directly perceive and contact each other for on-site assistance.

[0046] In the embodiment, the collaborative control system can also replace the subordinate systems with systems in different professional fields, and is applied to safety control of fixed places such as houses, shopping malls, workshops, construction sites and forests, or a plurality of types of subordinate systems are selected across themes, and the overall system serves as a general multi-system collaborative platform.

[0047] The present application discloses the following technical effects:

[0048] The present application can greatly reduce the probability of small probability events that cannot be coped with when the multi-agent system acts together, the system can independently judge the situation and take further measures, realize the communication and contact between the data and information of the intelligent agent systems, realize mutual reference in formulating control decisions, and greatly improve the performance of multi-agent linkage.

[0049] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present application, which are used to illustrate the technical solutions of the present application, but not to limit the same, and the protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can make modifications or easily think of changes to the technical solutions recorded in the foregoing embodiments within the technical scope disclosed by the present application, or make equivalent replacements to some technical features; and these modifications, changes or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application. All of them should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A multi-agent cooperative control system based on dynamic event triggering, characterized in that, The application relates to a multi-agent system event-triggered cooperative control method and device. The application comprises a collection unit, a cooperative control unit, a dynamic event triggering unit, a communication unit and an execution unit. The collection unit is used for sampling a consistent error multi-agent system to obtain a sampling result. The cooperative control unit is used for data processing of the sampling result and issuing a cooperative control instruction according to the data processing result. The dynamic event triggering unit is used for dynamic event triggering judgment according to the cooperative control instruction and issuing a triggering instruction. The communication unit is used for transmitting the data processing result and the triggering instruction. The execution unit is used for executing the data processing result according to the triggering instruction to complete multi-agent cooperative control. The cooperative control unit comprises a data processing module, a data storage module and a cooperative control module. The data processing module is used for digital-analog conversion of the sampling result to obtain a conversion result. The data storage module is used for storing the conversion result. The cooperative control module is used for multi-agent state coordination according to the stored conversion result and issuing a cooperative control instruction based on the coordination result. The cooperative control module is further connected with a plurality of subordinate agent systems to form a superior system to control the plurality of subordinate systems; the subordinate system is a subsystem of the overall system or an external system; the subsystem refers to an internal component of the system which can be deeply integrated with the superior system or other subsystems in software and hardware resources; the external system refers to an external system which has the same status as the subsystem, accepts the control of the central system or assists but does not belong to the system; The multi-agent state coordination comprises formulating a decision according to the stored conversion result and transmitting a cooperative control instruction to a subordinate unit according to the decision result. The dynamic event triggering unit comprises a condition constraint module and a triggering module. The condition constraint module is used for setting a dynamic event triggering condition and performing dynamic event triggering judgment on the conversion result according to the cooperative control instruction. The triggering module is used for issuing a triggering instruction according to the judgment result. The dynamic triggering mechanism comprises the following steps: The state of the consistent error system is sampled every interval random seconds to obtain a periodically sampled state; an event triggering condition is defined; when the difference between the current periodically sampled state and the last triggering state violates the defined triggering condition, a new triggering time is generated; a cost function and a terminal constraint which approximate the optimal cost function and terminal constraint of a neural network are designed by using the universal approximation characteristic of the neural network; an approximate event-triggered differential game cooperative guidance law is obtained; an updating law is obtained according to the gradient descent method to complete the updating step of the guidance law; and an approximate optimal control strategy is obtained based on the system event-triggered constraint condition. The communication unit comprises a first communication module and a second communication module. The first communication module is used for transmitting the data processing result. The second communication module is used for transmitting the triggering instruction. The first communication module and the second communication module adopt a wireless connection mode for communication. The execution unit comprises an intelligent terminal and an execution module. The intelligent terminal comprises a plurality of intelligent systems; each intelligent system comprises a plurality of agents. The execution module is configured to execute a differential countermeasure guidance law on the data processing result according to the trigger instruction, and complete multi-agent cooperative control. The types of the intelligent system include a state monitoring system, a human-computer interaction system, a recorder, a radar system, a remote communication system, an artificial intelligence system, a machine training system, a database, a wireless search system; wherein the human-computer interaction system includes a switch, a keyboard, an electronic screen, a voice system, a brain-computer system or a combination thereof, and can be used for managing and using the system, or can be used for acquiring data of life characteristics, behavior characteristics or psychological characteristics of a person, or has the above functions; and different users can assist each other in communication through a data center, or the system is provided with a Bluetooth or other type of wireless search system, allowing users to directly perceive and communicate with each other, so as to provide on-site assistance.

Images