Distributed management method based on multi-agent system
By designing follower event-triggered pulse protocols and controlling pulse inputs in a multi-agent system, asynchronous non-periodic control is realized, the problems of waste of resources and insufficient processing capabilities in periodic control are solved, and resource utilization and system synchronization performance are improved.
Patent Information
- Application Number
- CN202411065492.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-05
- Publication Date
- 2024-11-12
AI Technical Summary
Multi-agent systems have resource waste problems in periodic control, and the agile's computing and communication processing capabilities are limited, resulting in low resource utilization efficiency and poor synchronization.
A follower event-triggered pulse protocol is designed to calculate the follower's control pulse input, and a multi-agent system model consisting of leaders and followers is built to realize asynchronous and non-periodic control.
Resource utilization is significantly improved through non-periodic control, system flexibility and robustness are enhanced, and task coordination and system stability are ensured.
Smart Images

Figure CN118945236A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of multi-agent systems, and in particular to a distributed management method based on multi-agent systems. Background Art
[0002] In the past decade, due to the widespread application of multi-agent systems (MASs) such as multi-robot systems, sensor networks, chaotic systems, etc. in the engineering field, their synchronization has attracted more and more attention. Usually, each agent shares information with adjacent agents through a type of communication and performs some tasks as a group, showing high flexibility and robustness. In coordinated control, the communication between agents in a multi-agent system plays an important role, among which the time-triggered sampling scheme is the most widely used in traditional cooperative control. Under the time-triggered sampling scheme, the agents exchange state information at a fixed period, and the controller executes control instructions at the same periodic interval. This periodic control strategy will lead to resource waste. At the same time, since most agents are only equipped with some simple digital microprocessors, the processing capabilities of the agents in computing and communication will be limited. Summary of the invention
[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides a distributed management method based on a multi-agent system, which is used to solve the problem of resource waste in the periodic control of each agent, namely the leader and the follower, in the multi-agent system in the prior art, so as to improve resource utilization efficiency and enhance the performance of the multi-agent system.
[0004] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:
[0005] The distributed management method based on a multi-agent system includes the following steps:
[0006] S1. Design a pulse protocol triggered by follower events;
[0007] S2, calculating the follower's control pulse input according to the pulse protocol triggered by the follower event;
[0008] S3. Based on the event-triggered pulse protocol and the follower's control pulse input, a multi-agent system model including leaders and followers is constructed to realize asynchronous and non-periodic control of the multi-agent system.
[0009] Furthermore, the pulse protocol triggered by the follower event in step S1 is:
[0010]
[0011] Among them, p i (t) represents the pulse protocol triggered by the ith follower event at time t, N irepresents the neighbor set of the ith follower, a ij represents the pulse gain from the ith follower to the jth follower, x j (t) represents the state of the j-th follower at the t-th time, x i (t) represents the state of the i-th follower at the t-th time, b i represents the internal pulse gain of the ith follower, and x0(t) represents the state of the leader at the tth moment.
[0012] Furthermore, the calculation formula of the control pulse input of the follower in step S2 is:
[0013]
[0014] Among them, u i (t) represents the control pulse input of the ith follower, γ represents the pulse gain matrix, δ represents the Dirac function, i.e., the instantaneous pulse, represents the i-th follower at the time when the k-th event is triggered, and N represents the total number of followers.
[0015] Furthermore, the multi-agent system model including leaders and followers constructed in step S3 is:
[0016]
[0017] in, represents the continuous state of the leader at time t, A represents the system matrix, represents the continuous state of the ith follower at time t, Δx i Represents the state change of the i-th follower.
[0018] The present invention has the following beneficial effects:
[0019] 1. The distributed management method based on the multi-agent system proposed in the present invention, by designing a pulse protocol triggered by follower events, enables each follower to exchange status information and execute control instructions only when an event is triggered, thereby avoiding the waste of resources caused by a fixed period. In particular, when the computing and communication capabilities of each follower are limited, this non-periodic control can significantly improve resource utilization; at the same time, the pulse protocol allows the follower to dynamically adjust its triggering conditions according to actual conditions, so that the entire multi-agent system can more flexibly respond to various complex environments and task requirements, thereby improving the adaptability and robustness of the multi-agent system;
[0020] 2. Using the pulse protocol triggered by follower events to realize asynchronous and non-periodic control of multiple agents, i.e. multiple followers, helps to improve the overall synchronization performance of the system and ensures that each follower can complete the task in a coordinated manner;
[0021] 3. By carefully designing the pulse protocol triggered by follower events and the control pulse input of the followers, precise control of the multi-agent system can be achieved, thereby improving the overall performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] Figure 1 This is a flow chart of the distributed management method based on a multi-agent system proposed in the present invention. DETAILED DESCRIPTION
[0023] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.
[0024] like Figure 1 As shown, the distributed management method based on the multi-agent system includes the following steps:
[0025] S1. Design a pulse protocol triggered by follower events.
[0026] In this embodiment, the purpose of designing the pulse protocol for follower event triggering is to define the triggering conditions of the follower in the event triggering, so that the follower is always under the leadership control of the leader, specifically: the state of the leader will be perceived or received by the follower, and serve as the basis for the follower to adjust its own state; and the main task of the follower is to follow the state change of the leader, and use the designed pulse protocol to adjust its own state to keep synchronization or near synchronization with the leader. In addition, such a design also allows each follower to exchange state information and execute control instructions only when the event is triggered, avoiding the waste of resources caused by a fixed period, especially when the computing and communication capabilities of each follower are limited, this non-periodic control can significantly improve resource utilization; at the same time, the pulse protocol allows the follower to dynamically adjust its triggering conditions according to the actual situation, so that the entire multi-agent system can more flexibly respond to various complex environments and task requirements, improving the adaptability and robustness of the multi-agent system.
[0027] Specifically, the pulse protocol triggered by the follower event in step S1 is:
[0028]
[0029] Among them, p j (t) represents the pulse protocol triggered by the ith follower event at time t, N i represents the neighbor set of the ith follower, a ijrepresents the pulse gain from the ith follower to the jth follower, x j (t) represents the state of the j-th follower at the t-th time, x i (t) represents the state of the i-th follower at the t-th time, b i represents the internal pulse gain of the ith follower, and x0(t) represents the state of the leader at the tth moment.
[0030] S2. Calculate the follower's control pulse input according to the pulse protocol triggered by the follower event.
[0031] In this embodiment, since the traditional time-triggered sampling method requires the leader and each agent, i.e., the follower, to exchange state information at a fixed period, this will lead to a waste of resources, especially when the state of each follower does not change frequently; therefore, the present invention triggers pulse control only when necessary through a pulse protocol triggered by a follower event, thereby realizing asynchronous and non-periodic control, thereby improving the flexibility and efficiency of the system; at the same time, it can ensure that the follower is always within the control range of the leader, and the entire multi-agent system can maintain a stable operating state. Therefore, the pulse protocol triggered by the follower event defines the conditions for triggering the control pulse, which are usually based on the current state of the follower, the neighbor state, and the leader state and the relative relationship between them. When a specific condition is met, the follower will trigger a control pulse; and according to the pulse protocol triggered by the follower event, the control pulse input of each follower is calculated, and this input is usually an instantaneous pulse (such as a Dirac function), which will affect the state of the follower at the triggering moment; at the same time, the intensity and direction of the control pulse input are determined by the pulse gain matrix and other related parameters.
[0032] Specifically, the calculation formula of the control pulse input of the follower in step S2 is:
[0033]
[0034] Among them, u i (t) represents the control pulse input of the ith follower, γ represents the pulse gain matrix, δ represents the Dirac function, i.e., the instantaneous pulse, represents the i-th follower at the time when the k-th event is triggered, and N represents the total number of followers.
[0035] S3. Based on the event-triggered pulse protocol and the follower's control pulse input, a multi-agent system model including leaders and followers is constructed to realize asynchronous and non-periodic control of the multi-agent system.
[0036] In this embodiment, a multi-agent system model including a leader and followers is constructed to describe the dynamic behavior of the system under asynchronous and non-periodic control, including the continuous state of the leader, the continuous state of the followers and the interaction between them.
[0037] Specifically, the multi-agent system model including leaders and followers constructed in step S3 is:
[0038]
[0039] in, represents the continuous state of the leader at time t, A represents the system matrix, represents the continuous state of the ith follower at time t, Δx i Represents the state change of the i-th follower.
[0040] In summary, the distributed management method based on a multi-agent system proposed in the present invention calculates the control pulse input through a pulse protocol triggered by follower events, which can realize asynchronous and non-periodic control of the multi-agent system, achieve resource optimization, and enhance the stability and flexibility of the system.
[0041] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.
[0042] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.
Claims
1. A distributed management method based on a multi-agent system, characterized in that: The following steps are involved: S1. Design a pulse protocol triggered by follower events; S2, calculating the follower's control pulse input according to the pulse protocol triggered by the follower event; S3. Based on the event-triggered pulse protocol and the follower's control pulse input, a multi-agent system model including leaders and followers is constructed to realize asynchronous and non-periodic control of the multi-agent system.
2. The distributed management method based on a multi-agent system according to claim 1, characterized in that: The pulse protocol triggered by the follower event in step S1 is: Among them, p i (t) represents the pulse protocol triggered by the ith follower event at time t, N i represents the neighbor set of the ith follower, a ij represents the pulse gain from the ith follower to the jth follower, x j (t) represents the state of the j-th follower at the t-th time, x i (t) represents the state of the i-th follower at the t-th time, b i represents the internal pulse gain of the ith follower, and x0(t) represents the state of the leader at the tth moment.
3. The distributed management method based on a multi-agent system according to claim 2, characterized in that: The calculation formula of the follower's control pulse input in step S2 is: Among them, u i (t) represents the control pulse input of the ith follower, γ represents the pulse gain matrix, δ represents the Dirac function, i.e., the instantaneous pulse, represents the i-th follower at the time when the k-th event is triggered, and N represents the total number of followers.
4. The distributed management method based on a multi-agent system according to claim 3, characterized in that: The multi-agent system model including leaders and followers constructed in step S3 is: in, represents the continuous state of the leader at time t, A represents the system matrix, represents the continuous state of the ith follower at time t, Δx i Represents the state change of the i-th follower.
