Safety consistency control method for multi-agent system

Abstract

The invention belongs to the field of multi-agent system control and particularly relates to a safety consistency control method for a multi-agent system. For any normal agent in the multi-agent system, the control method comprises steps: a sequence containing own information and neighbor delay information is ascended, the state values in the sequence that are not in the safety domain are removed, and a new sequence after preprocessing is thus obtained; the mean and the standard deviation of the new sequence are calculated with one standard deviation away from the mean as a constraint condition; the state values in the new sequence that are not in the constraint condition are eliminated until all kept state values in the new sequence meet the constraint condition; and the kept neighborinformation is used for own state updating. Malicious value judgment can be carried out on al neighbor information of a normal agent, and the anti-attack capability of the system is enhanced.

Description

technical field [0001] The invention belongs to the field of multi-agent system control, and is particularly aimed at a security consistency control method for a multi-agent system containing malicious agents, and specifically a security consistency control method for a multi-agent system. Background technique [0002] Multi-agent systems (MASs for short) is a new distributed computing technology. It has developed rapidly since its appearance in the 1970s, and has become a thinking method and tool for complex system analysis and simulation. [0003] In recent years, research on coordinated control of multi-agent systems inspired by swarm behavior has been widely used in wireless sensor networks, UAV formations, satellite formations, and swarm spacecraft sounding detection, and has become a research hotspot in the current control field. However, the actual environment in which the system exists is becoming more and more complex. As we all know, delays are ubiquitous in the n...

AI Technical Summary

Problems solved by technology

This method only selects the median of the neighbor information sequence for state update, which increases the deviation of its own state update; it needs to meet a high degree of network connectivity, that is, the r-robust graph; the network topology adopts a fixed communication topology, which cannot cope with flexible and scalable changed malicious node

[0007] (2) Remove the maximum or minimum value in the neighbor information sequence through adaptive parameters, and update the state with the retained information; for example, a security system with the ability to resist malicious attacks and communication delays proposed in Chinese patent CN105467839A Convergence control method, this method adaptively reduces the number of received information by constructing adaptive pruning parameters, and finally achieves the state consistency of all normal agents, but this method also needs to meet high network connectivity, That is, the r-robust graph; the number of malicious nodes that can be resisted is single, that is, when the network connectivity satisfies the 2-robust graph, any normal node can only be attacked by one malicious node

[0008] In summary, the current security consensus algorithms do not judge the optimal value of all neighbor information, which increases the deviation of their own state updates and the influence of malicious nodes on the state of the multi-agent system

In addition, the topology composed of normal nodes in the system requires high connectivity, which increases the computing and communication capabilities of each node and increases the control cost.

Images