Multi-agent consistency control method with model uncertainty

Abstract

The invention discloses a multi-agent consistency control method with the model uncertainty, which reduces the calculation processing load of an agent processor and the communication frequency betweenmultiple agents, can reduce the data communication amount each time, can greatly reduce the network transmission load in the communication mechanism with reduced communication frequency and reduced single data communication traffic, can replace the original protocol for using numerical differentiation to approximately calculate a time lag state derivative, and selects an appropriate sampling period to ensure the consistency so as to improve the stability of the system; and the multi-agent consistency control method also adopts a consistency control scheduling scheme combining a self-running control strategy and a self-adjusting control strategy based on model uncertainty, thereby effectively solving the problem that the multi-agent consistency control methods relying on real-time continuous information exchange are difficult to implement in a real network application condition or multi-agent collaborative control confusion occurs in the prior art.

Description

technical field [0001] The invention relates to the technical field of multi-agent control, in particular to a multi-agent consistency control method with model uncertainty. Background technique [0002] In recent years, with the cross-development of many disciplines such as computer network, electronic communication, biotechnology, automatic control and artificial intelligence, multi-agent network, as the core of distributed system control technology, has increasingly attracted the attention and interest of many researchers. . Multi-agent network is not only an important branch of artificial intelligence research, but also a very important direction in the research of complex systems. [0003] First, the definition of an agent is given. An agent refers to a single device or machine with independent sensors, processors, and communication elements, and has certain self-organization, learning, and reasoning abilities. Therefore, a multi-agent system refers to a communication...

AI Technical Summary

Problems solved by technology

If the multi-agent system runs in a stable and stable state without sampling, the disadvantages of this time-continuously scheduled information sampling and interactive control method will appear, that is, one is to waste the computing resources of the agent processor, The second is to waste the network bandwidth resources for communication between multi-agents

[0006] Not only that, the existing multi-agent system needs to rely on the consistent control mode of real-time continuous information exchange, which also encounters challenges and difficulties in practical application

Because, in many outdoor real network conditions, due to the limitation of the network communication environment, the network communication bandwidth and transmission rate are limited, and even effective information transmission cannot be carried out in some time periods, so the precise state of each agent controller It is difficult to achieve real-time and continuous mutual exchange of information values, which makes it difficult to implement the multi-agent consensus control method that relies on real-time continuous information exchange in many real network application conditions, or problems such as multi-agent cooperative control confusion