Leader-following bipartite consensus of multiple uncertain Euler-Lagrange systems under deception attacks

Highlights：

• In the context of the previous effort, we encounter another problem in reality, namely electromagnetic interference in complex environments or hacking during remote control, which can lead to deception attacks in the transmission of signals between agents. In order to solve this problem, we add filters to the distributed observer and apply an adaptive solution to eliminate the attacks, while the filters do not affect the stability of the system when the attacks are not present. Furthermore, because of the adaptive presence, we can simply estimate the upper bound on the number of attacks that may be encountered during the operation in advance of the operation, and based on this upper bound, our filter will always work properly.

• As far as we know, interference in complex environments can exist between not only agents, but also systems. Then we add a parameter to the control input based on an adaptive distributed observer to eliminate the perturbation of this attack and thus make the system more secure.

摘要

•When the environment is complex and humans can only rely on agents to enter, it is usually hard for us to get accurate global information, i.e. ELSs are uncertain. Based on this, we make the system controllable by using an autonomous linear system, called an exosystem, to generate a large number of signals, and an adaptive distributed observer to observe the state of each agents, thus making a more reliable estimate of the global information.•In the context of the previous effort, we encounter another problem in reality, namely electromagnetic interference in complex environments or hacking during remote control, which can lead to deception attacks in the transmission of signals between agents. In order to solve this problem, we add filters to the distributed observer and apply an adaptive solution to eliminate the attacks, while the filters do not affect the stability of the system when the attacks are not present. Furthermore, because of the adaptive presence, we can simply estimate the upper bound on the number of attacks that may be encountered during the operation in advance of the operation, and based on this upper bound, our filter will always work properly.•As far as we know, interference in complex environments can exist between not only agents, but also systems. Then we add a parameter to the control input based on an adaptive distributed observer to eliminate the perturbation of this attack and thus make the system more secure.