当前位置: X-MOL 学术IEEE Trans. Cybern. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Supervisory Nonlinear State Observers for Adversarial Sparse Attacks
IEEE Transactions on Cybernetics ( IF 11.8 ) Pub Date : 2020-05-29 , DOI: 10.1109/tcyb.2020.2992762
Liwei An 1 , Guang-Hong Yang 1
Affiliation  

This article investigates the problem of secure state estimation for continuous-time linear systems in the presence of sparse sensor attacks. Compared with the existing results, the attacked sensor set can be changed by adversaries against secure estimation. To address the more erratic attacks, a novel supervisory state observer is proposed, which employs a bank of candidate nonlinear subobservers and a switching logic administrated by a monitoring function to select the active subobserver at every instant of time. Based on the stability analysis of switched systems, it is proven that the supervisory observer asymptotically converges to a neighborhood of the true system state in the presence of the sensor attacks and bounded disturbances. A simulation example is given to substantiate the theoretical results.

中文翻译:

对抗性稀疏攻击的监督非线性状态观察器

本文研究了在存在稀疏传感器攻击的情况下连续时间线性系统的安全状态估计问题。与现有结果相比,被攻击的传感器集可以被攻击者改变以对抗安全估计。为了解决更不稳定的攻击,提出了一种新颖的监督状态观察器,它采用一组候选非线性子观察器和一个由监控功能管理的切换逻辑,以在每个时刻选择活动的子观察器。基于切换系统的稳定性分析,证明了在存在传感器攻击和有界干扰的情况下,监督观测器渐近收敛到真实系统状态的邻域。给出了一个仿真例子来证实理论结果。
更新日期:2020-05-29
down
wechat
bug