Active Bayesian Inference for Robust Control under Sensor False Data Injection Attacks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Andersson, Axel, Dán, György
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918442970382336
author Andersson, Axel
Dán, György
author_facet Andersson, Axel
Dán, György
contents We present a framework for bridging the gap between sensor attack detection and recovery in cyber-physical systems. The proposed framework models modern-day, complex perception pipelines as bipartite graphs, which combined with anomaly detector alerts defines a Bayesian network for inferring compromised sensors. An active probing strategy exploits system nonlinearities to maximize distinguishability between attack hypotheses, while compromised sensors are selectively disabled to maintain reliable state estimation. We propose a threshold-based probing strategy and show its effectiveness via a simplified partially observable Markov decision process (POMDP) formulation. Experiments on an inverted pendulum under single and multi-sensor attacks show that our method significantly outperforms outlier-robust and prediction-based baselines, especially under prolonged attacks.
format Preprint
id arxiv_https___arxiv_org_abs_2604_11410
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Active Bayesian Inference for Robust Control under Sensor False Data Injection Attacks
Andersson, Axel
Dán, György
Machine Learning
Systems and Control
We present a framework for bridging the gap between sensor attack detection and recovery in cyber-physical systems. The proposed framework models modern-day, complex perception pipelines as bipartite graphs, which combined with anomaly detector alerts defines a Bayesian network for inferring compromised sensors. An active probing strategy exploits system nonlinearities to maximize distinguishability between attack hypotheses, while compromised sensors are selectively disabled to maintain reliable state estimation. We propose a threshold-based probing strategy and show its effectiveness via a simplified partially observable Markov decision process (POMDP) formulation. Experiments on an inverted pendulum under single and multi-sensor attacks show that our method significantly outperforms outlier-robust and prediction-based baselines, especially under prolonged attacks.
title Active Bayesian Inference for Robust Control under Sensor False Data Injection Attacks
topic Machine Learning
Systems and Control
url https://arxiv.org/abs/2604.11410