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

TL;DR

This paper introduces a Bayesian framework with active probing to detect and recover from sensor false data injection attacks in cyber-physical systems, improving robustness and reliability.

Contribution

It develops a novel active probing strategy leveraging system nonlinearities and Bayesian inference to enhance attack detection and sensor recovery.

Findings

Significantly outperforms baseline methods in experiments.

Effective in detecting prolonged sensor attacks.

Utilizes a threshold-based probing strategy within a POMDP framework.

Abstract

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.