Diagnosis-guided Attack Recovery for Securing Robotic Vehicles from Sensor Deception Attacks

TL;DR

DeLorean is a comprehensive framework that detects, diagnoses, and recovers robotic vehicle sensors from complex deception attacks, ensuring operational resilience even under multi-sensor compromises.

Contribution

The paper introduces DeLorean, a novel unified approach combining attack diagnosis and targeted recovery for sensor deception attacks in robotic vehicles.

Findings

Successfully recovers RVs from SDAs in 93% of cases.

Effectively identifies targeted sensors using causal analysis.

Demonstrates robustness against multi-sensor deception attacks.

Abstract

Sensors are crucial for perception and autonomous operation in robotic vehicles (RV). Unfortunately, RV sensors can be compromised by physical attacks such as sensor tampering or spoofing. In this paper, we present DeLorean, a unified framework for attack detection, attack diagnosis, and recovering RVs from sensor deception attacks (SDA). DeLorean can recover RVs even from strong SDAs in which the adversary targets multiple heterogeneous sensors simultaneously. We propose a novel attack diagnosis technique that inspects the attack-induced errors under SDAs, and identifies the targeted sensors using causal analysis. DeLorean then uses historic state information to selectively reconstruct physical states for compromised sensors, enabling targeted attack recovery under single or multi-sensor SDAs. We evaluate DeLorean on four real and two simulated RVs under SDAs targeting various sensors, and we find that it successfully recovers RVs from SDAs in 93% of the cases.