Kalman Filter (KF) is widely used in various domains to perform sequential learning or variable estimation. In the context of autonomous vehicles, KF constitutes the core component of many Advanced Driver Assistance Systems (ADAS), such as Forward Collision Warning (FCW). It tracks the states (distance, velocity etc.) of relevant traffic objects based on sensor measurements. The tracking output of KF is often fed into downstream logic to produce alerts, which will then be used by human drivers to make driving decisions in near-collision scenarios. In this paper, we study adversarial attacks on KF as part of the more complex machine-human hybrid system of Forward Collision Warning. Our attack goal is to negatively affect human braking decisions by causing KF to output incorrect state estimations that lead to false or delayed alerts. We accomplish this by sequentially manipulating measure ments fed into the KF, and propose a novel Model Predictive Control (MPC) approach to compute the optimal manipulation. Via experiments conducted in a simulated driving environment, we show that the attacker is able to successfully change FCW alert signals through planned manipulation over measurements prior to the desired target time. These results demonstrate that our attack can stealthily mislead a distracted human driver and cause vehicle collisions.


    Access

    Download


    Export, share and cite



    Title :

    Sequential Attacks on Kalman Filter-based Forward Collision Warning Systems


    Contributors:
    Ma, Yuzhe (author) / Sharp, Jon (author) / Wang, Ruizhe (author) / Fernandes, Earlence (author) / Zhu, Xiaojin (author)

    Publication date :

    2020


    Remarks:

    Accepted by AAAI21


    Type of media :

    Preprint


    Type of material :

    Electronic Resource


    Language :

    English




    Demo: Sequential Attacks on Kalman Filter-based Forward Collision Warning Systems

    Ma, Yuzhe / Sharp, Jon / Wang, Ruizhe et al. | TIBKAT | 2021


    Vehicle-mounted forward collision early warning method and system based on OpenCv Kalman filter

    HE ZHAODAN / SUN WENJIAN | European Patent Office | 2015

    Free access

    Forward collision warning

    ANNIGERI ROHIT / SRINIVASAN SHARAN / LAI KEVIN et al. | European Patent Office | 2025

    Free access

    Anti-Collision Warning of Offshore Wind Farm Based on Kalman Filter

    Dai, Cheng / Xie, Haibo / Bai, Weiwei | IEEE | 2023


    Child forward collision warning

    MELL BOAZ / HAREL ADAM / RAICHELGAUZ IGAL | European Patent Office | 2024

    Free access