Black-box Adversarial Attacks on CNN-based SLAM Algorithms
This addresses a security problem for autonomous navigation systems by exposing critical vulnerabilities in SLAM, though it is incremental as it focuses on a specific algorithm and dataset.
The paper tackles the vulnerability of CNN-based SLAM algorithms to adversarial attacks by introducing black-box perturbations on RGB images, resulting in tracking failure in up to 76% of frames on the TUM dataset.
Continuous advancements in deep learning have led to significant progress in feature detection, resulting in enhanced accuracy in tasks like Simultaneous Localization and Mapping (SLAM). Nevertheless, the vulnerability of deep neural networks to adversarial attacks remains a challenge for their reliable deployment in applications, such as navigation of autonomous agents. Even though CNN-based SLAM algorithms are a growing area of research there is a notable absence of a comprehensive presentation and examination of adversarial attacks targeting CNN-based feature detectors, as part of a SLAM system. Our work introduces black-box adversarial perturbations applied to the RGB images fed into the GCN-SLAM algorithm. Our findings on the TUM dataset [30] reveal that even attacks of moderate scale can lead to tracking failure in as many as 76% of the frames. Moreover, our experiments highlight the catastrophic impact of attacking depth instead of RGB input images on the SLAM system.