Adversarial Driving: Attacking End-to-End Autonomous Driving
arXiv:2103.09151 · doi:10.1109/IV55152.2023.10186386
Abstract
As research in deep neural networks advances, deep convolutional networks become promising for autonomous driving tasks. In particular, there is an emerging trend of employing end-to-end neural network models for autonomous driving. However, previous research has shown that deep neural network classifiers are vulnerable to adversarial attacks. While for regression tasks, the effect of adversarial attacks is not as well understood. In this research, we devise two white-box targeted attacks against end-to-end autonomous driving models. Our attacks manipulate the behavior of the autonomous driving system by perturbing the input image. In an average of 800 attacks with the same attack strength (epsilon=1), the image-specific and image-agnostic attack deviates the steering angle from the original output by 0.478 and 0.111, respectively, which is much stronger than random noises that only perturbs the steering angle by 0.002 (The steering angle ranges from [-1, 1]). Both attacks can be initiated in real-time on CPUs without employing GPUs. Demo video: https://youtu.be/I0i8uN2oOP0.
arXiv commentAccepted by IEEE Intelligent Vehicle Symposium, 2023
References in corpus (9)
- Explaining and Harnessing Adversarial Examples
- End to End Learning for Self-Driving Cars
- A Survey of Autonomous Driving: Common Practices and Emerging Technologies
- Robot Operating System 2: Design, Architecture, and Uses In The Wild
- A Survey of End-to-End Driving: Architectures and Training Methods
- A Review of Adversarial Attack and Defense for Classification Methods
- Trajectory-guided Control Prediction for End-to-end Autonomous Driving: A Simple yet Strong Baseline
- Evaluating Adversarial Attacks on Driving Safety in Vision-Based Autonomous Vehicles
- Adversarial Attacks, Regression, and Numerical Stability Regularization