Kidnapping Deep Learning-based Multirotors using Optimized Flying Adversarial Patches

TL;DR

This paper presents a novel multi-robot adversarial attack system using flying patches to manipulate deep learning-based pose estimation in autonomous multirotors, demonstrating physical kidnapping of a robot in real-world flights.

Contribution

It introduces flying adversarial patches mounted on robots to perform physical attacks on drone perception systems, extending the concept to multi-robot adversarial systems.

Findings

Methods effectively optimize multiple patches and positions.

Scalable attack strategies for multiple patches.

Successful physical kidnapping in real-world flights.

Abstract

Autonomous flying robots, such as multirotors, often rely on deep learning models that make predictions based on a camera image, e.g. for pose estimation. These models can predict surprising results if applied to input images outside the training domain. This fault can be exploited by adversarial attacks, for example, by computing small images, so-called adversarial patches, that can be placed in the environment to manipulate the neural network's prediction. We introduce flying adversarial patches, where multiple images are mounted on at least one other flying robot and therefore can be placed anywhere in the field of view of a victim multirotor. By introducing the attacker robots, the system is extended to an adversarial multi-robot system. For an effective attack, we compare three methods that simultaneously optimize multiple adversarial patches and their position in the input image. We show that our methods scale well with the number of adversarial patches. Moreover, we demonstrate physical flights with two robots, where we employ a novel attack policy that uses the computed adversarial patches to kidnap a robot that was supposed to follow a human.