DoubleStar: Long-Range Attack Towards Depth Estimation based Obstacle Avoidance in Autonomous Systems

Zhou, Ce; Yan, Qiben; Shi, Yan; Sun, Lichao

doi:10.48550/arxiv.2110.03154

Cited by 3 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bad actors can modify the lane-keeping system by installing dirty road patches and ultimately causing the ADV to drift away from its lane (Sato et al, 2021). Jamming the cameras' modules (Panoff et al, 2021) or LIDAR spoofing attacks (Zhou et al, 2021) to inject false obstacle depth lead to false sensor data and hence causes the ADV data processing chain to compute erroneous control commands. In these cases, the health of the sensors is not compromised hence remain undetected by traditional fault detection schemes.…”

mentioning

confidence: 99%

Toward resilient autonomous driving—An experience report on integrating resilience mechanisms into the Apollo autonomous driving software stack

Lucchetti

Graczyk

Völp

2023

Front. Comput. Sci.

View full text Add to dashboard Cite

Autonomous driver assistance systems (ADAS) have been progressively pushed to extremes. Today, increasingly sophisticated algorithms, such as deep neural networks, assume responsibility for critical driving functionality, including operating the vehicle at various levels of autonomy. Elaborate obstacle detection, classification, and prediction algorithms, mostly vision-based, trajectory planning, and smooth control algorithms, take over what humans learn until they are permitted to control vehicles and beyond. And even if humans remain in the loop (e.g., to intervene in case of error, as required by autonomy levels 3 and 4), it remains questionable whether distracted human drivers will react appropriately, given the high speed at which vehicles drive and the complex traffic situations they have to cope with. A further pitfall is trusting the whole autonomous driving stack not to fail due to accidental causes and to be robust against cyberattacks of increasing sophistication. In this experience report, we share our findings in retrofitting application-agnostic resilience mechanisms into an existing hardware-/software-stack for autonomous driving—Apollo—as well as where application knowledge helps improve existing resilience algorithms. Our goal is to ultimately decrease the vulnerability of autonomously driving vehicles to accidental faults and attacks, allowing them to absorb and tolerate both, as well as to come out of them at least as secure as before the attack has happened. We demonstrate replication and rejuvenation on the driving stack's Control module and indicate how this resilience can be extended both downwards to the hardware level, as well as upwards to the prediction and planning modules.

show abstract

mentioning

confidence: 99%

Toward resilient autonomous driving—An experience report on integrating resilience mechanisms into the Apollo autonomous driving software stack

Lucchetti

Graczyk

Völp

2023

Front. Comput. Sci.

View full text Add to dashboard Cite

show abstract

“…Several works have used non-GPS techniques for a hostile takeover of UAVs. These include the use of lasers to activate obstacle detection and avoidance systems [250], attacking the data-link between the radio-controller and the UAV [198] and [188] where the authors showcase the vulnerabilities present in a popular UAV platform from Parrot [12] as a result of a poorly configured wireless network.…”

Section: Related Workmentioning

confidence: 99%

Towards secure autonomous aerial vehicle navigation

Sathaye

View full text Add to dashboard Cite

The modern aerial vehicle ecosystem relies heavily on various wireless communication and navigation technologies that are often unauthenticated and vulnerable to adversarial interference. This thesis evaluates the security of three critical components of automation systems in modern aerial vehicles: instrument landing system (ILS), aviation datalink applications like controller-pilot datalink communications (CPDLC), and satellite navigation systems such as global positioning system (GPS).First, we demonstrate the feasibility of manipulating ILS instruments using commercially available software-defined radio (SDR) and a closed-loop ILS spoofer capable of manipulating spoofing signals based on the aircraft's position. In the second part, we propose a spoof-and-jam strategy to manipulate flight crew decision-making by implementing a reactive jammer for aviation datalink applications with a fast reaction time and high jamming success rate. The third part investigates the feasibility of controlling unmanned aerial vehicle (UAV) movements solely by GPS spoofing, highlighting the challenges of achieving a complete UAV takeover without causing crashes. We explore generic and UAV-specific strategies to control the UAV's speed and direction, successfully taking over consumer-grade UAVs from leading manufacturers. Finally, to address security issues surrounding satellite navigation, we design and implement SemperFi, a single antenna GPS receiver capable of tracking legitimate GPS signals and autonomously recovering from spoofing attacks. Sem-perFi leverages the extended Kalman filter (EKF) sensor-fusion mechanism built into most existing UAVs and a custom-designed legitimate signal retriever module to recover from attacks with high accuracy and a fast recovery time.Overall, we address critical security concerns in modern aerial vehicles and pave the way for developing secure autonomous technologies by leveraging interconnected and tightly coupled sensors and individual systems.

show abstract

Proposed method for building an anti-drone system for the protection of facilities important for state security

Łukasiewicz¹,

Kobaszyńska-Twardowska²

2022

Security and Defence Quarterly

View full text Add to dashboard Cite

Unmanned aerial vehicles (UAVs) pose a threat to buildings and facilities important to the security of the state. As they are able to operate like individual aircraft, the number of ways they can be used for terrorist activity is practically unlimited. Anyone in charge of a facility that is crucial for the reliable functioning of a state is obliged to ensure an acceptable level of security. Since drones can be used to attack protected structures, they need to be protected by an anti-drone system. The paper proposes a method for assessing the effectiveness of systems for detecting and neutralising unmanned aerial vehicles. In order to suggest a new method for assessing the effectiveness of anti-drone systems, an analysis of the scientific literature and other documents describing existing anti-drone systems has been carried out. Attacks involving the use of drones, both in wartime and in incidents of terrorism, are also analysed and existing anti-drone solutions assessed. Because there are a variety of technical solutions for the detection and neutralisation of drones, and different location and weather conditions, a universal method is proposed based on probability calculations and neutralisation of drones, using mathematical formulas. This method allows for the effectiveness of the entire anti-drone system to be assessed on the basis of measuring the probability of detection and neutralisation of drones in real conditions. The proposed method allows the effectiveness of the currently existing anti-drone systems to be evaluated and for new methods for detecting and neutralising drones to be proposed. This method, based on mathematical calculations, enables software to be written for simulating anti-drone systems on computers and for the effectiveness of these systems to be confirmed before their construction in a protected facility.

show abstract

DoubleStar: Long-Range Attack Towards Depth Estimation based Obstacle Avoidance in Autonomous Systems

Cited by 3 publications

References 20 publications

Toward resilient autonomous driving—An experience report on integrating resilience mechanisms into the Apollo autonomous driving software stack

Toward resilient autonomous driving—An experience report on integrating resilience mechanisms into the Apollo autonomous driving software stack

Towards secure autonomous aerial vehicle navigation

Proposed method for building an anti-drone system for the protection of facilities important for state security

Contact Info

Product

Resources

About