Evaluating the Cybersecurity Risks of Cooperative Ramp Merging in a Mixed Traffic Environment
Xuanpeng Zhao (firstname.lastname@example.org), University of California, RiversideShow Abstract
Ahmed Abdo, University of California, Riverside
Xishun Liao, University of California, Riverside
Matthew Barth, University of California, Riverside
Guoyuan Wu, University of California, Riverside
The Connected and Automated Vehicle (CAV) technology has demonstrated the potential to greatly improve transportation mobility and energy efficiency. Such ubiquitous vehicular connectivity also opens a new door for cyber-attacks. In this study, we target a representative cooperative traffic management application, i.e., highway on-ramp merging, in a mixed traffic environment. We develop our threat model with two trajectory spoofing strategies on CAVs to create traffic congestion, and we also devise an attack-resilient strategy for system defense. Then, we leverage VENTOS, a state-of-the-art CAV-based cybersecurity simulation platform, to evaluate the cybersecurity risks of the attack and the performance of the proposed defense strategy. A comprehensive case study has been conducted across different traffic congestion levels (in terms of volume-to-capacity ratio), penetration rates of CAVs, and attack rates. The results show that that the performance of mobility and energy decreases when the attack rate increases. With proposed defense algorithm, the cyber-attack resilience of the system can be well improved.
PREDICTION-BASED GNSS SPOOFING ATTACK DETECTION FOR AUTONOMOUS VEHICLES
Sagar Dasgupta, University of AlabamaShow Abstract
Mizanur Rahman, University of Alabama, Tuscaloosa
Mhafuzul Islam, Clemson University
Mashrur Chowdhury, Clemson University
Global Navigation Satellite System (GNSS) provides Positioning, Navigation, and Timing (PNT) services for autonomous vehicles (AVs) using satellites and radio communications. Due to the lack of encryption, open-access of the coarse acquisition (C/A) codes, and low strength of the signal, GNSS is vulnerable to spoofing attacks compromising the navigational capability of the AV. A spoofed attack is difficult to detect as a spoofer (attacker who performs spoofing attack) can mimic the GNSS signal and transmit inaccurate location coordinates to an AV. In this study, we have developed a prediction-based spoofing attack detection strategy using the long short-term memory (LSTM) model, a recurrent neural network model. The LSTM model is used to predict the distance traveled between two consecutive locations of an autonomous vehicle. In order to develop the LSTM prediction model, we have used a publicly available real-world comma2k19 driving dataset. The training dataset contains different features (i.e., acceleration, steering wheel angle, speed, and distance traveled between two consecutive locations) extracted from the controlled area network (CAN), GNSS, and inertial measurement unit (IMU) sensors of AVs. Based on the predicted distance traveled between the current location and the immediate future location of an autonomous vehicle, a threshold value is established using the positioning error of the GNSS device and prediction error (i.e., maximum absolute error) related to distance traveled between the current location and the immediate future location. Our analysis revealed that the prediction-based spoofed attack detection strategy can successfully detect the attack in real-time.
A Data Driven Method for Falsified Vehicle Trajectories Identification by Anomaly Detection
Shihong Huang, University of Michigan, Ann ArborShow Abstract
Yiheng Feng (email@example.com), Purdue University
Henry Liu, University of Michigan, Ann Arbor
The vehicle-to-infrastructure (V2I) communications enable a wide range of new applications, which bring prominent benefits to the transportation system. Malicious attackers can potentially launch falsified data attacks against V2I applications to jeopardize the traffic operation and cause safety and mobility issues, by leveraging the new technology. As a result, to ensure the benefits brought by the V2I applications, it is critical to establish defense models to protect the applications from falsified data attacks. This paper proposes a data-driven method for identifying falsified trajectories. A trajectory embedding model, inspired by the word embedding model from the natural language processing (NLP) community, is developed that generates vector representations of vehicle trajectories, which can be used to compute the similarity between trajectories. The proposed method consists of two steps. In the first step, historical trajectory data are used to train a neural network to obtain the vector representations of trajectories. The second step computes a distance matrix between each pair of trajectories and identifies falsified trajectories using a hierarchical clustering algorithm. Simulation experiments show that the proposed method has a very high detection rate ( 99.0%) under different attack goals with varying connected vehicle (CV) penetration rates, while the false alarm rate remains low ( 6.6%). It has great potential to be implemented in different trajectory-based applications such as traffic state estimation and traffic signal control, and safeguard the CV system.
Minimum Detectable Error in Identifying Vehicle Malicious and Erroneous Misbehavior: Sensor-Based Misbehavior Detection Study
Boon Teck Ong, Booz Allen Hamilton, Inc.Show Abstract
Joshua Kolleda, Booz Allen Hamilton, Inc.
Saleh Mousa, Texas Department of Transportation
Scott Andrews, Cogenia Partners, LLC
Dennis Fleming, Cogenia Partners, LLC
James Marousek, Booz Allen Hamilton, Inc.
Mahsa Ettefagh, Booz Allen Hamilton, Inc.
Purser Sturgeon, Southwest Research Institute
Diego Lodato, Booz Allen Hamilton, Inc.
James Goldsmith, Booz Allen Hamilton, Inc.
This research establishes the minimum detectable error (MDE) boundary for relative position between the observer and status vehicles (SV) using vehicle sensor and GPS error profile from field tests and established minimum standards. The results demonstrated that the MDE increases in the lateral direction (side-to-side) with the increase in relative distance between the observer and SVs while remaining the same in the longitudinal direction (front-to-back). The implementation of Sensor-Based Misbehavior Detection (SBMD) with current specifications without GPS, sensors, and heading improvements can detect two-lane lateral offset malicious misbehavior and one vehicle length longitudinal offset malicious misbehavior with a probability above 99%. For one-lane lateral offset malicious misbehavior, there is a detection probability of approximately 30% at 70 meters distance and approximately 60% at 10 meters distance between observer and SV. Major improvements to sensors and heading error yield an improved detection for one-lane lateral offset malicious misbehavior with a probability approximately 60% for all distances while major improvements to GPS yield 100% detection of one-lane lateral offset malicious misbehavior. SBMD with current accuracy specifications without GPS, sensor, and heading improvements can detect GPS equipment failure where the GPS error profile is expanded to 3 times the current SAE J2945 specifications with a probability of approximately 42% at 70 meters distance and approximately 48% at 10 meters distance between observer and SV. This research found that major improvements to vehicle sensor and heading error reduces the MDE boundary lateral radius spread, thus decreasing the correlation between range and MDE boundary size.
DISCLAIMER: All information shared in the TRB Annual Meeting Online Program is subject to change without notice. Changes, if necessary, will be updated in the Online Program and this page is the final authority on schedule information.