Vehicle trust management for connected vehicles : final research report.

Lee, Insup; Bezzo, Nicola; Chang, Jian · 2016 · ROSA P / Carnegie-Mellon University. Technologies for Safe and Efficient Transportation University Transportation Center

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This research report addresses the vulnerability of autonomous and connected vehicles to cyber-attacks that compromise sensor data, thereby threatening vehicular safety. As vehicles increasingly rely on sensors for functions like adaptive cruise control and obstacle avoidance, malicious attacks on these sensors can lead to dangerous outcomes. The study focuses on developing a resilient control framework that allows a vehicle to maintain its desired state even when one or more sensor measurements are corrupted. The work specifically targets ground vehicles and aims to detect attacks and mitigate their effects through security-aware, attack-resilient estimators. The methodology employs a Recursive Adaptive Estimator (RAE), which modifies the standard Kalman Filter approach to accommodate potential sensor attacks. The RAE includes a "shield procedure" that adjusts the covariance matrix associated with measurement errors. When a measurement deviates significantly from the predicted state estimate, the algorithm increases the uncertainty weight of that specific measurement, effectively reducing its influence on the state estimation. The system is hierarchical: sensors measure environmental variables correlated with speed, a security module performs attack detection and state estimation, and a PID controller uses the estimated velocity to drive actuators. For experimental validation, the researchers used a ground vehicle platform equipped with three independent speed sensors: GPS and left and right encoders. They extracted a seventh-order dynamical model of the vehicle through tests on various indoor and outdoor surfaces. The findings demonstrate that the RAE algorithm successfully detects and mitigates malicious attacks on individual sensors. Extensive simulations in Matlab/Simulink and hardware implementations showed that the vehicle could reach and maintain its desired cruise speed even when one sensor was compromised. Specifically, the system could estimate the correct state and maintain performance as long as fewer than half of the sensors (N/2) were under attack. When an attack was injected, the algorithm added weight to the noise variance of the corrupted measurement, decreasing its trustworthiness and allowing the remaining valid sensors to guide the control loop. The report also outlines a trust management framework for Vehicle-to-Vehicle (V2V) communications, where trust is computed as a triple consisting of reputation, confidence, and default values based on static vehicle information and context. The significance of this work lies in providing a practical method for enhancing the cybersecurity of autonomous vehicles without requiring complete system redesigns. By demonstrating that control systems can remain stable despite sensor spoofing, the research highlights the importance of integrating security into the control loop. The authors conclude that while modern vehicles offer increased comfort and autonomy, they lack inherent security against hackers. Future work will expand these techniques to multi-vehicle networks using V2V and Vehicle-to-Infrastructure (V2I) protocols, as well as more complex applications like waypoint navigation and passenger automobiles.

Key finding

The recursive attack-resilient estimator detected and removed a spoofed or attacked sensor and maintained the desired cruise speed whenever fewer than half the sensors were compromised.

Methodology

modeling

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (8 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• v2x connected vehicle

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Methodological Resource: validation psychometrics