Keywords
Vehicle, Simulation, Trust, Autonomous

Technologies
Python, Simpy, ETSI

Vehicular communication is essential for autonomous driving, but malicious actors can broadcast false warnings, hide genuine hazards, or spoof their location. This project explores how autonomous vehicles can intelligently decide which other vehicles and messages it can trust. I built an ETSI‑standard simulation where vehicles exchange Cooperative Awareness Messages (CAMs) and Decentralized Environmental Notification Messages (DENMs). A portion of vehicles execute attacks such as false positives, false negatives, position spoofing, and replay.

Six trust models were developed, each adding sophistication. No Trust accepts all messages as a baseline. Simple Memory maintains trust scores via sensor verification. DST uses Dempster‑Shafer theory to represent belief, disbelief, and uncertainty. Basic Plausibility adds physics‑based checks. Cross‑Validation enables vehicles to share evidence and penalise unconfirmed reports. Organisational‑Social incorporates fleet reputation while ensuring direct distrust overrides.