Adversarial Attack Injection in Carla PDF Print E-mail

The tool is a custom version of the LearningByCheating autonomous driving agent and related suite, which has been integrated with the IBM Adversarial Robustness Toolbox (ART) for the injection of 4 attacks on the RGB camera:

  • Spatial Transformation (STA),
  • HopSkipJump (HSJ),
  • Basic Iterative Method (BIM),
  • NewtonFool (NF).
The software is available at https://github.com/piazzesiNiccolo/myLbc

The software is included in the "awesome carla" repo:  https://github.com/Amin-Tgz/awesome-CARLA

Niccolò Piazzesi, "Attacchi verso sistemi di apprendimento in ambito autonomous driving: studio e implementazione in ambienti simulati (in Italian)", Bachelor Thesis at the University of Florence. Supervisor: Andrea Ceccarelli. Link to the Thesis: http://rcl.dimai.unifi.it/publication/show/912-2


Resilient Computing Lab, 2011

Joomla - Realizzazione siti web