Fooling a self-driving car with mirrors on traffic cones
SMRTR summary
European researchers demonstrated how mirrors can manipulate self-driving cars by tricking LIDAR sensors, making vehicles either fail to detect real obstacles (Object Removal Attack) or brake for nonexistent objects (Object Addition Attack). Using simple mirrors on traffic cones, they achieved up to 74% success rate in creating phantom obstacles, revealing a significant vulnerability in autonomous vehicle systems that could potentially cause dangerous driving behaviors in real traffic situations.
SMRTR provides this summary for quick context. The original article belongs to Daily.dev.
Read the original article