Published:
Building CARLA from source on Windows is notoriously painful — outdated guides, cryptic errors, and hours of compilation. This post is based on this excellent guide by wambitz, combined with my own personal experience, additional fixes, and troubleshooting tips for issues I ran into that weren’t covered elsewhere.
Published:
Autonomous vehicles (AVs)—including self-driving cars, robots, and drones—rely heavily on multi-modal sensor pipelines to operate safely. However, these sensors are known to be vulnerable to adversarial attacks. A major gap in current research is the lack of a systematic ecosystem view: how exactly do sensor-induced errors propagate through the interconnected modules of an AV to eventually cause physical harm?
Published:
In the rapidly evolving field of autonomous systems, visual perception is the cornerstone of safe navigation. A recent paper titled “Investigating Physical Latency Attacks Against Camera-Based Perception”, accepted to the 2025 IEEE Symposium on Security and Privacy (S&P), introduces a novel and concerning threat vector: Detstorm.
Published:
Autonomous Target Tracking (ATT) systems, especially in drones, are becoming increasingly common for surveillance and photography. However, their reliance on visual tracking algorithms introduces significant security vulnerabilities. A new paper titled “FlyTrap: Physical Distance-Pulling Attack Towards Camera-based Autonomous Target Tracking Systems” exposes a critical flaw that allows attackers to manipulate drones into dangerous proximity.
Published:
This winter, I embarked on an epic cross-country road trip, driving over 2,200 miles from Columbus, Ohio to Los Angeles, California. It was an unforgettable journey through the changing landscapes of America.