Assuring Safe Navigation and Network Operations of Autonomous Ships

Autonomous marine vehicles face escalating cyber-security threats as connectivity and automation increase attack surfaces. Cybercriminals have increased security intrusions targeting the marine industry by 400% during the COVID-19 epidemic. This research is a work-in-progress study investigating the feasibility of developing a modular battleship simulator and a simultaneous navigation and security artificial intelligence (AI) monitoring system with machine learning (ML) technologies. We used a simple but representative threat model based on GPS spoofing and integrity attacks against the weapons system and ICS network operations. Results indicate the feasibility of the modular battleship simulator and the associated AI monitoring system. The simulator utilizes maritime vehicle physics, rudimentary weapons systems operations, and ICS network operations. Most importantly, we demonstrate the ability of AI monitors to guard against navigation and network operation anomalies that would jeopardize a ship and/or its associated mission.