Robust communication, cyber security and artificial intelligence for security in RPAS

This report summarizes the results of the first year's work on information security in the project Unmanned vehicles and cyber threats within the FoT thematic area Autonomous vehicles for territorial surveillance. We focus on remotely operated vehicles for territorial surveillance in a twenty-year perspective. We analyze robust communication for control and sensor data as well as for flight management and map out basic communication requirements for remote piloting. A cursory analysis suggests that uninterrupted remote control may be difficult to achieve over the entire theater of operations if faced with competent adversarial electronic warfare measures. Navigation relying on data from satellites is also easily disrupted. The cyber security analysis is grounded in understanding how vulnerabilities arise from a life cycle perspective and what capabilities that can be expected from different types of opponents. Examples of cyber-attacks on unmanned vehicles are analyzed and a fundamental threat analysis is carried out. We show how autonomous artificial intelligence can be used for building robust multi-layered cyber security. Specifically, we study how an AI supervisor aboard the craft can detect attacks at different system levels and intervene in emergencies when the operator is unable to communicate or otherwise control the situation.