ADIDRUS: Adversarial Detection, Inference, and Deceptive Response for Unmanned Systems
The increasing prevalence of unmanned aerial vehicles (UAVs) in both military and civilian applications has made them an attractive target for cyber attacks. ADIDRUS is our solution to the growing necessity to secure these systems in a proactive manner.
Instances of attacks in the past were successful primarily due to their use of deception, both social and technological. ADIDRUS is a platform-agnostic, software-based solution currently in development that hardens UAVs against attacks through the use of deception and counter-deception turned against the adversary.
With ADIDRUS, we consider the following techniques in defending critical unmanned systems:
- Deception detection: We extensively monitor likely targets for deceptive techniques, such as input from various sensors, to ensure correct operation.
- Defensive counter-deception: We delay and avoid attacks using techniques that involve misdirection and decoys deflect attempting attackers.
- Active counter-deception: We execute a strategy that deceives the attacker into thinking that the attack has succeeded, while keeping full control of the unmanned system, giving us insight on the attacker’s objectives and the ability to influence their beliefs and actions.
ADIDRUS is the epitome of risk management and resilience strategy. Our operating assumption is that using deception and counter-deception against adversaries is the only way to level the playing field. As the security community increasingly moves to a view that playing perimeter defense is outdated and outclassed, solutions like ADIDRUS will become more the norm. For now, however, the ADIDRUS approach is a distinct competitive advantage because it is proactive, not reactive.