Cyber Algorithm Halts Malicious Robotic Assault

Home Research Cyber Algorithm Halts Malicious Robotic Assault
ai cyber algorithm

Researchers in Australia have developed a cutting-edge algorithm capable of swiftly halting a man-in-the-middle (MitM) cyberattack on an unmanned military robot.


By leveraging deep learning neural networks to emulate human brain behavior, experts in artificial intelligence from Charles Sturt University and the University of South Australia (UniSA) trained the robot’s operating system to recognize the signature of a MitM eavesdropping cyberattack, a tactic employed by attackers to disrupt ongoing conversations or data transfers.


The algorithm, assessed in real-time using a replica of a United States army combat ground vehicle, achieved a remarkable 99% success rate in thwarting malicious attacks. It also demonstrated its effectiveness by maintaining false positive rates of under 2%. These impressive findings have been documented in IEEE Transactions on Dependable and Secure Computing.


Professor Anthony Finn, an autonomous systems researcher at UniSA, highlighted that this algorithm outperforms various other recognition methods used globally to detect cyberattacks. Collaborating with the US Army Futures Command, Professor Finn and Dr. Fendy Santoso from Charles Sturt Artificial Intelligence and Cyber Futures Institute executed a man-in-the-middle cyberattack on a GVT-BOT ground vehicle, training its operating system to identify such an attack.


Prof Finn explained, “The robot operating system (ROS) is highly susceptible to data breaches and electronic hijacking due to its extensive networking. The rise of Industry 4, characterized by advancements in robotics, automation, and the Internet of Things, necessitates robots to collaborate and share information through cloud services. Unfortunately, this makes them susceptible to cyberattacks. The silver lining, however, is that computing speed doubles every few years, allowing for the development and implementation of advanced AI algorithms to safeguard systems from digital threats.”


Dr. Santoso emphasized that despite its widespread use and numerous advantages, the robot operating system often neglects security concerns within its coding due to encrypted network traffic data and limited integrity-checking capabilities. He expressed, “Benefitting from the strengths of deep learning, our intrusion detection framework exhibits robustness and exceptional precision. This platform excels at handling vast datasets, rendering it well-suited for fortifying expansive, data-driven systems operating in real-time, such as ROS.”


Prof Finn and Dr. Santoso have plans to test their intrusion detection algorithm on various robotic platforms, including drones, which exhibit faster and more intricate dynamics compared to ground robots.