An algorithm developed by researchers at Charles Sturt University and the University of South Australia shows high potential in detecting and blocking man-in-the-middle (MitM) attacks on military robots. The new algorithm stands out for its amazing success rate and can prevent such cyberattacks in seconds.
The research teams used neural networks and the process of Deep Learning to develop the algorithm, with the goal of emulating how the human brain works. Through this training, the robotic operating system of a military robot could be programmed to identify the signature of a MitM attack. The innovative power of the algorithm lies in its ability to autonomously detect whether a potential attacker has managed to sneak into an existing data connection and log activities, for example to control the robot.
The robot operating system (ROS), which has a high degree of vulnerability to data breaches and electronic attacks due to its high level of interconnectivity, was the researchers’ main focus, according to Professor Anthony Quinn Finn, an expert in autonomous systems at UniSA.
Details of the algorithm and its operation are detailed in a scientific paper titled “Trusted Operations of a Military Ground Robot in the Face of Man-in-the-Middle Cyber-Attacks Using Deep Learning Convolutional Neural Networks: Real-Time Experimental Outcomes.” The paper, published in open access in the journal IEEE Transactions on Dependable and Secure Computing, reports on tests of the algorithm on a model of a U.S. Army combat robot. Here, the algorithm was able to successfully defend against malicious attacks 99% of the time, with a false positive rate of less than 2%.
In a statement, Professor Anthony Finn confirmed that the algorithm developed performed better than any other detection technique currently in use worldwide for such cyberattacks. Dr. Fendy Santoso, a fellow at the Charles Sturt Artificial Intelligence and Cyber Futures Institute and co-author of the study, praises the system as robust and highly accurate in detecting intruders.
The further research path Finn and Santoso plan to take involves testing the algorithm on various robotic platforms, with a particular focus on drones, which present an additional level of challenge due to their rapid movement and more complex control functions. This approach once again underscores the importance and potential of this new algorithm for the future of global cybersecurity.
