Researchers from Charles Sturt University (Australia) and the University of South Australia (UniSA) developed a novel algorithm that allows to block a Man-in-the-Middle (MitM) attack on an unmanned military robot in a matter of seconds.
The researchers trained a robot’s operating system to identify the signatures of a MitM eavesdropping cyber attack, which involves interrupting an existing data transfer between an unmanned robot and a control center.
Based on the concepts of deep learning convolutional neural networks (CNNs), the new method is meant to reduce the vulnerability of the Robot Operating System (ROS), a well-known middleware platform widely used in both civilian and military robots.
“The robot operating system (ROS) is extremely susceptible to data breaches and electronic hijacking because it is so highly networked,” the researchers said. “The advent of Industry 4, marked by the evolution in robotics, automation, and the Internet of Things, has demanded that robots work collaboratively, where sensors, actuators and controllers need to communicate and exchange information with one another via cloud services. The downside of this is that it makes them highly vulnerable to cyberattacks.”
The new algorithm has been tested on a replica of a US Army combat ground vehicle and shown to be 99% successful at preventing a malicious attack, with false positive rates of less than 2%.
The experts collected network data of the robot operating under “legitimate and malicious conditions” and used that to train a convolutional neural network (CNN) to try and identify attack traffic.
The researchers plan to further test their algorithm on different robotic platforms, such as drones, whose dynamics are faster and more complex compared to a ground robot.