AI Algorithm Against Cyberattacks on Robots

image provided by pixabay

This post is also available in: עברית (Hebrew)

Australian researchers have developed an AI algorithm to detect and stop a cyberattack on a military robot in seconds using deep-learning neural networks.

Deep learning neural networks mimic how the human brain works, and are used to train the robot’s operating system to recognize the signature of a man-in-the-middle (MitM) cyberattack, which involves hackers intercepting and altering the communication between two parties.

According to Interesting Engineering, the researchers tested their algorithm on a replica of a US Army combat ground vehicle and found it was 99% effective in preventing a malicious attack. The system also had a low false positive rate of less than 2%, meaning it did not mistake everyday communication for an attack.

The study was published in IEEE Transactions on Dependable and Secure Computing and claims that this algorithm outperforms other globally used methods of detecting cyberattacks.

Professor Anthony Finn, an autonomous systems researcher at UniSA was one of the leaders of the research. He states that the robot operating system (ROS) is prone to cyberattacks because it is highly networked, and explains that Industry 4, characterized by advancements in robotics, automation, and the Internet of Things, requires robots to work together, where sensors, actuators, and controllers communicate and share information via cloud services. This makes them very vulnerable to cyberattacks.

Dr Santoso, an AI and cyber futures expert at Charles Sturt University and co-leader of the research, said the robot operating system needs adequate security measures in its coding scheme due to encrypted network traffic data and limited integrity-checking capability. He stated that their intrusion detection framework, which leverages the benefits of deep learning, is robust and highly accurate. He also said the system can handle large datasets suitable for securing large-scale and real-time data-driven systems such as ROS.

Interesting Engineering reports that the researchers plan to test their algorithm on other robotic platforms, like drones which have faster and more complex dynamics than a ground robot.