2023-10-13 14:45:16
There is a growing fear that attack-capable military robots could be cracked by cybercriminals in peacetime and used to carry out terrorist attacks or other crimes. Some researchers have been working on this threat, with the aim of finding a solution that can also serve to stop hackers from the opposing side in the event of a war.
Fendy Santoso, from Charles Sturt University in Australia, and Anthony Finn, from the University of South Australia, have designed an algorithm that can stop a cyberattack of the type known as “Man-in-the-Middle (MitM) in a matter of seconds. ), into a fully autonomous military robot in a matter of seconds.
In an attack of this type, someone enters a communication or data transfer between the two legitimate parties and from there can access the devices involved and alter their operation.
In an experiment that used deep learning neural networks (a form of artificial intelligence) to simulate the behavior of the human brain, researchers trained a robot’s operating system to learn to recognize the “signature” of a cyber attack. MitM type.
The algorithm, tested in real time on a replica of a US Army ground combat robot, had a 99% success rate in preventing such attacks and a false positive rate of less than 2%, all of which demonstrates the effectiveness of the system.
The robot used in the research experiments. (Photo: Fendy Santoso / Charles Sturt University)
Santoso and Finn present the technical details of their protection system in the academic journal IEEE Transactions on Dependable and Secure Computing, under the title “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.” (Source: NCYT from Amazings)
#control #military #robot