Navigating the modern skies is not just a matter of skill, but also of managing an avalanche of data in real time. Imagine being the pilot in command of an airplane, with dozens of displays flashing and only a moment to make a crucial decision.
What if you had a digital co-pilot, ready to collaborate with you, understand your decisions and intervene only when necessary? MIT just unveiled Air Guardian (here's the research): it is not the usual artificial intelligence that we know, but a real flight companion.
Air Guardian: the evolution of the digital co-pilot
The era in which artificial intelligence was relegated to the role of a simple emergency alert system is now behind us. The MIT CSAIL laboratory decided to take AI to the next level, turning it into an active and collaborative co-pilot.
Do you all remember the US Airways Flight 1549 crash in 2009? The legendary Clint Eastwood also did it a film. The pilot Chesley “Sully” Sullenberger to land an Airbus A320 directly into the Hudson River, with an incredible landing that saved 155 lives in addition to his own. Well, an AI expert revealed that, if the pilot had had more time to evaluate the situation, he could have avoided the ditching. What if he had an AI system like Air Guardian at his side?
Beyond simple monitoring
Traditional AI systems for aircraft were designed as warning systems. They are there, silently, monitoring flight data and only intervening when something goes wrong. Air Guardian goes further. It not only monitors the plane, but also its “driver”. It uses its training to track the pilot's eye movements, understanding where he is looking and how much attention he is paying to each display.
Air Guardian is based on advanced algorithms and what is called a "liquid neural network", an extremely flexible network capable of adapting even after being trained. This ability to adapt allows the AI to build a model of what is happening from second to second and literally learn to cooperate with the pilot.
The pilot in command, AI as co-pilot
Without mincing words: if the pilot is not paying attention to a crucial piece of data or is dwelling too much on an irrelevant detail, the AI intervenes. In recent tests, the collaboration between pilot and Air Guardian reduced risks and improved navigation precision.
Ramin Hasani, an MIT CSAIL-affiliated researcher and inventor of liquid neural networks, said: “This system represents the innovative approach of human-centered AI in aviation. Our use of liquid neural networks offers a dynamic and adaptive approach, ensuring that AI does not replace human judgment, but complements it, leading to greater safety and collaboration in the skies.”
In other words: for MIT, solutions like Air Guardian make the future of aviation safer and more collaborative. The focus remains on the human being, but AI officially enters the cabin.
Would you take a plane that no longer has two human pilots but only one, supported by an AI co-pilot? Let me know on the Futuroprossimo.it social channels.
