Carding 4 Carders
Professional
- Messages
- 2,724
- Reaction score
- 1,588
- Points
- 113
Find out how MIT is changing the rules of the aviation game.
MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) is developing an artificial intelligence system for aircraft called Air Guardian. This system actively cooperates with the pilot, turning the computer into part of the team, and not just an emergency reserve.
Modern aircraft can provide the pilot with an exciting experience, but in certain situations, controlling them can become a challenge. Takeoff, landing, flying in crowded airspace, or sudden malfunctions can result in the pilot having to process a huge amount of data in a fraction of a second.
One example of a difficult situation was the collision of US Airways Flight 1549 with a flock of birds on takeoff from LaGardie Airport in New York on January 15, 2009. Pilot Chesley "Sully" Sullenberger made the decision to land an Airbus A320 on the Hudson River, saving the lives of 155 passengers and crew. However, experts say that with an artificial intelligence system on board, such a decision could have been avoided.
Traditionally, artificial intelligence in aviation is considered as an emergency warning system. However, MIT's Air Guardian offers a different approach. The system tracks not only the parameters of the aircraft, but also the actions of the pilot, acting like a real co-pilot. It analyzes the pilot's eye movements and creates "attention maps" to understand what the pilot is paying attention to.
The basis of this system is "liquid neural networks" that can adapt even after training. This allows the AI to build a model of what is happening from second to second and learn how to cooperate with the pilot.
As a result, the pilot remains the main person operating the aircraft, using his experience and knowledge. At the same time, Air Guardian monitors the pilot's attention and intervenes if something threatens the safety of the flight.
Recent tests have shown that the combination of a human and a machine reduces the risks in flight and increases the success of navigation between target points.
Ramin Hasani, a researcher at MIT CSAIL and inventor of liquid neural networks, stated: "This system represents an innovative approach to aviation with a focus on humans and AI support. Our approach provides a dynamic, adaptive solution, ensuring that AI not only replaces human judgment, but also complements it, increasing security and cooperation in the sky."
MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) is developing an artificial intelligence system for aircraft called Air Guardian. This system actively cooperates with the pilot, turning the computer into part of the team, and not just an emergency reserve.
Modern aircraft can provide the pilot with an exciting experience, but in certain situations, controlling them can become a challenge. Takeoff, landing, flying in crowded airspace, or sudden malfunctions can result in the pilot having to process a huge amount of data in a fraction of a second.
One example of a difficult situation was the collision of US Airways Flight 1549 with a flock of birds on takeoff from LaGardie Airport in New York on January 15, 2009. Pilot Chesley "Sully" Sullenberger made the decision to land an Airbus A320 on the Hudson River, saving the lives of 155 passengers and crew. However, experts say that with an artificial intelligence system on board, such a decision could have been avoided.
Traditionally, artificial intelligence in aviation is considered as an emergency warning system. However, MIT's Air Guardian offers a different approach. The system tracks not only the parameters of the aircraft, but also the actions of the pilot, acting like a real co-pilot. It analyzes the pilot's eye movements and creates "attention maps" to understand what the pilot is paying attention to.
The basis of this system is "liquid neural networks" that can adapt even after training. This allows the AI to build a model of what is happening from second to second and learn how to cooperate with the pilot.
As a result, the pilot remains the main person operating the aircraft, using his experience and knowledge. At the same time, Air Guardian monitors the pilot's attention and intervenes if something threatens the safety of the flight.
Recent tests have shown that the combination of a human and a machine reduces the risks in flight and increases the success of navigation between target points.
Ramin Hasani, a researcher at MIT CSAIL and inventor of liquid neural networks, stated: "This system represents an innovative approach to aviation with a focus on humans and AI support. Our approach provides a dynamic, adaptive solution, ensuring that AI not only replaces human judgment, but also complements it, increasing security and cooperation in the sky."
