Machine learning in avionics

1. A method for managing a flight of an aircraft, comprising the steps of:

receiving with a processor data from recordings of the flight of an aircraft; said data comprising data from sensors and/or data from onboard avionics;

generating a learning model by training the learning model on a large number of flight recordings relating to an aircraft type;

further generating the learning model by generating links between a state of an aircraft and flight parameters based on the large number of flight recordings relating to the aircraft type;

determining with the processor flight parameters representing an aircraft state of the aircraft at a point N on a basis of the received data;

providing with the processor the flight parameters representing the aircraft state of the aircraft at a point N+1, wherein the aircraft state at point N+1 being determined by inputting the flight parameters representing the aircraft state at point N to the learning model modelling links between state of an aircraft and flight parameters, the model being learnt by means of machine learning by training on the large number of flight recordings; and

using the flight parameters representing the aircraft state at point N+1 for trajectory prediction to assess aircraft performance, optimize flight operations, implement flight simulation, assist in mission management, assist in a piloting of the aircraft, and/or adjust avionic systems.

2. The method according to claim 1 , the step of determining the state of the aircraft state of the aircraft at point N+1 on the basis of the aircraft state of the aircraft at point N comprising the steps of:

determining the flight parameters SEP (Specific Excess Power), FF (Fuel Flow) and N1 on the basis of the aircraft state at point N by applying a model learnt by machine learning; and

determining the aircraft state at point N+1 on the basis of values of flight parameters SEP, FF and N1 by means of trajectory calculation,

wherein the SEP value denotes energy available for the aircraft to climb, the FF value denotes a variation in a fuel weight and the N1 value denotes a speed of rotation of a first stage of an engine, which influences fuel consumption.

3. The method according to claim 1 , the machine learning being unsupervised.

4. The method according to claim 1 , the machine learning being supervised.

5. The method according to claim 1 , the machine learning being performed offline.

6. The method according to claim 1 , the machine learning being performed online.

7. The method according to claim 1 , the machine learning comprising one or more algorithms selected from the algorithms comprising: support-vector machines; classifiers; neural networks; decision trees and/or steps from statistical methods such as Gaussian mixture models, logistic regression, linear discriminant analysis and/or genetic algorithms.

8. A computer program product, said computer program product comprising non-transitory code instructions for performing the steps of the method according to claim 1 when said computer program product is executed on a computer.

9. A system for implementing the steps of the method according to claim 1 , the system comprising one or more avionic systems such as a flight management system FMS and/or an electronic flight bag EFB.

10. The system according to claim 9 , further comprising one or more neural networks chosen from the neural networks comprising: an artificial neural network; an acyclic artificial neural network; a recurrent neural network; a feedforward neural network; a convolutional neural network; and/or a generative adversarial neural network.

11. The method according to claim 1 , wherein the flight parameters comprise at least SEP (Specific Excess Power) values, FF (Fuel Flow) values, and N1 values.

12. A method for managing a flight of an aircraft, comprising the steps of:

receiving with a processor data from recordings of the flight of an aircraft; said data comprising data from sensors and/or data from onboard avionics;

generating a learning model by training the learning model on a large number of flight recordings relating to an aircraft type utilizing as input a state of an aircraft and flight parameters as an output; and

further generating the learning model by generating links between the state of an aircraft and the flight parameters based on the large number of flight recordings relating to the aircraft type utilizing as input the state of the aircraft and the flight parameters as the output;

determining with the processor flight parameters representing an aircraft state of the aircraft at a point N on a basis of the received data;

providing with the processor the flight parameters representing the aircraft state of the aircraft at a point N+1, wherein the aircraft state at point N+1 being determined by inputting the flight parameters representing the aircraft state at point N to the learning model modelling links between state of an aircraft and flight parameters, the model being learnt by means of machine learning by training on the large number of flight recordings; and

using the flight parameters representing the aircraft state at point N+1 for trajectory prediction to assess aircraft performance, optimize flight operations, implement flight simulation, assist in mission management, assist in a piloting of the aircraft, and/or adjust avionic systems.

13. The method according to claim 12 , wherein the flight parameters comprise at least SEP (Specific Excess Power) values, FF (Fuel Flow) values, and N1 values.

14. A method for managing a flight of an aircraft, comprising the steps of:

receiving with a processor data from recordings of the flight of an aircraft; said data comprising data from sensors and/or data from onboard avionics;

generating a learning model by training the learning model on a large number of flight recordings relating to a specific aircraft;

further generating the learning model by generating links between a state of an aircraft and flight parameters based on the large number of flight recordings relating to the specific aircraft;

determining with the processor flight parameters representing an aircraft state of the aircraft at a point N on a basis of the received data;

providing with the processor the flight parameters representing the aircraft state of the aircraft at a point N+1, wherein the aircraft state at point N+1 being determined by inputting the flight parameters representing the aircraft state at point N to the learning model modelling links between state of an aircraft and flight parameters, the model being learnt by means of machine learning by training on the large number of flight recordings; and

using the flight parameters representing the aircraft state at point N+1 for trajectory prediction to assess aircraft performance, optimize flight operations, implement flight simulation, assist in mission management, assist in a piloting of the aircraft, and/or adjust avionic systems.

15. The method according to claim 1 , further comprising:

generating the learning model by training the learning model on a large number of flight recordings relating to a specific aircraft utilizing as input a state of an aircraft and flight parameters as an output; and

further generating the learning model by generating links between the state of an aircraft and the flight parameters based on the large number of flight recordings relating to the specific aircraft utilizing as input the state of the aircraft and the flight parameters as the output.

16. The method according to claim 15 , wherein the flight parameters comprise at least SEP (Specific Excess Power) values, FF (Fuel Flow) values, and N1 values.