Method for determining a flight plan
A computer implemented method to generate a flight plan for a flight mission of an aircraft by calculating, based on an optimization criterion, successive states, e.g., flight way points, along the flight plan to generate a four dimensional flight route including the aircraft latitude, longitude, altitude and time, and aircraft performance constraints results.
1. A computer-implemented method configured to determine a flight plan for an aircraft mission in a mission planning system, the mission planning system including:
a routing engine adapted to manage a set of solvers configured to solve different domain models in response to a routing request,
a database module comprising at least one database related to an aircraft and/or a mission context associated with the aircraft mission, and
a performance module in communication with the database module and the routing engine, wherein the performance module is configured to calculate aircraft performance constraints;
wherein the computer-implemented method comprises:
in a step A, providing to the routing engine a routing request according to the aircraft mission and information defining a context where the mission is to be performed, the routing request comprising at least an optimization criteria,
in a step B, transforming, by the routing engine, the routing request into a domain model to be solved, wherein the domain model comprises a cost function according to an optimization criteria,
in a step C, sending the domain model from the routing engine to a solver of the set of solvers to initiate the solver,
in a step D, determining, by the solver, for each of a plurality of states of the aircraft, wherein each of the states comprises at least aircraft latitude, aircraft longitude, aircraft altitude and time, and the states include an initial state, at least one intermediate state and a final state determined by the routing request, and wherein a next of the states of the aircraft as follows:
calculating, by the performance module, aircraft performance constraints according to the state of the aircraft and other parameters from the mission context information, and sending the aircraft performance constraints to the solver, and determining, by the solver, the next states according to the aircraft performance constraints and the cost function,
in a step E, selecting, by the solver, the best route solution from determined next states, and
in a step F, providing, by the routing engine, the flight plan comprising the selected route solution being a four-dimensional route including the aircraft latitude, longitude, altitude and time, and aircraft performance constraints results.
2. The computer-implemented method according to claim1 , wherein the performance module comprises a machine learning module trained for determining the aircraft performance constraints.
3. The computer-implemented method according to claim 1 , wherein the method further comprises before the step c) selecting by the routing engine the best solver for solving the routing request depending on domain characteristics.
4. The computer-implemented method according to claim 1 , wherein the routing request further comprises information of: departure and arrival aircraft condition, aircraft type and configuration, path geometry rules, or any combination of them.
5. The computer-implemented method according to claim 1 , wherein the mission context information comprises at least one of: terrain and obstacle databases, weather data, aircraft performance databases, air navigation databases, and environment data in which the mission is to be performed,
wherein the context information is comprised in the database module.
6. The computer-implemented method according to claim 1 , wherein the step b) comprises modelling the aircraft in a start state of the aircraft and creating a cost function according to the optimization criteria.
7. The computer-implemented method according to claim 1 , wherein the step b) further comprises a step of determining, by the routing engine, possible aircraft states according to the aircraft type and possible movements of the aircraft.
8. The computer-implemented method according to claim 1 , wherein the step a) is performed by an application programming interface of the mission planning system.
9. The computer-implemented method according to claim 8 , wherein the performance module determines performance constrains including at least one of kinematic constraints and fuel consumption of the aircraft, and the step d) of determining the next states by the solver is performed according to the at least one kinematic constraints and the fuel consumption.
10. The computer-implemented method according to claim 1 , wherein each solver can be a heuristic based solution including at least one of a: A* search algorithm, a reinforcement learning based solver, a genetic programming solution, a particle optimization based solver, and a Monte Carlo tree search algorithm.
11. The computer-implemented method according to claim 1 , wherein the method comprises a further step of validating the performance of the selected route solution by the performance module.
12. The computer-implemented method according to claim 1 , wherein the method further comprises a step of encoding the selected route solution by the routing engine, wherein encoding the route comprises geometry encoding and formatting.
13. A computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of claim 1 .
14. A computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out the method of claim 1 .
15. A mission planning system comprising means for carrying out the steps of the method of claim 1 .