EP-4741833-A1 - SYSTEM AND METHOD FOR MONITORING AN AIRCRAFT STATE

Abstract

Methods and systems for monitoring an aircraft state are provided. A method includes acquiring a primary value indicative of the aircraft state using a first sensor, and generating a preliminary synthetic value indicative of the aircraft state using other data acquired from a second sensor having a different operating principle than the first sensor. A corrective value to be applied to the preliminary synthetic value is computed using a computer-implemented model having been trained using machine learning and training data associating previous values of the other data with previous corrective values. The corrective value is applied to the preliminary synthetic value to generate a corrected synthetic value indicative of the aircraft state. The corrected synthetic value may be used as a basis to control the aircraft when the primary value is deemed unreliable or unavailable.

Inventors

• TESSIER, CLAUDE

Assignees

• Bombardier Inc.

Dates

Publication Date

20260513

Application Date

20251106

Claims (15)

1. A method of operating an aircraft, the method comprising: acquiring primary pneumatic data indicative of an aircraft state using a sensor interacting with an air stream outside of the aircraft during flight of the aircraft; generating preliminary synthetic data indicative of the aircraft state using non-pneumatic data from a source other than the sensor; generating a correction to be applied to the preliminary synthetic data using a computer-implemented model having been trained using machine learning and training data associating previous non-pneumatic data with previous corrections; applying the correction to the preliminary synthetic data to generate corrected synthetic data indicative of the aircraft state; determining that the primary pneumatic data is unreliable or unavailable; and after determining that the primary pneumatic data is unreliable or unavailable, performing an action onboard the aircraft based on the corrected synthetic data.
2. The method as defined in claim 1, wherein the action includes controlling a flight control surface of the aircraft.
3. The method as defined in claim 2, comprising, before determining that the primary pneumatic data is unreliable or unavailable, controlling the flight control surface of the aircraft based on the primary pneumatic data.
4. The method as defined in any one of claims 1 to 3, comprising comparing the primary pneumatic data to the corrected synthetic data when determining that the primary pneumatic data is unreliable or unavailable.
5. The method as defined in any one of claims 1 to 4, wherein the action includes communicating the corrected synthetic data to a flight crew of the aircraft.
6. The method as defined in any one of claims 1 to 5, wherein applying the correction to the preliminary synthetic data includes adding a corrective value to a synthetic value of the preliminary synthetic data.
7. The method as defined in any one of claims 1 to 5, wherein applying the correction to the preliminary synthetic data includes multiplying a corrective value with a synthetic value of the preliminary synthetic data.
8. The method as defined in claim 6 or claim 7, comprising, before applying the correction to the preliminary synthetic data: determining that the corrective value exceeds a prescribed value; and adjusting the corrective value to the prescribed value.
9. The method as defined in any one of claims 1 to 8, wherein generating the correction includes: using the preliminary synthetic data to identify an operating point of the aircraft within an operating envelope of the aircraft; and computing the correction based on the operating point within the operating envelope.
10. The method as defined in any one of claims 1 to 9, wherein the aircraft state includes an airspeed of the aircraft.
11. The method as defined in any one of claims 1 to 9, wherein the aircraft state includes a Mach number at which the aircraft is travelling.
12. The method as defined in any one of claims 1 to 9, wherein the aircraft state includes an angle of attack of the aircraft.
13. The method as defined in any one of claims 1 to 9, wherein the aircraft state includes a side slip angle of the aircraft.
14. The method as defined in any one of claims 1 to 9, wherein the aircraft state includes a dynamic pressure of air through which the aircraft is travelling.
15. The method as defined in any one of claims 1 to 9, wherein the aircraft state includes a temperature of air through which the aircraft is travelling.

Description

TECHNICAL FIELD The disclosure relates generally to aircraft, and more particularly to monitoring an aircraft state and to operating the aircraft. BACKGROUND During flight of a modern fly-by-wire aircraft, an air data system of the aircraft generates safety-critical air data parameters that may be used by a flight control system of the aircraft, or that may be relied upon by a flight crew operating the aircraft. A traditional approach for providing the required reliability in such safety-critical system includes having a plurality of redundant measurement channels that work in parallel and that introduce design and manufacturing differences, and independence between the channels. In combination with failure detection and isolation logic, the redundancy provides data availability, and the dissimilarity provides data integrity. Unfortunately, this approach is penalizing in terms of weight, power demand, volume, installation effort, development complexity and cost. This approach also may not necessarily cover all common failure modes so the channels may be prone to similar causes of failure. Improvement is desirable. SUMMARY In one aspect, the disclosure describes a method of operating an aircraft. The method comprises: acquiring primary pneumatic data indicative of an aircraft state using a sensor interacting with an air stream outside of the aircraft during flight of the aircraft;generating preliminary synthetic data indicative of the aircraft state using non-pneumatic data from a source other than the sensor;generating a correction to be applied to the preliminary synthetic data using a computer-implemented model having been trained using machine learning and training data associating previous non-pneumatic data with previous corrections;applying the correction to the preliminary synthetic data to generate corrected synthetic data indicative of the aircraft state;determining that the primary pneumatic data is unreliable or unavailable; andafter determining that the primary pneumatic data is unreliable or unavailable, performing an action onboard the aircraft based on the corrected synthetic data. The action may include controlling a flight control surface of the aircraft. The method may comprise, before determining that the primary pneumatic data is unreliable or unavailable, controlling the flight control surface of the aircraft based on the primary pneumatic data. The method may comprise comparing the primary pneumatic data to the corrected synthetic data when determining that the primary pneumatic data is unreliable or unavailable. The action may include communicating the corrected synthetic data to a flight crew of the aircraft. Applying the correction to the preliminary synthetic data may include adding a corrective value to a synthetic value of the preliminary synthetic data. Applying the correction to the preliminary synthetic data may include multiplying a corrective value with a synthetic value of the preliminary synthetic data. The method may comprise, before applying the correction to the preliminary synthetic data: determining that the corrective value exceeds a prescribed value; and adjusting the corrective value to the prescribed value. Generating the correction may include: using the preliminary synthetic data to identify an operating point of the aircraft within an operating envelope of the aircraft; and computing the correction based on the operating point within the operating envelope. The aircraft state may include an airspeed of the aircraft. The aircraft state may include a Mach number at which the aircraft is travelling. The aircraft state may include an angle of attack of the aircraft. The aircraft state may include a side slip angle of the aircraft. The aircraft state may include a dynamic pressure of air through which the aircraft is travelling. The aircraft state may include a temperature of air through which the aircraft is travelling. The aircraft state may include a pressure altitude of the aircraft. The aircraft state may include a vertical speed of the aircraft. The aircraft state may include an air density outside of the aircraft. The aircraft state may include an aircraft weight. Embodiments may include combinations of the above features. In another aspect, the disclosure describes a method of monitoring an aircraft state. The method comprises: acquiring a primary value indicative of the aircraft state using a first sensor;computing a preliminary synthetic value indicative of the aircraft state using other data acquired via a second sensor having a different operating principle than the first sensor;computing a corrective value to be applied to the preliminary synthetic value using a computer-implemented model having been trained using machine learning and training data associating previous values of the other data with previous corrective values; andapplying the corrective value to the preliminary synthetic value to generate a corrected synthetic value indicative of the air