Air data indicating device and method of calibrating the same

Abstract

This invention relates to an air data indication and calibration device (1) and method for a vertical takeoff and landing (VTOL) aircraft, specifically a helicopter (50), for providing airspeed information and altitude information of the VTOL aircraft. The VTOL aircraft includes: a pitot tube device (51) for determining stagnation pressure at the location of the pitot tube device (51); and a static pressure orifice device (52) for determining static pressure at the location of the static pressure orifice device (52). The air data indication device (1) includes an airspeed and altitude determination module (2) for determining the airspeed and altitude of the VTOL aircraft in real time based on flight data using a regressor (3) obtained by training an artificial neural network through training data.

Description

Technical Field

[0001] This invention relates to an air data indication device for a vertical takeoff and landing (VTOL) aircraft (specifically a helicopter), providing airspeed and altitude information for the VTOL aircraft. The VTOL aircraft includes: a pitot tube device for determining the stagnation pressure at the location of the pitot tube device and providing pitot tube data including the stagnation pressure information at the location of the pitot tube device; and a static pressure orifice device for determining the static pressure at the location of the static pressure orifice device and providing static pressure orifice data including the static pressure information at the location of the static pressure orifice device. Furthermore, this invention also relates to a method for calibrating such an air data indication device. Background Technology

[0002] As is well known, air data indication devices belong to the technical field initially mentioned. They are included in the so-called pitot-barrel system implemented in vertical takeoff and landing (VTOL) aircraft. This pitot-barrel system uses a pitot tube device installed on the VTOL aircraft to measure the stagnation pressure at the location of the pitot tube device, and uses a barometric pressure port device installed on the VTOL aircraft to measure the static pressure at the location of the barometric pressure port device. Using the air data indication device, the airspeed and altitude of each VTOL aircraft are determined using pitot tube data (including stagnation pressure information measured using the pitot tube device at the location of the pitot tube device on the VTOL aircraft) and barometric pressure port data (including static pressure information measured using the barometric pressure port device at the location of the barometric pressure port device on the VTOL aircraft). The airspeed and altitude of the VTOL aircraft are then displayed to the VTOL aircraft pilot, or the data, including the airspeed and altitude information of the VTOL aircraft, is forwarded to the VTOL aircraft's autopilot to enable the VTOL aircraft to fly automatically in autopilot mode.

[0003] These Pitot hydrostatic systems are crucial for the safety of VTOL aircraft. If a pilot is unaware of the correct airspeed and altitude, or even relies on incorrect airspeed and altitude values, they might attempt a fly maneuver, potentially causing the VTOL to crash. Similarly, if data containing incorrect airspeed and / or incorrect altitude information is relayed to the VTOL's autopilot, it could also lead to a crash.

[0004] Therefore, the Pitot hydrostatic system needs to reliably display the airspeed and altitude of the VTOL aircraft to the pilot with an accuracy at least as required by regulatory specifications. To achieve this, the air data indication device needs to reliably determine the airspeed and altitude of the VTOL aircraft with an accuracy at least as required by regulatory specifications and provide accurate information about the airspeed and altitude of the VTOL aircraft.

[0005] However, in vertical takeoff and landing (VTOL) aircraft, the stagnation pressure at the pitot tube location and the static pressure at the pressure orifice location depend not only on the actual airspeed and altitude of the VTOL aircraft, but also on the current flight configuration and flight conditions. More specifically, airflow and turbulence exist around the VTOL aircraft, hence the placement of pitot tubes and pressure orifices. These airflows and turbulence at the pitot tube and pressure orifice locations vary with the VTOL aircraft's flight configuration and flight conditions. Therefore, at a given airspeed and altitude, the stagnation pressure measured by the pitot tube and the static pressure measured by the pressure orifice depend on the current flight configuration and flight conditions. Therefore, to provide airspeed and altitude to the VTOL aircraft pilot with acceptable accuracy, the aircraft's air data indication devices must be calibrated.

[0006] Methods for calibrating air data indication devices are described, for example, in Denis Hamel and Alex Kolarich's *Vertical Flight Society*, 76... th The Annual Forum & Technology Display, Oct 06-08, 2020, featured a paper titled "GPS-BASED Airspeed Calibration for Rotocraft: Generalized Application for All Flight Regimes." This paper addresses the calibration of the Pitot hydrostatic system. Therefore, according to this paper, the GPS-based real airspeed method, employing appropriate execution and analysis techniques, is recognized as the most practical in terms of equipment and efficiency for providing complete airspeed system calibration.

[0007] However, there is no known calibration method that can calibrate the airspeed and altitude values ​​determined by the air indicators of a vertical takeoff and landing aircraft to the accuracy required by regulatory specifications for all flight configurations and flight conditions.

[0008] In addition to the calibration issues mentioned above, the stagnation pressure at the pitot tube device location and the static pressure at the static pressure orifice device location also depend on the positions of the pitot tube device and the static pressure orifice device on the VTOL aircraft. For example, it is generally observed that the closer the pitot tube device and the static pressure orifice device are to the fuselage of the VTOL aircraft, the greater the deviation of the stagnation pressure measured by the pitot tube device and the static pressure measured by the static pressure orifice device from their effective values. However, when designing new VTOL aircraft that include a pitot tube static pressure system, the deviations of the measured stagnation pressure and static pressure from their effective values ​​are so large that calibration cannot achieve the required accuracy. Therefore, it is often necessary to repeatedly reposition the pitot tube device and the static pressure orifice device. This typically leads to multiple design iterations. In each iteration, after flight testing with a prototype of such a VTOL aircraft, the air data indication device of the employed pitot tube system is recalibrated to verify the accuracy of the pitot tube system, until a design of a VTOL aircraft that meets the regulatory requirements for the pitot tube hydrostatic system, and the arrangement of the pitot tube equipment and hydrostatic orifice equipment on the VTOL aircraft, is found. Because these iterations are very time-consuming and costly, it is worthwhile to reduce the number of iterations required. One known method to reduce the number of iterations is to place the pitot tube equipment and hydrostatic orifice equipment in locations outside the critical pressure field, specifically away from the VTOL aircraft fuselage. However, the disadvantages of doing so are high cost and safety concerns for ground personnel. Invention Overview

[0010] The object of this invention is to create an air data indicator device belonging to the initially mentioned technical field and a method for calibrating such an air data indicator device, such that the air data indicator device can provide correct airspeed and altitude values ​​with the accuracy required by regulatory specifications for all flight configurations and flight conditions, and makes the design process for vertical takeoff and landing aircraft cheaper and safer.

[0011] The solution of the present invention is described by the features of claim 1. According to the present invention, the air data indicator can be connected to a pitot tube device for receiving pitot tube data provided by the pitot tube device, and the air data indicator can be connected to a barometric orifice device for receiving barometric orifice data provided by the barometric orifice device. Therefore, the air data indicator includes an airspeed and altitude determination module, which is used to determine the airspeed and altitude of a vertical takeoff and landing (VTOL) aircraft in real time based on flight data using a regressor obtained by training an artificial neural network with training data. The flight data includes at least: pitot tube data, barometric orifice data, vertical speed data including VTOL vertical speed information, and pitch attitude angle data including VTOL pitch attitude angle information.

[0012] According to the present invention, the air data indication device is used for vertical takeoff and landing (VTOL) aircraft, specifically helicopters. Therefore, it is not important whether the VTOL aircraft or helicopter is manned or unmanned. If the VTOL aircraft or helicopter is manned, the pilot can be located on the aircraft or helicopter, or, if the VTOL aircraft or helicopter can be remotely controlled, the pilot can be located on the ground. If the VTOL aircraft or helicopter is unmanned, it is advantageously remotely controlled. In all these cases, the VTOL aircraft or helicopter may also include an autopilot that relies on a Pitot hydrostatic system, etc., including the air data indication device according to the present invention.

[0013] According to the present invention, an air data indication device is used to provide information about the airspeed of a vertical takeoff and landing (VTOL) aircraft and information about the altitude of the VTOL aircraft. Therefore, the air data indication device is advantageously adapted to provide airspeed information of the VTOL aircraft determined by an airspeed and altitude determination module, and is also adapted to provide altitude information of the VTOL aircraft determined by the same module. Thus, in one example, the air data indication device includes a display for displaying the airspeed and altitude of the VTOL aircraft. In this example, the airspeed and altitude information of the VTOL aircraft are provided by the air data indication device by displaying the airspeed and altitude on the display. In another example, the air data indication device includes an output port for outputting output data including the airspeed and altitude information of the VTOL aircraft. In this other example, the output port may be connected to the autopilot of the VTOL aircraft. Alternatively, the output port may be connected to other devices, such as the onboard computer of the VTOL aircraft, for transmitting the output data from the air data indication device to the onboard computer for displaying the airspeed and altitude on a display controlled by the onboard computer. Alternatively, in cases where the VTOL aircraft can be remotely controlled by a remote controller, the output port can be connected to the VTOL aircraft's remote controller to transmit output data from the air data indication device to the remote controller for displaying airspeed and altitude. Therefore, the air data indication device can, for example, be located within the VTOL aircraft and transmit output data to the remote controller, or it can, for example, be located within the remote controller and control and receive pitot tube data and barometric pressure data from the VTOL aircraft. In either variant with the output port, the VTOL aircraft's airspeed and altitude information are both provided by the air data indication device as output data.

[0014] According to the present invention, the air speed and altitude determination device includes an airspeed and altitude determination module adapted to determine the airspeed and altitude of a vertical takeoff and landing (VTOL) aircraft in real time based on flight data using a regressor obtained by training an artificial neural network with training data. In one example, the airspeed and altitude determination module is a computer program product running on a computing unit (e.g., the VTOL aircraft's control computer, or another computing unit separate from the VTOL aircraft's control computer). In another example, the airspeed and altitude determination module is a computing unit adapted to determine the airspeed and altitude of a VTOL aircraft in real time based on flight data using a regressor obtained by training an artificial neural network with training data.

[0015] Regardless of whether the airspeed and altitude determination module is a computer program product or a computing unit, it is adapted to utilize a regressor obtained by training an artificial neural network with training data to determine the airspeed and altitude of the VTOL aircraft in real time based on flight data. Therefore, during the operation of the airspeed and altitude determination module, flight data is advantageously fed into the regressor in real time to determine the airspeed and altitude of the VTOL aircraft based on the flight data. Thus, in one example, the regressor outputs the airspeed and altitude of the VTOL aircraft. In another example, the airspeed and altitude of the VTOL aircraft can be calculated by the airspeed and altitude determination module based on the output of the regressor. In the latter example, the output of the regressor can be, for example, calibrated stagnation pressure and calibrated static pressure, from which the airspeed and altitude determination module can calculate the airspeed and altitude of the VTOL aircraft.

[0016] According to the present invention, the flight data includes at least: pitot tube data, hydrostatic orifice data, vertical velocity data including vertical velocity information of the vertical takeoff and landing aircraft, and pitch attitude angle data including pitch attitude angle information of the vertical takeoff and landing aircraft.

[0017] Since the Pitot tube data includes stagnation pressure information at the location of the Pitot tube device, the Pitot tube data can be, for example: stagnation pressure measured by the Pitot tube device and expressed in any unit, an uncalibrated airspeed calculated based on the measured stagnation pressure and the measured static pressure, or a pre-calibrated airspeed. The pre-calibrated airspeed can, for example, be an uncalibrated airspeed calibrated using known location calibration from a lookup table.

[0018] Since the static pressure orifice data contains information about the static pressure at the orifice location, the static pressure orifice data can be, for example, the static pressure output by the orifice measured by the orifice and expressed in any unit, the uncalibrated height calculated based on the measured static pressure, or the pre-calibrated height. This pre-calibrated height can, for example, be a pre-calibrated height calibrated using known location calibration from a lookup table.

[0019] Since vertical velocity data includes vertical speed information for VTOL aircraft, it can be obtained from static orifice data, for example, by utilizing the change in static orifice data over time. This could be, for instance, the change in static pressure (expressed in arbitrary units) measured and output by the orifice over time, the change in uncalibrated altitude over time calculated based on the measured static pressure, or the change in pre-calibrated altitude over time. In either case, this could be the change within each predefined time unit.

[0020] Since pitch attitude angle data includes pitch attitude angle information for a vertical takeoff and landing (VTOL) aircraft, it can be obtained, for example, from the VTOL aircraft's Air Data Attitude and Heading Reference System (ADAHRS), specifically from a gyroscope flight instrument or a microelectrochemical system (MEMS) gyroscope. To receive pitch attitude angle data, a flight data indication device can advantageously be connected to a pitch attitude angle data providing unit that provides the pitch attitude angle data. This pitch attitude angle data providing unit can, for example, be a gyroscope flight instrument or a microelectrochemical system (MEMS) gyroscope. However, the pitch attitude angle data providing unit can also be a computer that receives pitch attitude angle data from another unit or the aforementioned gyroscope flight instrument or microelectrochemical system (MEMS) gyroscope. Therefore, the pitch attitude angle data providing unit can, for example, be part of the VTOL aircraft's Air Data Attitude and Heading Reference System (ADAHRS).

[0021] In the method for calibrating an air data indicator according to the present invention, the airspeed and altitude determination modules of the air data indicator are calibrated using a regressor obtained by training a neural network with training data, wherein the training data includes training datasets. Therefore, each training dataset is related to flight conditions and includes flight data obtained during flight of a vertical takeoff and landing (VTOL) aircraft, specifically a helicopter, under the respective flight conditions, for which the air data indicator is calibrated. Each training dataset includes reference output data corresponding to the expected output of the regressor during flight of the respective type of VTOL aircraft under its respective flight conditions, wherein the flight data includes at least: pitot tube data, obtained from the pitot tube device of the respective type of VTOL aircraft during flight of the respective type of VTOL aircraft under its respective flight conditions; barometric pressure data, obtained from the barometric pressure device of the respective type of VTOL aircraft during flight of the respective type of VTOL aircraft under its respective flight conditions; vertical speed data, including information on the vertical speed of the respective type of VTOL aircraft during flight of the respective type of VTOL aircraft under its respective flight conditions; and pitch attitude angle data, including information on the pitch attitude angle of the respective type of VTOL aircraft during flight of the respective type of VTOL aircraft under its respective flight conditions.

[0022] Therefore, it is advantageous that for each training dataset, flight data for that specific type of VTOL aircraft has been simultaneously acquired and advantageously recorded. Similarly, it is advantageous that for each training dataset, flight data for that specific type of VTOL aircraft has been acquired and advantageously recorded during the same test flight period of a single maneuver flight under specific flight conditions. This allows for the combination of different types of flight data for each training dataset. These different training datasets can be obtained from the same test flight period or from different test flight periods. In the latter case, it is advantageous to use different training datasets obtained from the same type of VTOL aircraft. However, different training datasets can also be obtained from different test flight periods of different VTOL aircraft of the same type.

[0023] In all these variations, for each training dataset, reference output data corresponding to the expected output of the regressor during flight of the respective type of VTOL aircraft under their respective flight conditions can be obtained or recorded simultaneously, along with the flight data of the respective training dataset. However, the reference output data can also be obtained by computation based on the flight data of the respective training dataset.

[0024] Therefore, for example, during one or more test flights of a specific type of VTOL aircraft, flight data and reference output data can be recorded by measurement. Subsequently, the recorded flight data and reference output data can be used to calibrate the air data indication devices (AVIDs) of each VTOL aircraft, as well as the AVIDs of other types of VTOL aircraft. In another example, flight data is recorded by measurement, and reference output data is calculated based on the flight data. Alternatively, flight data can be recorded by measurement during one or more test flights for approving a specific VTOL aircraft, and subsequently, the recorded flight data and the recorded or calculated reference output data can be used to calibrate the AVIDs of each VTOL aircraft.

[0025] The air data indication device according to the invention and the method for calibrating the air data indication device according to the invention have the following advantages: they can quickly provide calibrated air data indication devices that provide correct airspeed and altitude values ​​for all flight configurations and flight conditions, and have the accuracy required by regulatory specifications. Therefore, the solution according to the invention improves the safety of VTOL aircraft operation by providing the correct airspeed and altitude values ​​to the pilot and autopilot of the VTOL aircraft in real time. Furthermore, the air data indication device according to the invention and the method for calibrating the air data indication device according to the invention enable the design of VTOL aircraft with a cheaper and safer design process, requiring only minor, or even no, rearrangement iterations of the pitot tube and hydrostatic ports. Moreover, compared to using known air data indication devices, the solution according to the invention allows the pitot tube and hydrostatic ports to be positioned closer to the fuselage of the VTOL aircraft, avoiding components protruding too far from the fuselage and reducing the risk of injury to aircraft operators. Furthermore, the solution of the present invention utilizes the existing pitot tube equipment and hydrostatic holes of various vertical takeoff and landing aircraft, and can be easily implemented in existing vertical takeoff and landing aircraft. After obtaining training data through test flights using various vertical takeoff and landing aircraft, and after training artificial neural networks using the training data, the air data indicator can be easily calibrated.

[0026] Preferably, the flight data includes roll attitude angle data, which includes roll attitude angle information of the VTOL aircraft. The advantage is that the air data device can be calibrated more accurately using the same calibration methods. Since the roll attitude angle data includes roll attitude angle information of the VTOL aircraft, the roll attitude angle data can be obtained, for example, from the VTOL aircraft's Air Data Attitude and Heading Reference System (ADAHRS), specifically from a gyroscope flight instrument or a microelectrochemical system (MEMS) gyroscope.

[0027] However, alternatively, the flight data may not include roll attitude angle data. The advantage of this is that the calibration of the air data unit is simpler and requires less computing power.

[0028] Advantageously, the flight data includes sideslip angle data, which includes information on the sideslip angle of the VTOL aircraft. This has the advantage that the air data device can be calibrated more precisely using methods for calibrating the air data device. In a variant that allows for more precise calibration of the air data device, the flight data includes: roll attitude angle data (including information on the roll attitude angle of the VTOL aircraft) and sideslip angle data (including information on the sideslip angle of the VTOL aircraft). Since the sideslip angle data contains information on the sideslip angle of the VTOL aircraft, the sideslip angle data can be obtained, for example, from a dedicated sensor. For example, a dedicated sideslip sensor can be used, which calculates the sideslip angle using the total pressure measured at two additional hydrostatic orifices (one pointing to the right side of the VTOL aircraft and one pointing to the left side).

[0029] However, flight data may not include sideslip angle data. The advantage of this is that the calibration of air data devices is simpler and requires less computing power.

[0030] Preferably, the flight data includes lateral acceleration data, which includes information on the lateral acceleration of the VTOL aircraft, specifically the amount of lateral acceleration. This has the advantage that the air data device can be calibrated more accurately using methods for calibrating air data devices. Since the lateral acceleration data includes information on the lateral acceleration of the VTOL aircraft, the lateral acceleration data can be obtained, for example, from the VTOL aircraft's Air Data Attitude and Heading Reference System (ADAHRS), specifically from one or more accelerometers.

[0031] In the first variant, the air data device can be calibrated more precisely. The flight data includes: roll attitude angle data, which includes information on the roll attitude angle of the VTOL aircraft; and lateral acceleration data, which includes information on the lateral acceleration of the VTOL aircraft, specifically the amount of lateral acceleration. In the second variant, the air data device can also be calibrated more precisely. The flight data includes: lateral acceleration data, which includes information on the lateral acceleration of the VTOL aircraft, specifically the amount of lateral acceleration; and sideslip angle data, which includes information on the sideslip angle of the VTOL aircraft. In the third variant, the air data device can be calibrated more precisely. The flight data includes: roll attitude angle data, which includes information on the roll attitude angle of the VTOL aircraft; sideslip angle data, which includes information on the sideslip angle of the VTOL aircraft; and lateral acceleration data, which includes information on the lateral acceleration of the VTOL aircraft, specifically the amount of lateral acceleration.

[0032] Alternatively, however, the flight data may lack lateral acceleration data. The advantage of this is that the calibration of the air data unit is simpler and requires less computing power.

[0033] Advantageously, the Pitot hydrostatic system includes an air data indicator, a Pitot tube device, and a hydrostatic orifice according to the invention.

[0034] Preferably, the vertical takeoff and landing (VTOL) aircraft (specifically a helicopter) includes the air data indication device according to the invention. The VTOL aircraft advantageously further includes a pitot tube and a barometric vent. In this case, the VTOL aircraft advantageously includes the aforementioned pitot barometric vent system, which includes the air data indication device according to the invention. However, alternatively, the VTOL aircraft may not have a pitot tube or a barometric vent. In any case where the VTOL aircraft includes the air data indication device according to the invention, the VTOL aircraft advantageously further includes a display for displaying the airspeed of the VTOL aircraft based on airspeed information received from the air data indication device, and for displaying the altitude of the VTOL aircraft based on altitude information received from the air data indication device.

[0035] However, in alternatives to these variations, the air data indicator according to the invention is separate from the mentioned Pitot hydrostatic system and from the mentioned vertical takeoff and landing aircraft.

[0036] If the VTOL aircraft is an unmanned aerial vehicle (UAV), then the air data indication device (AFID) is advantageously part of the combination of the VTOL aircraft and the remote controller used to control it. In this case, the AFID can be located either in the VTOL aircraft or in the remote controller. In the latter case, the AFID is separate from the VTOL aircraft because it is located in the remote controller.

[0037] Alternatively, the air data indicator can be manufactured and sold completely independently of the VTOL aircraft and independent of any remote controller. Therefore, the air data indicator can be adapted, for example, to be installed in a VTOL aircraft, or it can be adapted to be installed in a remote controller.

[0038] Preferably, in the method for calibrating the air data indication device, in each training dataset, reference output data corresponding to the expected output of the regressor of that type of VTOL aircraft during flight under its respective flight conditions is obtained from one or more reference sensors of the VTOL aircraft during flight under its respective flight conditions. This has the advantage that the reference output data can be obtained reliably and in an easily understandable manner. The one or more reference sensors are advantageously calibrated sensors arranged on the VTOL aircraft. Thus, the one or more reference sensors can be sensors arranged on the VTOL aircraft, used only to obtain reference output data during flight testing under their respective flight conditions. After the flight testing is completed, the one or more reference sensors can be removed from the VTOL aircraft. However, in one variation, the one or more reference sensors are permanently installed on the VTOL aircraft.

[0039] Whether the reference output data is identical to the data output by one or more reference sensors, or whether the reference output data is calculated based on the data output by one or more reference sensors, is irrelevant. In either case, one or more reference sensors can be positioned, for example, on the nose boom of a VTOL aircraft, or, during flight testing, on a trailing bomb towed by the VTOL aircraft, to obtain the reference output data. Such a nose boom and such a trailing bomb allow one or more reference sensors to be positioned remotely from the fuselage of the VTOL aircraft, enabling them to provide the most accurate data possible to obtain the reference output data. Therefore, the data obtained is particularly accurate when one or more reference sensors are calibration sensors.

[0040] Depending on the desired output of the returner during flight of the vertical takeoff and landing (VTOL) aircraft type under its respective flight conditions, the reference output data can be obtained, for example, from airspeed and pressure altitude sensors that have been calibrated to account for the effects of wind and turbulence, from dynamic pressure and static pressure altitude sensors that have been calibrated to account for the effects of wind and turbulence, from total pressure and static pressure altitude sensors that have been calibrated to account for the effects of wind and turbulence, or from ground speed and rate of climb sensors (such as GPS and inertial units).

[0041] However, alternatively, the reference output data may be obtained by computation based on flight data from various training datasets or by other means. For example, reference output data can be obtained by calibrating flight data using GPS calibration methods, as described in Denis Hamel and Alex Kolarich's *Vertical Flight Society's 76*. th The article “GPS-BASED Airspeed Calibration for Rotocraft: Generalized Application for All Flight Regimes” was presented at the Annual Forum & Technology Display, Oct 06-08, 2020, Virtual.

[0042] Advantageously, for at least one flight condition in the flight condition list, the training data includes at least one training dataset associated with the respective flight condition, wherein the flight condition list includes level flight at a first level flight speed, climb at a first rate of climb and a first climb speed, and descent at a first rate of descent and a first descent speed. Thus, in the variant of vertical climb, the first climb speed is 0 knots. Furthermore, in the variant of vertical descent, the first descent speed is 0 knots. More advantageously, for each of at least two flight conditions in the flight condition list, the training data includes at least one training dataset associated with the respective flight condition. Even more advantageously, for each of at least three flight conditions in the flight condition list, the training data includes at least one training dataset associated with the respective flight condition. Therefore, the more diverse the flight conditions in the flight condition list associated with the training data including at least one training dataset, the more accurately the air data indication device can be calibrated.

[0043] Advantageously, the flight conditions list also includes level flight at a second level flight speed, which is greater than the first level flight speed. Therefore, the training data advantageously includes at least two training datasets, one related to level flight under flight conditions at the first level flight speed, and the other related to level flight under flight conditions at the second level flight speed. This allows for more precise calibration of the air data indicator. This precision can be further improved if the first and second level flight speeds are substantially equidistant (e.g., 100 knots and 200 knots). However, the first and second level flight speeds can also be unequally equidistant.

[0044] Alternatively, the flight conditions list may not include level flight at a second level flight speed, where the second level flight speed is greater than the first level flight speed.

[0045] Advantageously, the flight conditions list also includes climbs at a second rate of climb and a first rate of climb, wherein the second rate of climb is greater than the first rate of climb. Therefore, the training data advantageously includes at least two training datasets, one of which relates to climbs under flight conditions at the first rate of climb and the first rate of climb, and the other of which relates to climbs under flight conditions at the second rate of climb and the first rate of climb. This allows for more accurate calibration of the air data indication device. This accuracy can be further improved if the first and second rates of climb are substantially equidistant (e.g., 500 ft / min and 1,000 ft / min). However, the first and second rates of climb can also be unequally equidistant.

[0046] The flight conditions list advantageously includes climbs at a first rate of climb and a second climb speed, where the second climb speed is greater than the first climb speed. Therefore, the training data advantageously includes at least two training datasets, one of which relates to climbs under flight conditions at the first rate of climb speed and the other to climbs under flight conditions at the first rate of climb speed and the second climb speed. This allows for more accurate calibration of the air data indication device. This accuracy can be further improved if the first and second climb speeds are substantially equidistant (e.g., 50 knots and 100 knots). However, the first and second climb speeds can also be unequally equidistant.

[0047] Advantageously, the flight condition list also includes climbs at a second rate of climb and a second climb speed. This allows for more precise calibration of the air data indication device. This precision can be further improved if the first and second climb speeds are substantially equal in interval (e.g., 50 knots and 100 knots). However, the first and second climb speeds can also be unequal in interval.

[0048] Alternatively, the flight conditions list may not include climbs at a second rate of climb, where the second rate of climb is greater than the first rate of climb.

[0049] The list of flight conditions advantageously includes descents at a second rate of descent and a first descent speed, wherein the second rate of descent is greater than the first rate of descent. Therefore, the training data advantageously includes at least two training datasets, one of which relates to descents at the first rate of descent and the first descent speed, and the other of which relates to descents at the second rate of descent and the first descent speed. This allows for more precise calibration of the air data indicator. This precision can be further improved if the first and second rates of descent are substantially equidistant (e.g., 700 ft / min and 1400 ft / min). However, the first and second rates of descent can also be unequally equidistant.

[0050] The list of flight conditions advantageously includes descents at a third rate of descent and a first descent speed, wherein the third rate of descent is greater than the second rate of descent. Therefore, the training data advantageously includes at least three training datasets, one of which relates to descents at the first rate of descent and the first descent speed, another to descents at the second rate of descent and the first descent speed, and yet another to descents at the third rate of descent and the second descent speed. This has the advantage of allowing for more precise calibration of the air data indicator. This precision can be further improved if the first, second, and third rates of descent are substantially equidistant (e.g., 700 ft / min, 1400 ft / min, and 2100 ft / min). However, the first, second, and third rates of descent can also be unequally equidistant.

[0051] The list of flight conditions advantageously includes descents at a first descent rate and a second descent flight speed, wherein the second descent flight speed is greater than the first descent flight speed. Therefore, the training data advantageously includes at least two training datasets, one of which is correlated with descents at the first descent rate at the first descent flight speed, and the other of which is correlated with descents at the first descent rate at the second descent flight speed. This allows for more accurate calibration of the air data indication device. Accuracy can be further improved if the first and second descent flight speeds are substantially equidistant (e.g., 50 knots and 100 knots). However, the first and second descent flight speeds can also be unequally equidistant.

[0052] The flight condition list advantageously includes a descent at a second descent rate and a second descent speed. This allows for more precise calibration of the air data indication device. Accuracy can be further improved if the first and second descent speeds are substantially equal in interval (e.g., 50 knots and 100 knots). However, the first and second descent speeds can also be unequal in interval.

[0053] The flight condition list advantageously further includes descents at a third descent rate and a second descent speed, wherein the third descent rate is greater than the second descent rate. Therefore, the training data advantageously includes at least three training datasets, one of which relates to descents at the first descent rate and the second descent speed, another to descents at the second descent rate and the second descent speed, and yet another to descents at the third descent rate and the second descent speed. This allows for more accurate calibration of the air data indication device.

[0054] Particularly advantageous is that the training data advantageously includes at least six training datasets, one of which relates to flight condition descent at a first rate of descent and a first descent speed, one of which relates to flight condition descent at a second rate of descent and a first descent speed, one of which relates to flight condition descent at a third rate of descent and a first descent speed, one of which relates to flight condition descent at a first rate of descent and a second descent speed, one of which relates to flight condition descent at a second rate of descent and a second descent speed, and one of which relates to flight condition descent at a third rate of descent and a second descent speed. This allows for more accurate calibration of the air data indication device.

[0055] However, in one variant, the flight conditions list does not include descent at a third descent rate, where the third descent rate is greater than the second descent rate.

[0056] Independent of the descent speed, if the list of flight conditions includes a descent at a second rate of descent or even a third rate of descent, then the flight condition for a descent with the highest rate of descent is advantageously associated with unpowered descent (sometimes also called autorotational descent), while one or more other flight conditions for a descent with a rate of descent are associated with powered descent.

[0057] Alternatively, the flight condition list may not include descent at a second descent rate, where the second descent rate is greater than the first descent rate. Furthermore, at the second descent speed, the flight condition list can be implemented without any descent. Advantageously, the first level flight speed, second level flight speed, first rate of climb, second rate of climb, first climb speed, second climb speed, first descent rate, second descent rate, and third descent rate, first descent speed, and second descent speed described above are constants. However, as described below, one or more of the first level flight speed, second level flight speed, first rate of climb, second rate of climb, first climb speed, second climb speed, first descent rate, second descent rate, third descent rate, first descent speed, and second descent speed can vary over time with the application of acceleration in either direction to simulate one of the corresponding flight configurations described below.

[0058] Preferably, each training dataset is associated with a flight configuration and includes flight data obtained during flight in the respective flight configuration and flight conditions. Each training dataset is associated with the flight conditions of a vertical takeoff and landing (VTOL) aircraft of the respective type, wherein the air data indication device is calibrated for that type of VTOL aircraft. The training data includes at least one training dataset associated with at least one of the following flight configurations: center of gravity, longitudinally fully forward center of gravity, longitudinally fully rearward center of gravity, longitudinally slightly forward center of gravity, and longitudinally slightly rearward center of gravity.

[0059] Therefore, each training dataset is advantageously associated with one flight condition and one flight configuration. Thus, the flight condition is advantageously one of the aforementioned list of flight conditions, and the flight configuration is advantageously one of the aforementioned list of flight configurations.

[0060] Advantageously, for each of at least two flight configurations in the flight configuration list, the training data includes at least one training dataset associated with the respective flight configuration. More advantageously, for each of at least three flight configurations in the flight configuration list, the training data includes at least one training dataset associated with the respective flight configuration. Even more advantageously, for each of at least four flight configurations in the flight configuration list, the training data includes at least one training dataset associated with the respective flight configuration. Most advantageously, for each flight configuration in the flight configuration list, the training data includes at least one training dataset associated with the respective flight configuration.

[0061] In a more advantageous variant, for each of at least two flight configurations in the flight configuration list, the training data includes at least one training dataset related to both the respective flight conditions and the respective flight configuration. More advantageously, for each of at least three flight configurations in the flight configuration list, the training data includes a training dataset related to both the respective flight conditions and the respective configuration. Even more advantageously, for each of at least four flight configurations in the flight configuration list, the training data includes at least one training dataset related to both the respective flight conditions and the respective flight configuration. Most advantageously, for each flight condition of each flight configuration in the flight configuration list, the training data includes at least one training dataset related to both the respective flight conditions and the respective configuration.

[0062] If the flight configuration is center of gravity, then the VTOL aircraft can be physically configured in a center of gravity configuration, or the VTOL aircraft can be physically configured in a different configuration, such as a slightly rearward center of gravity configuration, and the center of gravity configuration can be simulated to fly with constant forward acceleration, such that the pitch angle is equal to that of the center of gravity configuration. This simulation of the center of gravity configuration can be achieved under all flight conditions in the above list of flight conditions.

[0063] If the flight configuration is a longitudinally forward center of gravity configuration, then the VTOL aircraft can be physically configured in a longitudinally forward center of gravity configuration, or the VTOL aircraft can be physically configured in a different configuration, such as an intermediate center of gravity configuration, and simulated to fly with a longitudinally forward center of gravity at a constant forward acceleration, such that the pitch angle is equal to that of the longitudinally forward center of gravity configuration. This simulation of a longitudinally forward center of gravity configuration can be achieved under all flight conditions in the above list of flight conditions.

[0064] If the flight configuration is a longitudinally fully rearward center of gravity configuration, then the VTOL aircraft can be physically configured in a longitudinally fully rearward center of gravity configuration, or the VTOL aircraft can be physically configured in a different configuration, such as an intermediate center of gravity configuration, and simulated to fly with a constant rearward acceleration at a longitudinally fully rearward center of gravity configuration, such that the pitch angle is equal to that of the fully rearward center of gravity configuration. This simulation of a longitudinally fully rearward center of gravity configuration can be achieved under all flight conditions in the above list of flight conditions.

[0065] If the flight configuration is a slightly forward center of gravity in the longitudinal direction, then the VTOL aircraft can be physically configured with a slightly forward center of gravity in the longitudinal direction, or the VTOL aircraft can be physically configured with a different configuration, such as an intermediate center of gravity configuration, and simulated to fly with a slightly forward center of gravity in the longitudinal direction with constant forward acceleration, so that the pitch angle is equal to the slightly forward center of gravity configuration. This simulation of a slightly forward center of gravity configuration in the longitudinal direction can be achieved under all flight conditions in the above list of flight conditions.

[0066] If the flight configuration is a slightly rearward longitudinal center of gravity, then the VTOL aircraft can be physically configured with a slightly rearward longitudinal center of gravity configuration, or the VTOL aircraft can be physically configured with a different configuration, such as an intermediate center of gravity configuration, and simulated to fly with a slightly rearward longitudinal center of gravity at a constant rearward acceleration, so that the pitch angle is equal to the slightly rearward longitudinal center of gravity configuration. This simulation of a slightly rearward longitudinal center of gravity configuration can be achieved under all flight conditions in the above list of flight conditions.

[0067] Simulating one flight configuration from the flight configuration list while simultaneously configuring the VTOL aircraft physically as another flight configuration from the list has the advantage of allowing the collection of training datasets related to both flight configurations in a single test flight. Therefore, it is particularly advantageous to obtain or record training data separately when configuring the VTOL aircraft physically as one flight configuration from the list (e.g., a center-of-gravity configuration) while simultaneously simulating other flight configurations from the list, so as to obtain or record training datasets related to other flight configurations in the list besides the one configured for VTOL aircraft.

[0068] However, as an alternative to these variants, the training data may consist of a training dataset associated with only one flight configuration, specifically one flight configuration from the list of flight configurations.

[0069] Advantageously, the regressor is a neural network regressor, specifically a fully connected neural network regressor, including an output layer.

[0070] Advantageously, the neural network regressor is a fully connected feedforward neural network regressor. The advantage of this is that the neural network regressor is easier to operate because information only propagates forward within the network. However, an alternative is a recurrent neural network regressor.

[0071] Advantageously, the output layer has a hyperbolic tangent (Tanh) transfer function. This is beneficial because the output of the transfer function provides values ​​between -1 and +1, allowing penalties to be imposed on nodes rather than simply preventing them from triggering. This provides a wider output range, resulting in more accurate calibration of the air data indicator. Furthermore, the derivative of the hyperbolic tangent (Tanh) is easy to compute.

[0072] However, the output layer may have a transfer function different from the hyperbolic tangent (Tanh).

[0073] Advantageously, the regressor includes at least two hidden layers. This allows for more accurate calibration of the air data indicator compared to a regressor with only one hidden layer, or even no hidden layers at all.

[0074] However, the regressor may include only one hidden layer or no hidden layer at all.

[0075] Preferably, each of the at least two hidden layers contains at least 32 neurons, more preferably at least 60 neurons, and most preferably at least 100 neurons. With at least two hidden layers, each containing at least 32 neurons, good calibration of the air data indicator can be achieved. However, the more neurons each of the at least two hidden layers contains, the more accurate the calibration of the air data indicator can be. However, the more neurons each of the at least two hidden layers contains, the greater the computational power required to calibrate the air data indicator. Therefore, it is advantageous that each of the at least two hidden layers contains fewer than 300 neurons. Nevertheless, alternatively, each of the at least two hidden layers may contain fewer than 32 neurons, 300 neurons, or even more than 300 neurons.

[0076] Preferably, the first of at least two hidden layers has a ReLU transfer function. This has the advantage of enabling reliable calibration of air data indicators while requiring relatively little computational power.

[0077] Advantageously, each of the at least two hidden layers, except for the last one, has a ReLU transfer function. This allows for reliable calibration of air data indicators while requiring relatively little computational power. This advantage is particularly pronounced when the regressor contains more than two hidden layers.

[0078] Advantageously, each of at least two hidden layers has a ReLU transfer function. This allows for reliable calibration of the air data indicator while requiring relatively little computational power. However, in a variant, the last of the at least two hidden layers has a continuously distinguishable transfer function (specifically a Tanh transfer function). This allows for more precise calibration of the air data indicator.

[0079] In all the variants with ReLU transfer functions mentioned above, a transfer function different from the ReLU transfer function can be used. For example, in the variants with ReLU transfer functions mentioned above, the transfer function is the GeLu transfer function or the SiLu transfer function, instead of the ReLU transfer function.

[0080] Advantageously, in each training dataset, the value for each type of flight data (e.g., pitot tube data, barometric orifice data, etc.) is obtained by filtering a data stream of flight data of the respective type, measured under the respective flight conditions and respective flight configurations (if applicable), for a length of at least 0.5 seconds, particularly advantageously at least 1 second, and more advantageously at least 2 seconds. The advantage of this is that each training dataset contains reliable and representative figures for each type of flight data, representing the flight conditions and flight configurations relevant to each training dataset. In a preferred variant of these variants, in each training dataset, the value for each type of flight data (e.g., pitot tube data, barometric orifice data, etc.) is obtained by filtering a data stream of flight data of the respective type, measured under the respective flight conditions and respective flight configurations (if applicable), for a length of 10 seconds or less.

[0081] Therefore, when filtering the data stream to obtain corresponding values ​​for each type of flight data, in one instance, the filter used is the average value of the data stream for each type of flight data, calculated using the filter length of the data stream. In another instance, the filter used is a low-pass filter applied to the data stream for each type of flight data over the filter length of the data stream, where the output value of the low-pass filter is, for example, 2Hz.

[0082] In one alternative, in each training dataset, the values ​​for each type of flight data (e.g., pitot tube data, barometric orifice data, etc.) are obtained by filtering the data stream of flight data of the respective type, measured under the respective flight conditions and configurations (if applicable), with a length of less than 0.5 seconds or greater than 10 seconds. In another alternative, the training datasets do not include values ​​for each type of flight data obtained by filtering the data stream of the respective type of flight data.

[0083] Advantageously, further training data can be used to validate the calibration. This allows for control over the safety of the calibration. To achieve this, it is advantageous to use approximately 80% to 90% of the training data for calibration, and the remaining approximately 20% to 10% for validation.

[0084] However, alternatively, this method may not require such verification.

[0085] Other advantageous embodiments and combinations of features will be derived from the following detailed description and the claims as a whole. Attached Figure Description

[0086] The accompanying drawings used to explain the implementation scheme are shown below:

[0087] Figure 1 This is a simplified schematic diagram of the Pitot hydrostatic system of a vertical takeoff and landing (VTOL) aircraft (specifically a helicopter), used to indicate the airspeed and altitude of the VTOL aircraft to the pilot. The Pitot hydrostatic system includes an air data indication device according to the present invention.

[0088] Figure 2 This is a simplified schematic side view of the front of a helicopter, exemplified by a vertical takeoff and landing aircraft equipped with an air data indication device according to the present invention.

[0089] Figure 3a , Figure 3b The data is recorded during the two test flights of the helicopter. Training data can be obtained from this data for calibrating the air data indication device of the present invention using the method according to the present invention.

[0090] Figure 4 Data recorded during the third test flight was used to verify the calibration of the air data indication device according to the present invention, and

[0091] Figure 5 These are detailed views of data from three flight maneuvers conducted during the third test flight, to illustrate in more detail the method of the present invention and the operation of the air data indication device of the present invention.

[0092] In the accompanying drawings, the same parts are referred to by the same reference numerals.

[0093] Preferred implementation scheme

[0094] Figure 1A simplified schematic diagram of a pitot tube pressure system 100 for vertical takeoff and landing (VTOL) aircraft (specifically, helicopters) is shown, used to indicate the airspeed and altitude of the VTOL aircraft to the pilot. The pitot tube pressure system 100 includes a pitot tube device 51 for determining the stagnation pressure at the location of the pitot tube device 51 and providing pitot tube data including stagnation pressure information at the location of the pitot tube device 51. Furthermore, the pitot tube pressure system 100 includes a static pressure orifice device 52 for determining the static pressure at the location of the static pressure orifice device 52 and providing static pressure orifice data including static pressure information at the location of the static pressure orifice device 52. Additionally, the pitot tube pressure system 100 includes an air data indicating device 1 for VTOL aircraft (specifically helicopters) of the present invention for providing airspeed information and altitude information of the VTOL aircraft. The Pitot hydrostatic system 100 also includes a display 101 for displaying the airspeed and altitude of the VTOL aircraft, thereby instructing the VTOL aircraft pilot on the airspeed and altitude of the VTOL aircraft.

[0095] When the Pitot tube pressure system 100 is applied to a vertical takeoff and landing (VTOL) aircraft, it is incorporated into the VTOL aircraft's Attitude and Heading Reference System (ADAHRS). Thus, the pitot tube device 51 and the pressure port device 52 of the Pitot tube pressure system 100 are mounted on the VTOL aircraft, thereby belonging to the VTOL aircraft. Furthermore, a display 101 is arranged to display the VTOL aircraft's airspeed and altitude to the pilot. Therefore, if the pilot is to be seated in the VTOL aircraft's cockpit, the display 101 is located in the cockpit and thus belongs to the VTOL aircraft. However, in cases where the VTOL aircraft is remotely controllable, the display 101 can be arranged in the remote controller used to control the VTOL aircraft. In this case, the display belongs to the remote controller.

[0096] When the pitot tube system 100 is applied to a vertical takeoff and landing (VTOL) aircraft, the air data indicator 1 is connected to the pitot tube device 51 to receive pitot tube data provided by the pitot tube device 51, and is connected to the static pressure port device 52 to receive static pressure port data provided by the static pressure port device 52. Furthermore, the air data indicator 1 is connected to the display 101 to provide the VTOL aircraft's airspeed and altitude information to the display 101 for display to the pilot. Therefore, the air data indicator 1 can, for example, be arranged in the VTOL aircraft and thus become part of the VTOL aircraft, or, if the VTOL aircraft is remotely controllable, the air data indicator 1 can be arranged in the remote controller. In either case, if the VTOL aircraft is equipped with an autopilot to enable automatic flight, the air data indicator 1 is connected to the autopilot to provide the autopilot with information about the VTOL aircraft's airspeed and altitude.

[0097] The air data indicator 1 of the present invention can be manufactured and sold separately from the vertical takeoff and landing aircraft and the pitot tube static pressure system 100. However, since the air data indicator 1 can be part of the vertical takeoff and landing aircraft or part of the aforementioned pitot tube static pressure system 100, the air data indicator 1 can be connected to the pitot tube device 51 to receive pitot tube data provided by the pitot tube device 51, and can be connected to the static pressure orifice device 52 to receive static pressure orifice data provided by the static pressure orifice device 52.

[0098] According to the present invention, the air speed and altitude determination module 1 includes an airspeed and altitude determination module 2. The airspeed and altitude determination module 2 is adapted to determine the airspeed and altitude of a vertical takeoff and landing (VTOL) aircraft in real time based on flight data using a regressor 3 obtained by training an artificial neural network with training data. In one example, the airspeed and altitude determination module 2 is a computer program product running on a computing unit (e.g., the VTOL aircraft's control computer, or another computing unit separate from the VTOL aircraft's control computer). In another example, the airspeed and altitude determination module 2 is a computing unit (such as a computer on a VTOL aircraft, or a computer contained in a remote controller) adapted to determine the airspeed and altitude of a VTOL aircraft in real time based on flight data using a regressor 3 obtained by training an artificial neural network with training data.

[0099] Flight data includes at least: pitot tube data, hydrostatic orifice data, vertical speed data including vertical speed information of the VTOL aircraft, and pitch attitude angle data including pitch attitude angle information of the VTOL aircraft.

[0100] Since the pitot tube data includes stagnation pressure information at the location of the pitot tube device 51, the pitot tube data can be, for example, the stagnation pressure measured by the pitot tube device 51 and expressed in arbitrary units, the uncalibrated wind speed calculated based on the measured stagnation pressure and the measured static pressure, or the precalibrated wind speed. The precalibrated airspeed can, for example, be the uncalibrated airspeed calibrated using known location calibration from a lookup table.

[0101] Since the static pressure orifice data includes static pressure information at the location of the static pressure orifice device 52, the static pressure orifice data can be, for example, the static pressure measured by the static pressure orifice device 52 and expressed in arbitrary units, the uncalibrated height calculated based on the measured static pressure, or the pre-calibrated height. The pre-calibrated height can be, for example, the uncalibrated height calibrated using known location calibration in a lookup table.

[0102] Since vertical velocity data includes vertical speed information for VTOL aircraft, it can be obtained from static pressure orifice data, for example, by utilizing the change in static pressure over time. This could be, for instance, the change in static pressure over time (expressed in arbitrary units) measured and output by static pressure orifice device 52, the change in uncalibrated altitude over time calculated based on the measured static pressure, or the change in pre-calibrated altitude over time. In either case, this could be the change within each predefined time unit.

[0103] Since pitch attitude angle data includes pitch attitude angle information of the VTOL aircraft, it can be obtained, for example, from the VTOL aircraft's Air Data Attitude and Heading Reference System (ADAHRS), specifically from a gyroscope flight instrument or a microelectrochemical system (MEMS) gyroscope. To receive pitch attitude angle data, the flight data indication device 1 is advantageously connected to a pitch attitude angle data providing unit that provides the pitch attitude angle data. This pitch attitude angle data providing unit can, for example, be a gyroscope flight instrument or a microelectrochemical system (MEMS) gyroscope. However, the pitch attitude angle data providing unit can also be a computer that receives pitch attitude angle data from another unit or the aforementioned gyroscope flight instrument or microelectrochemical system (MEMS) gyroscope. Therefore, the pitch attitude angle data providing unit can, for example, be part of the VTOL aircraft's Air Data Attitude and Heading Reference System (ADAHRS).

[0104] In one variant, the flight data further includes roll attitude angle data, which includes information on the roll attitude angle of the VTOL aircraft. Since the roll attitude angle data includes information on the roll attitude angle of the VTOL aircraft, it can be obtained, for example, from the VTOL aircraft's Air Data Attitude and Heading Reference System (ADAHRS), specifically from a gyroscope flight instrument or a microelectrochemical system (MEMS) gyroscope. To receive the roll attitude angle data, the flight data indication device is advantageously connected to a roll attitude angle data providing unit that provides the roll attitude angle data. Therefore, the roll attitude angle data providing unit can be the same unit as the aforementioned pitch attitude angle data providing unit, or it can be separate from the aforementioned pitch attitude angle data providing unit. The roll attitude angle data providing unit can, for example, be a gyroscope flight instrument or a microelectrochemical system (MEMS) gyroscope. However, the roll attitude angle data providing unit can also be a computer that receives the roll attitude angle data from another unit or the aforementioned gyroscope flight instrument or microelectrochemical system (MEMS) gyroscope. Therefore, the roll attitude angle data providing unit can, for example, be part of the air data attitude and heading reference system (ADAHRS) of a vertical takeoff and landing aircraft.

[0105] In further variants, the flight data also includes sideslip angle data, which includes information on the sideslip angle of the VTOL aircraft. In variants capable of more precise calibration of the air data unit, the flight data includes: roll attitude angle data (including information on the roll attitude angle of the VTOL aircraft) and sideslip angle data (including information on the sideslip angle of the VTOL aircraft). Because the sideslip angle data contains information on the sideslip angle of the VTOL aircraft, the sideslip angle data can be obtained, for example, from a dedicated sensor. For example, a dedicated sideslip sensor can be used, which calculates the sideslip angle using the total pressure measured at two additional hydrostatic orifices (one pointing to the right side of the VTOL aircraft and one pointing to the left side).

[0106] In another variation, the flight data also includes lateral acceleration data, which includes information about the lateral acceleration of the VTOL aircraft, specifically the amount of lateral acceleration. Since the lateral acceleration data includes information about the lateral acceleration of the VTOL aircraft, it can be obtained, for example, from the VTOL aircraft's Attitude and Heading Reference System (ADAHRS), specifically from one or more accelerometers. To receive the lateral acceleration data, the flight data indication device is advantageously connected to a lateral acceleration data providing unit that provides the lateral acceleration data.

[0107] In a variant capable of more precise calibration of the air data device, the flight data includes: roll attitude angle data, which includes information on the roll attitude angle of the VTOL aircraft, and lateral acceleration data, which includes information on the lateral acceleration of the VTOL aircraft, specifically the amount of lateral acceleration. In a further variant, where the air data device can also be calibrated more precisely, the flight data includes: lateral acceleration data, which includes information on the lateral acceleration of the VTOL aircraft, specifically the amount of lateral acceleration; and sideslip angle data, which includes information on the sideslip angle of the VTOL aircraft. In a third variant, where the air data device can be calibrated more precisely, the flight data includes: roll attitude angle data, which includes information on the roll attitude angle of the VTOL aircraft; sideslip angle data, which includes information on the sideslip angle of the VTOL aircraft; and lateral acceleration data, which includes information on the lateral acceleration of the VTOL aircraft, specifically the amount of lateral acceleration.

[0108] Figure 2 A simplified schematic side view of the front of a helicopter 50, exemplified as a vertical takeoff and landing aircraft having the air data indicator 1 of the present invention, is shown. The helicopter 50 is equipped with a pitot tube static pressure system 100 and therefore includes the air data indicator 1 of the present invention. Thus, a pitot tube device 51 is mounted on the right side of the front of the helicopter 50, while static pressure port devices 52 are mounted on the left and right sides of the front of the helicopter. These static pressure port devices 52 are interconnected. Both the pitot tube device 51 and the static pressure port devices 52 are arranged and positioned in a manner known in the art. The pitot tube device 51 is used to determine the stagnation pressure at the location of the pitot tube device 51 and provides pitot tube data including stagnation pressure information at the location of the pitot tube device 51. The interconnected static pressure port devices 52 are used to determine the static pressure at the location of the static pressure port devices 52 and provide static pressure port data including static pressure information at the location of the static pressure port devices 52. The air data indicator 1 is connected to the pitot tube device 51 to receive the pitot tube data provided by the pitot tube device 51 and is connected to the static pressure port devices 52 to receive static pressure port data provided by the static pressure port devices 52. Furthermore, the air data indicator 1 is adapted to obtain vertical speed data, including helicopter vertical speed information, from the static pressure hole data by determining the change of static pressure hole data over time.

[0109] The helicopter 50 also includes an air data attitude and heading reference system (ADAHRS) 56, which is known in the art. The air data indication device 1 is connected to the ADASHRS 56 and is used to receive pitch attitude angle data determined and provided by the ADASHRS 56.

[0110] In helicopter 50, air data indicator 1 is connected to display 101 of pitot static pressure system 100. Figure 2(Not shown in the image), it is used to provide the airspeed and altitude information of the helicopter 50 to the display 101, so as to display the airspeed and altitude to the pilot of the helicopter 50 through the display 101. In addition, the air data indication device 1 is connected to the autopilot of the helicopter 50 and is used to provide the autopilot with the airspeed and altitude information of the helicopter 50. Therefore, Figure 2 The helicopter 50 shown can be flown by a pilot and an autopilot using an air data indicator 1 calibrated by the method of the present invention.

[0111] In order to obtain reference output data for training data during flight testing, the helicopter 50 was equipped with... Figure 2 The nose boom 53 is shown. A reference pitot tube device 54 and a reference static pressure orifice device 55 are mounted at the front end of the nose boom 53. Because the nose boom 53, the reference pitot tube device 54, and the reference static pressure orifice device 55 are installed in a noncritical pressure field at a distance from the fuselage of the helicopter 50, the effects of airflow and turbulence around the helicopter 50, which could distort the stagnation pressure and static pressure measured using the reference pitot tube device 54 and the reference static pressure orifice device 55, are minimized. Based on the data obtained from the reference pitot tube device 54 and the reference static pressure orifice device 55, a calibrated reference wind speed can be calculated in a manner known in the art, and a calibrated reference static pressure can be calculated based on the reference static pressure orifice device 55, in a manner known in the art. Therefore, both the calibrated reference wind speed and the calibrated reference static pressure are calibrated in a manner known in the art. Because the reference pitot tube device 54 and the reference barometric jack device 55 are positioned at the top of the nose boom 53, and because they are calibrated in a manner known in the art, they are known to provide values ​​well within the range of accuracy required by regulatory specifications for all flight configurations and flight conditions. In addition to positioning the reference pitot tube device 54 and the reference barometric jack device 55 at the top of the nose boom 53, they can also be mounted in towed missiles to obtain reference output data. This would also provide reference output data with the accuracy required by regulatory specifications.

[0112] Through two test flights of helicopter 50, training data has been obtained for training the neural network to obtain regressor 3, thereby calibrating air data indication device 1. The data recorded during these two test flights, such as... Figure 3a and 3bAs shown. The flight data for training includes: pitot tube data, barometric orifice data, vertical velocity data including vertical velocity information of helicopter 50, and pitch attitude angle data including pitch attitude angle information of helicopter 50. This flight data has been obtained from pitot tube device 51, barometric orifice device 52, and ADASHRS 56 as previously described. In the embodiment shown here, the flight data does not include roll attitude angle data, sideslip data, and lateral acceleration data. When one, two, or all three of the roll attitude angle data, sideslip data, and lateral acceleration data are added to the flight data, the calibration of air indicator 1 is more accurate than that shown in this example.

[0113] Figure 3a The data shown is from the first test flight of the helicopter 50 during its two test flights. Figure 3b The data shown is from the second test flight of the helicopter 50, one of two test flights conducted. Training data was obtained from this data. Figure 3a and 3b The six charts show the data recorded during their respective test flights, which are related to the time (in seconds) of each test flight.

[0114] exist Figure 3a , Figure 3b The top chart shows the uncalibrated airspeed (in knots) calculated using Pitot tube data and static orifice data obtained from Pitot tube device 51 and static orifice device 52, using the following formula:

[0115]

[0116] Where IAS is the uncalibrated airspeed shown in the figure, and p SSL Standard sea level pressure p SSL =101325Pa, ρ SSL The air density ρ at standard sea level SSL =1.225kg / m 3 γ is the specific heat capacity ratio of air, γ = 1.4, p t_e For the uncalibrated stagnation pressure obtained from the Pitot tube device 51, p s_e The uncalibrated static pressure is obtained from the static pressure orifice device 52.

[0117] exist Figure 3a , Figure 3b The second graph shows the vertical velocity (in feet per minute) calculated from the uncalibrated static pressure. Furthermore, in these two graphs, the third graph shows pitch attitude angle data obtained from ADASHRS56, while the fourth graph shows the uncalibrated altitude (in feet) calculated from the static pressure obtained from the static pressure orifice device 52. Therefore, the uncalibrated altitude can be calculated from the static pressure p obtained from the static pressure orifice device 52.s_e Calculate using the following formula:

[0118]

[0119] Where p SSL Still at standard sea level pressure p SSL =101325 Pa, and a is the standard sea-level temperature lapse rate a = 0.001982 K / ft, g c = 32.17 Ibm / slug, where g is the acceleration due to gravity, g = 31.174049 ft / sec 2 And R = 96.0340 (ft·lbf) / (lbm·K).

[0120] exist Figure 3a , Figure 3b The bottom two charts show the calibrated reference airspeed (in knots) and calibrated reference altitude (in feet) obtained from the reference pitot tube device 54 and reference hydrostatic orifice device 55 on the nose boom 53 of the helicopter 50, respectively. Thus, the calibrated reference airspeed and calibrated reference altitude are verified using known GPS calibration methods. In such GPS calibration methods, ideally, only GPS data calibrated for air density recorded during windless flight is used as calibration data along with the vertical speed. However, in this case, for windy conditions, the GPS quasi-static nose-and-tail wind method (quasi-static GPS) has been applied to eliminate the influence of wind. This method is described, for example, in the originally mentioned publication, in Vertical Flight Society's 76 th The Annual Forum & Technology Display, Oct 06-08, 2020, featured a presentation titled "GPS-BASED Airspeed Calibration for Rotocraft: Generalized Application for All Flight Regimes".

[0121] In both test flights, the Helicopter 50 was physically positioned in a neutral center-of-gravity configuration. During the test flights, the aircraft performed various maneuvers, such as level flight at different constant airspeeds, climb at different constant rates of climb, and descent at different rates of descent until a powered descent during autorotation. Furthermore, these maneuvers were performed at different forward and rearward accelerations to simulate different flight configurations with different centers of gravity, up to and including a fully longitudinally forward and fully longitudinally rearward center-of-gravity configuration. This completed the entire flight condition and configuration. It is important to note that at least two equally spaced rates of climb and at least three equally spaced rates of descent were covered, with the fastest descent being an autorotational descent and the slower rates of descent being powered descent.

[0122] from Figure 3a and Figure 3b The data shown comprises 6400 training datasets automatically extracted. For each training dataset, flight data and corresponding reference output data were extracted by averaging data streams longer than 2 seconds for each data type. Therefore, in each training dataset, each type of flight data (i.e., pitot tube data, barometric orifice data, vertical velocity data, and pitch attitude angle data), and each type of reference output data (i.e., reference airspeed and reference static pressure), is a single value obtained by averaging and filtering a 2-second data stream of each type of flight data measured under their respective flight conditions and configurations. Consequently, the 2-second time windows of adjacent training datasets overlap by 50%. Besides this example, the time window for averaging and filtering the data streams of each data type can be chosen to be longer than 2 seconds. For example, the time window can be chosen to be 0.5 seconds, 5 seconds, or 10 seconds. Furthermore, an alternative filtering method (e.g., low-pass filtering) can be used instead of averaging to filter the data streams of each data type.

[0123] The regressor 3 is trained using the training data to calibrate the airspeed and altitude determination module 2, thereby calibrating the air data indication device 1. The regressor 3 is a fully connected neural network regressor, comprising two hidden layers and one output layer. The transfer functions of the two hidden layers are ReLU transfer functions, while the transfer function of the output layer is a hyperbolic tangent (Tanh) function. Each of the two hidden layers contains 256 neurons. The regressor 3 used to calculate the calibrated airspeed (CAS) can be expressed by the following formula:

[0124] CAS=Tanh(C3·Relu(C2·Relu(C1·X+I1)+I2)+I3),

[0125] Where C1, C2, and C3 are the weight matrices of the first hidden layer, the second hidden layer, and the output layer, respectively, and I1, I2, and I3 are the constant vectors of the first hidden layer, the second hidden layer, and the output layer, respectively.

[0126] The air data indicator 1 was calibrated by training the regressor 3 with training data. The helicopter 50 underwent a third test flight to verify the calibration of the air data indicator 1. This brought the helicopter 50 back to a physically neutral center-of-gravity configuration. During the test flight, the aircraft performed various maneuvers, such as level flight at different constant airspeeds, climb at different constant rates of climb, and descent at different rates of descent until a powerless descent during autorotation. Furthermore, these maneuvers were performed at different forward and backward accelerations to simulate different flight configurations with different centers of gravity, up to and including a fully longitudinally forward and fully longitudinally backward center-of-gravity configurations. Thus, all flight conditions and configurations were finalized.

[0127] The data recorded during this third test flight, such as Figure 4 As shown. Therefore, Figure 4 Six charts are shown illustrating the relationship between data recorded during the third test flight and the time (in seconds) of the third test flight, with their configurations and... Figure 3a and Figure 3b The configuration shown in the first two test flights was similar. The top chart shows the uncalibrated airspeed (in knots) calculated based on pitot tube data and pressure port data obtained from pitot tube device 51 and pressure port device 52, calculated in the same way as... Figure 3a and Figure 3b The method shown is the same. Furthermore, the second graph shows the vertical velocity (in feet per minute) calculated from uncalibrated hydrostatic pressure, the third graph shows pitch attitude angle data obtained from ADASHRS56, and the fourth graph shows the uncalibrated altitude (in feet) calculated from hydrostatic pressure obtained from hydrostatic orifice device 52, the calculation method being the same as... Figure 3a and Figure 3b The method shown is the same.

[0128] and Figure 3a , Figure 3b on the contrary, Figure 4The bottom two charts show comparisons of airspeeds (in knots) determined in different ways and altitudes (in feet) determined in different ways. The dashed lines represent calibrated reference airspeeds and calibrated reference altitudes (in feet) obtained from the reference pitot tube device 54 and reference barometric gaiter device 55 on the nose boom 53 of the helicopter 50, as known in the art. Therefore, these dashed lines correspond to the reference output data from the third test flight. It is evident that the values ​​of these dashed lines fully comply with the accuracy required by regulatory specifications. The dotted lines represent airspeeds calibrated using classical GPS-based calibration methods known in the art, and altitudes calibrated using classical GPS-based calibration methods known in the art, both calculated from pitot tube data from the pitot tube device 51 and barometric gaiter data from the barometric gaiter device 52, and calibrated using a lookup table. Therefore, calibration was performed using a GPS-based calibration method described in the originally mentioned publication, Denis Hamel and Alex Kolarich in Vertical Flight Society's 76 th The Annual Forum & Technology Display, Oct 06-08, 2020, Virtual presented the article "GPS-BASED Airspeed Calibration for Rotocraft: Generalized Application for All Flight Regimes." Therefore, dotted lines were used for comparison with known classical calibration methods. Solid lines represent the airspeed determined using the calibrated air data indicator 1 and thus by means of a trained regressor 3, as well as the altitude determined using the calibrated air data indicator 1. Thus, the solid lines were determined as described above using flight data including pitot tube data from pitot tube device 51, hydrostatic orifice data from barometric orifice device 52, vertical velocity data, and pitch attitude angle data, with the aid of the trained regressor 3.

[0129] from Figure 4As can be seen from the bottom two charts, the solid line (from the calibrated air data indicator 1) and the dashed line (reference output data) correspond well, while the dotted line (from classical calibration) deviates the most from the other two lines. The airspeed residual is usually close to zero, but in the worst case of extreme flight conditions and rotor configurations, the airspeed residual can reach ±2 knots. This demonstrates that the method of the present invention for calibrating the air data indicator 1 is very effective. Specifically, the entire flight conditions and flight configurations have been covered since the third test flight. After the regressor 3 was determined as described above, the air data indicator 1, based on the pitot tube data from the pitot tube device 51 and the static pressure orifice data from the static pressure orifice device 52, can operate without the nose boom 53 to obtain the airspeed and altitude of the helicopter 50, so the nose boom 53 can be removed from the helicopter 50. Furthermore, the regressor 3 can also be used in the air data indicator of other helicopters of the same type as the helicopter 50 equipped with the pitot tube device 51 and the static pressure orifice device 52.

[0130] To illustrate the method of the present invention and the working effect of the air data indicator 1 in more detail, Figure 5 Detailed views of data from three maneuvers performed during the third test flight are shown. These charts illustrate... Figure 4 The enlarged and cropped portion of the data shown. Figure 5 The six charts on the left illustrate the dynamic descent process under a fully forward-center-of-gravity configuration, while Figure 5 The six charts in the middle show data on the rotational descent process under a fully forward-center-of-gravity configuration. Furthermore, Figure 5 The six charts on the right show the climb first in a fully forward center of gravity configuration and then in a fully rearward center of gravity configuration.

[0131] In summary, it is noteworthy that the present invention provides an air data indicator device belonging to the initially mentioned technical field and a method for calibrating such an air data indicator device, enabling the air data indicator device to provide correct airspeed and altitude values ​​with the accuracy required by regulatory specifications for all flight configurations and flight conditions, and making the design process for designing vertical takeoff and landing aircraft cheaper and safer.

Claims

1. An air data indication device (1) for a vertical takeoff and landing (VTOL) aircraft, for providing airspeed information of the VTOL aircraft and for providing altitude information of the VTOL aircraft, wherein the VTOL aircraft comprises: - Pitot tube device (51), used to determine the stagnation pressure at the location of the pitot tube device (51) and provide pitot tube data including information on the stagnation pressure at the location of the pitot tube device (51), and - Static pressure orifice device (52), used to determine the static pressure at the location of the static pressure orifice device (52) and to provide static pressure orifice data including information on the static pressure at the location of the static pressure orifice device (52). The air data indicator (1) is connected to the pitot tube device (51) for receiving the pitot tube data provided by the pitot tube device (51), and the air data indicator (1) is connected to the static pressure orifice device (52) for receiving the static pressure orifice data provided by the static pressure orifice device (52). The air data indication device (1) includes an airspeed and altitude determination module (2), which is adapted to determine real-time airspeed and altitude based on flight data using a regressor (3) obtained by training an artificial neural network through training data. -The airspeed of the vertical takeoff and landing aircraft, and -The altitude of the vertical takeoff and landing aircraft. The flight data includes at least: a) The Pitot data mentioned above b) The static pressure orifice data. c) Vertical speed data, including the vertical speed information of the vertical takeoff and landing aircraft, and d) Pitch attitude angle data, which includes pitch attitude angle information of the vertical takeoff and landing aircraft.

2. The air data indicating device (1) according to claim 1, characterized in that, The vertical takeoff and landing aircraft is a helicopter (50).

3. The air data indicating device (1) according to claim 1 or 2, characterized in that, The flight data includes roll attitude angle data, which includes information on the roll attitude angle of the vertical takeoff and landing aircraft.

4. The air data indicating device (1) according to claim 1, characterized in that, The flight data includes sideslip angle data, which includes information about the sideslip angle of the vertical takeoff and landing aircraft.

5. The air data indicating device (1) according to claim 1, characterized in that, The flight data includes lateral acceleration data, which includes information on the lateral acceleration of the vertical takeoff and landing aircraft.

6. The air data indicating device (1) according to claim 5, characterized in that, The flight data includes lateral acceleration data, which includes information on the amount of lateral acceleration of the vertical takeoff and landing aircraft.

7. A vertical takeoff and landing aircraft comprising an air data indication device (1) according to any one of claims 1-6.

8. The vertical takeoff and landing aircraft according to claim 7, characterized in that, The vertical takeoff and landing aircraft is a helicopter (50).

9. A method for calibrating the air data indicator (1) according to any one of claims 1-6, characterized in that, The airspeed and altitude determination module (2) of the air data indication device (1) is calibrated by obtaining a regressor (3) through training an artificial neural network using training data, which includes a training dataset. Each training dataset is associated with flight conditions and includes flight data obtained by a vertical takeoff and landing (VTOL) aircraft of a certain type during flight under its respective flight conditions, for which the air data indicator (1) is calibrated. Each training dataset includes reference output data corresponding to the expected output of the regressor (3) during flight of the vertical takeoff and landing aircraft of the type described under their respective flight conditions, wherein the flight data includes at least: a) Pitot tube data, obtained from the pitot tube device (51) of the vertical takeoff and landing aircraft of the type during flight under the respective flight conditions. b) Static pressure orifice data, obtained during flight under the respective flight conditions of the vertical takeoff and landing aircraft of the type, from the static pressure orifice device (52) of the vertical takeoff and landing aircraft of the type. c) Vertical speed data, including information on the vertical speed of the vertical takeoff and landing aircraft of the aforementioned type during flight under the respective flight conditions of the vertical takeoff and landing aircraft of the aforementioned type, and d) Pitch attitude angle data, including information on the pitch attitude angles of the vertical takeoff and landing aircraft of the type during flight under the respective flight conditions of the vertical takeoff and landing aircraft of the type.

10. The method according to claim 9, characterized in that, In each training dataset, the reference output data corresponding to the expected output of the regressor (3) during flight of the vertical take-off and landing aircraft of the type under its respective flight conditions is obtained from one or more reference sensors (54, 55) of the vertical take-off and landing aircraft of the type under its respective flight conditions.

11. The method according to claim 9 or 10, characterized in that, The training data includes at least one training dataset associated with each flight condition for at least one flight condition in the flight condition list, wherein the flight condition list includes: a) Level flight at the first level flight speed b) Climbing at a first rate of climb and a first climb speed, and c) Descent at a first descent rate and a first descent flight speed.

12. The method according to claim 9, characterized in that, Each training dataset is associated with a flight configuration and includes flight data obtained during flight under the respective flight configuration and flight conditions, wherein each training dataset for the flight conditions is associated with the vertical takeoff and landing aircraft of the type, wherein the air data indicator (1) is calibrated for the type of vertical takeoff and landing aircraft, and wherein the training data includes at least one training dataset associated with at least one of the following flight configurations: a) Center of gravity b) The center of gravity is completely forward in the longitudinal direction. c) The center of gravity is completely rearward in the longitudinal direction. d) The center of gravity is slightly forward longitudinally. e) The center of gravity is slightly rearward in the longitudinal direction.

13. The method according to claim 9, characterized in that, The regressor (3) is a neural network regressor, which includes an output layer.

14. The method according to claim 13, characterized in that, The regressor (3) is a fully connected neural network regressor, which includes an output layer.

15. The method according to claim 14, characterized in that, The output layer has a hyperbolic tangent (Tanh) transfer function.

16. The method according to claim 14 or 15, characterized in that, The regressor includes at least two hidden layers.

17. The method according to claim 16, characterized in that, Each of the at least two hidden layers contains at least 32 neurons.

18. The method according to claim 16, characterized in that, The first of the at least two hidden layers has a ReLU transfer function.

19. The method according to claim 17, characterized in that, The first of the at least two hidden layers has a ReLU transfer function.

20. The method according to claim 9, characterized in that, In each training dataset, the data for each type of flight data is obtained by filtering a data stream of at least 0.5 seconds in length, which is measured under the respective flight conditions.

Images