Method for altitude management of an aircraft equipped with an airborne radar

Description

[0001] The invention relates to a method for managing the altitude of an aircraft equipped with an airborne radar, such as an airplane, a helicopter, a drone, an airship, etc.

[0002] The present invention relates to a tactical decision support tool for operational personnel. It is a method for determining and controlling an improved target altitude for detecting targets, for example maritime or aerial, depending on the propagation conditions of the airborne radar waves.

[0003] The objective of the present invention is to have a rapid method for calculating a set altitude for using an airborne radar to satisfy the detection of targets of interest depending on the environment and the propagation conditions of the radar waves, and to have a step of controlling this set altitude by measuring the performance of the radar and correlating it with external data.

[0004] This problem is of interest to operational staff or operators, because, to date, it is an operation that they carry out manually and that costs them a lot of time and fuel, which implies a reduction in the duration of the operational mission.

[0005] The PREDEM software (disclosed for example in the document “A Prefiguration of Future Tactical Aids” by Y. Hurtaud, J. Claverie, M. Aïdonidis and E. Mandine, Dossier Observation des Côtes et des Océans: Sensors and Systems (OCOSS), 2008) developed by the company CS GROUP, and more precisely the “Surface Search” mode seems to address the problem.

[0006] Indeed, the "Surface Search" mode consists of advising the pilot of an aircraft on the best flight altitude to detect a target on the surface. To do this, the PREDEM software executes a certain number of times the propagation code of an advanced propagation model (APM) for "Advanced Propagation Model" in English, by varying the altitude of the transmitter (maximum of 20 altitudes). The inclination of the aircraft antenna as a function of the flight altitude can be taken into account at the level of the sensor database. The output screen is presented in the form of a set of horizontal bar graphs representing the chosen parameter (including the probability of detection) for each flight height.

[0007] However, this method has at least two drawbacks.

[0008] First, environmental propagation models are based solely on weather forecasts that are never compared with real environmental data.

[0009] Additionally, there is no control over the calculated altitude.

[0010] Also, if the environmental condition models are not valid ("atypical" propagation conditions), the estimated altitude is erroneous and the radar sensor operator is not informed.

[0011] One aim of the invention is to overcome the problems mentioned above.

[0012] According to one aspect of the invention, there is provided a method for managing the altitude of an aircraft equipped with an airborne radar comprising: a step of calculating in flight a set altitude at which the aircraft is positioned, comprising a first sub-step implementing an advanced propagation model APM using as input measurements of parameters representative of the surrounding atmosphere (measured meteorological data) and delivering as output an amplification / attenuation coefficient of the radar signals used as input to a second sub-step of calculating a set altitude from an estimate of the radar detection performance using a radar model; and a step of controlling the set altitude at which the aircraft is positioned using a comparison between an estimate of the radar detection performance using a radar model and a measurement of the radar detection performance.

[0013] In one embodiment, the method comprises a step of predetermined calculation of a recommended altitude comprising a first sub-step implementing the advanced propagation model APM using as input predetermined data of parameters representative of the surrounding atmosphere (predetermined meteorological data) and delivering as output an amplification / attenuation coefficient of the radar signals used as input to a second sub-step of calculation of a recommended altitude from an estimate of the radar detection performance and the radar model, said recommended altitude being provided as input to the step of in-flight calculation of a set altitude to limit the in-flight maneuvers necessary for the step of in-flight calculation of a set altitude around this recommended altitude to calculate the set altitude.

[0014] According to one mode of implementation, the predetermined data of parameters representative of the surrounding atmosphere are updated in flight of the aircraft.

[0015] In one implementation, radar performance includes a probability of detecting a target.

[0016] According to one method of implementation, said parameters representative of the surrounding atmosphere include temperature, pressure, humidity, wind speed, and wave height.

[0017] In one embodiment, the comparison between an estimate of the radar detection performance and a measurement of the radar detection performance of the set altitude control step uses a first neural network developing an attenuation / amplification coefficient of the radar signals from predetermined measurements and data of parameters representative of the surrounding atmosphere.

[0018] According to one embodiment, the comparison between an estimate of the radar detection performance and a measurement of the radar detection performance of the set altitude control step uses a second neural network developing a reflection coefficient and an impulsiveness of the surrounding sea clutter from predetermined data measurements of parameters representative of the surrounding atmosphere.

[0019] In one implementation mode, the estimation of the radar detection performance of the set altitude control step comprises an automatic estimation of the radar equivalent surface of the target from a location of a target detected by the radar and self-declared information of the nature and position of the target.

[0020] The self-reporting information of the nature and position of the target may be position and movement information of the surrounding automatic identification system (AIS) for maritime targets, and may be position and movement information of the automatic dependent surveillance system in broadcast mode (ADS-B) for air targets.

[0021] According to one embodiment, the estimation of the radar detection performance of the set altitude control step comprises a manual estimation by an operator of the radar equivalent surface (SER) of the target from attributes of the target including the length, shape, number and position of the superstructures of the target from the radar measurements.

[0022] In one embodiment, the estimation of the radar detection performance of the set altitude control step comprises a correlation of the automatic estimation of the radar equivalent surface and the manual estimation by an operator of the radar equivalent surface.

[0023] The invention will be better understood by studying a few embodiments described as non-limiting examples and illustrated by the appended drawings in which: There figure 1 schematically illustrates a method for managing the altitude of an aircraft equipped with an airborne radar, according to one aspect of the invention; The figure 2 schematically illustrates a predetermined calculation step of a recommended altitude of the management process of the figure 1 , according to one aspect of the invention; The figure 3 schematically illustrates a step of in-flight calculation of a set altitude of the management process of the figure 1 , according to one aspect of the invention; The figure 4 schematically illustrates a step of controlling the set altitude, a step of controlling the set altitude, according to another aspect of the invention; The figure 5 schematically illustrates a first neural network and a second neural network of the set altitude control step, according to other aspects of the invention; and The figure 6 schematically illustrates a step of controlling the set altitude a step of controlling the set altitude comprising the first neural network and / or the second neural network, according to another aspect of the invention.

[0024] Throughout the figures, elements with identical references are similar.

[0025] In the following description the examples described relate to maritime targets using position and movement information from the automatic identification system for surrounding vessels AIS, but can, alternatively, also be carried out for aerial targets, using position and movement information from the automatic dependent surveillance broadcast (ADS-B) system.

[0026] There figure 1 schematically represents a method for managing the altitude of an aircraft equipped with an airborne radar, according to one aspect of the invention.

[0027] The method comprises an optional step 1 of predetermined calculation of a recommended altitude, a step 2 of in-flight calculation of a target altitude, and a step 3 of checking the target altitude.

[0028] The optional step 1 of predetermined calculation of a recommended altitude can be performed in mission preparation.

[0029] There figure 2 schematically illustrates the optional step 1 of predetermined calculation of a recommended altitude of the management process of the figure 1 , according to one aspect of the invention.

[0030] The optional step 1 of predetermined calculation of a recommended altitude includes a first sub-step 4 implementing an advanced APM propagation model, for example as described in the solution developed in the document “A Prefiguration of Future Tactical Aids” cited previously, and based on knowledge of the forecast atmosphere and knowledge of the characteristics of the radar sensor and the carrier.

[0031] The advanced propagation model APM uses as input predetermined data of parameters representative of the surrounding atmosphere (meteorological data) and delivering as output a coefficient of amplification / attenuation of the radar signals. This coefficient of amplification / attenuation of the radar signals is used as input to a second sub-step 5 of calculation of a recommended altitude from an estimate of the radar detection performance and the radar model, the recommended altitude being provided as input to the step of in-flight calculation of a set altitude to limit the in-flight maneuvers necessary for step 2 of in-flight calculation of a set altitude around this recommended altitude to calculate the set altitude.

[0032] The second sub-step 5 of calculating a recommended altitude makes an estimate, for example from the detection performance estimates based on a statistical model of radar clutter and the output of the APM models for propagation losses in the atmosphere.

[0033] There figure 3 schematically illustrates step 2 of in-flight calculation of a set altitude of the management process of the figure 1 , according to one aspect of the invention.

[0034] Step 2 of in-flight calculation of a set altitude comprises a first sub-step 6 implementing an advanced propagation model APM using as input measurements of parameters representative of the surrounding atmosphere (measurements of meteorological data) and delivering as output an amplification / attenuation coefficient of the radar signals used as input to a second sub-step 7 of calculation of a set altitude from an estimate of the radar detection performance using a model of the radar.

[0035] For example, predetermined data of parameters representative of the surrounding atmosphere can be updated in flight of the aircraft, by a communication link from the aircraft to the ground.

[0036] Calculation 2 of the recommended altitude based on the information available in flight is carried out in operational conditions (in flight) and consists of integrating a prediction method 7 of the recommended altitude for using the radar based on the propagation conditions 6 (APM) into the on-board processing.

[0037] Prediction 7 can be autonomous, i.e. knowledge of the atmosphere for calculating the environment can be based solely on information from sensors on board the aircraft (temperature, pressure, hygrometry, etc.).

[0038] Prediction 7 may also use a recommended altitude provided by calculation step 1, on a recommended altitude calculated before the mission, when the aircraft is on the ground, making it possible to limit the in-flight maneuvers necessary for the in-flight calculation step of a set altitude around this recommended altitude to calculate the set altitude.

[0039] Alternatively, the prediction 7 can be connected, i.e. the knowledge of the atmosphere for calculating the environment can be based on information from the sensors on board the platform and on a recommended altitude calculated 1 during the mission, when the aircraft is in flight, the weather forecasts being updated in the event that the aircraft can receive this information.

[0040] The aircraft moves to the set altitude so that step 3 of the set altitude check can be carried out.

[0041] There figure 4 schematically represents step 3 of controlling the set altitude at which the aircraft has reached.

[0042] The airborne radar performs radar data acquisitions at the set altitude at which the aircraft is positioned.

[0043] A conventional processing 9 of the acquired radar data is carried out, and outputs observed detection performances or probabilities of detection of the maritime target at a given distance. Processing 9 also outputs a radar detection or location of a detected maritime target.

[0044] Radar detection is used for an automatic estimation 11 of the radar equivalent surface SER, depending on the size of the targets (if L < 10 m, SER = 10 m 2< , if L ∈ [10, 100], SER = 100 m 2< ), of the maritime target from an association 10 of the radar detection or location of a maritime target detected by the radar and the position and movement information of the automatic identification system of surrounding vessels AIS.

[0045] Radar detection may also, alternatively or in combination, be used for manual estimation 13 by an operator of the radar equivalent surface SER, for example by expertise of the radar operator, of the maritime target from attributes of the maritime target including the length, the course, the number and the position of the superstructures of the maritime target obtained 12 from the radar measurements.

[0046] When the method performs both the automatic estimation 11 of the radar equivalent surface SER and the manual estimation 13 by an operator of the radar equivalent surface SER, a correlation is performed, such as a weighted average, to make a resulting estimated radar equivalent surface SER.

[0047] Regarding the prediction of the SER (Equivalent Surface Area of ​​targets based on cooperative information (AIS 10 and / or observations 12), the performance of the processing is based on the reliability of the information used.

[0048] In this context, it is essential to know the type of vessel corresponding to the radar detection. This information can be accessed via AIS data transmitted by civilian vessels or through user experience based on what they see on their visualization tool.

[0049] AIS is a communications system based on the periodic sending of VHF (Very High Frequency) messages by ships. The SOLAS (Safety Of Life At Sea) convention requires that any ship meeting certain regulatory requirements be equipped with such a system.

[0050] There are two types of information transmitted by the AIS system: dynamic information, which is transmitted every 2 to 10 seconds, and static information, which is transmitted every 6 minutes. Static information includes the vessel's name, dimensions, and destination.

[0051] An association (geometric for example) is made between radar detections and AIS data. For this, windows are defined around each radar detection and according to a common time base, we associate, for each radar detection, the AIS track (if there is one) which is in the defined window.

[0052] The output data is the same information as a radar detection enriched with AIS information.

[0053] From the vessel's dimension information, it is possible to deduce a SER class.

[0054] Automatic estimation of the SER from AIS information or attributes is done, for example, from the length of the target: if the length is < 5 m → SER = 1 m 2< if the length is between 5 and 10 m → SER = 10 m 2< ...

[0055] Regarding observations, the user can obtain information (length, etc.) on a detection (via a distance profile or an ISAR image) and thus deduce, by experience, a class of SER.

[0056] SER prediction via AIS information is automatable, whereas it is not when the user intervenes in this processing chain.

[0057] A consistency check or comparison 15 is carried out between the performances (delivered by radar processing 9) in terms of actual detection capacity observed in flight on the SER targets estimated at a constant false alarm rate, and the expected detection performances on the estimated SERs 14.

[0058] The expected detection performance on the estimated SERs 14 uses the predictions of radar detection performance similar to those of sub-steps 5 and 7.

[0059] It is possible to predict radar detection performance from detailed knowledge of: the radar antenna (transmitted power, antenna gain, aperture angle, etc.); the waveform used (recurrence frequency, pulse duration, transmitted band, etc.); the processing used (probability of false alarm at the detector input); and the characteristics of the target (SER, speed, etc.); and making assumptions about the environment such as the average power and statistical distribution of clutter or sea disorder and the propagation pattern in the atmosphere.

[0060] After capitalizing a large number of real sea clutter data with the corresponding measured ground truths (temperature, pressure, wind speed, wave height, hygrometry, etc.), a learning of the different environmental parameters is carried out, which influence the detection performance predictions, namely: the propagation conditions, and the power and statistics of the sea clutter.

[0061] For this, two convolutional neural networks can be used, for example those made according to the U-NET described in the document “Convolutional networks for biomedical image segmentation”, by Ronneberger, O., Fischer, P., & Brox, T. (2015, October), International Conference on Medical image computing and computer-assisted intervention (pp. 234-241), to learn the behavior of these quantities.

[0062] There figure 5 synthesizes the network inputs and outputs to learn the fluctuation of the quantities influencing the detection parameters.

[0063] Thus, the use of one or both neural networks allows for a more precise comparison between actual and predicted detection performance and will take into account the fluctuating nature of these quantities.

[0064] There figure 6 represents the use of one or both neural networks in step 3 of controlling the set altitude.

[0065] The principle of the control chain is to say whether the altitude is valid (OK) or not (KO).

[0066] If the detection performance obtained is better than expected, it means that we are dealing with "favorable" phenomena and therefore there is no problem.

[0067] On the other hand, if it is the opposite, the operator is aware and can change altitude to see if these phenomena are present at other altitudes and find the altitude where there is consistency.

[0068] The objective is to automate this verification to carry out this step on a significant number of targets (for example 100) in order to consolidate this control phase.

[0069] The present invention provides a rapid and robust method for calculating the recommended altitude, without having any a priori assumptions about the propagation conditions in the environment and the sensor used, and to have a control loop for this altitude by correlating the performance of the radar sensor with external data.

[0070] The present invention makes it possible to solve the problem of altitude depending on propagation conditions which is very costly in terms of time during operational missions.

Claims

1. A method for managing the altitude of an aircraft provided with an airborne radar comprising: - a step (2) of in-flight computation of a setpoint altitude maintained by the aircraft, comprising a first sub-step (6) implementing an advanced propagation model APM, which, as input, uses measurements of parameters representing the surrounding atmosphere and, as output, provides an amplification / attenuation coefficient of the radar signals used as input for a second sub-step (7) of computing a setpoint altitude from an estimation of the radar detection performance capabilities using a model of the radar; and - a step (3) of controlling the setpoint altitude that has been maintained by the aircraft using a comparison (15) between an estimation of the detection performance capabilities of the radar using a model of the radar and a measurement of the detection performance capabilities of the radar.

2. The method according to claim 1, comprising a step (1) of predetermined computation of a recommended altitude comprising a first sub-step (4) implementing the advanced propagation model APM, which, as input, uses predetermined data of parameters representing the surrounding atmosphere and, as output, delivers an amplification / attenuation coefficient of the radar signals used as input for a second sub-step (5) of computing a recommended altitude from an estimation of the radar detection performance capabilities and of the model of the radar, said recommended altitude being provided as input for the step of in-flight computation of a setpoint altitude in order to limit the flight manoeuvres required for the step of in-flight computation of a setpoint altitude around this recommended altitude in order to compute the setpoint altitude.

3. The method according to claim 2, wherein the predetermined data of parameters representing the surrounding atmosphere is updated while the aircraft is in flight.

4. The method according to any of the preceding claims, wherein the performance capabilities of the radar comprise a probability of detecting a target.

5. The method according to any of the preceding claims, wherein said parameters representing the surrounding atmosphere comprise temperature, pressure, hygrometry, wind speed, and wave height.

6. The method according to any of the preceding claims, wherein the comparison (15) between an estimation of the detection performance capabilities of the radar and a measurement of the detection performance capabilities of the radar of the step of controlling the setpoint altitude uses a first neural network (16) establishing an attenuation / amplification coefficient of the radar signals from predetermined measurements and data of parameters representing the surrounding atmosphere.

7. The method according to any of the preceding claims, wherein the comparison (15) between an estimation of the detection performance capabilities of the radar and a measurement of the detection performance capabilities of the radar of the step of controlling the setpoint altitude uses a second neural network (17) establishing a reflection coefficient and an impulsivity of the surrounding sea clutter from predetermined data measurements of parameters representing the surrounding atmosphere.

8. The method according to any of the preceding claims, wherein the estimation of the detection performance capabilities of the radar of the step (3) of controlling the setpoint altitude comprises an automatic estimation (11) of the radar cross section (RCS) of the target from (10) a location of a target detected by the radar and self-declarative information (AIS, ADS-B) of the nature and the position of the target.

9. The method according to any of the preceding claims, wherein the estimation of the detection performance capabilities of the radar of the step (3) of controlling the setpoint altitude comprises a manual estimation (13), by an operator, of the radar cross section (RCS) of the target from attributes (12) of the target comprising the length, the shape, the number and the position of the superstructures of the target from the measurements of the radar.

10. The method according to claim 9, wherein the estimation of the detection performance capabilities of the radar of the step (3) of controlling the setpoint altitude comprises a correlation of the automatic estimation of the radar cross section and the manual estimation, by an operator, of the radar cross section.