Method of determining an angular direction of a target and system therefor

Abstract

A computer-implemented method for determining an angular direction of a target is provided. The method includes obtaining radar data from a radar system comprising an antenna with at least four subarrays. The method further includes using the radar data in a trained artificial intelligence (Al) model to determine the angular direction of the target.

Description

[0001] METHOD OF DETERMINING AN ANGULAR DIRECTION OF A TARGET AND SYSTEM THEREFOR

[0002] TECHNICAL FIELD

[0003] The presently disclosed subject matter relates to determining an angular direction of a target, and more particularly, to a method for determining an angular direction of a target based on data obtained from a radar system comprising an antenna with at least four subarrays and a system therefor.

[0004] BACKGROUND

[0005] Determining an angular direction of a target based on data obtained from a radar system is usually carried out, in known systems, by an algorithm called monopulse. Monopulse radar is a radar system that uses additional encoding of the radio signal received in the radar system, to provide accurate directional information. The name refers to its ability to extract angle information such as range and direction of a target, from a single signal pulse, as opposed to transmitting multiple narrow-beam pulses in different directions and looking for the maximum return (see Wikipedia "Monopulse radar", "Monopulse Antennas" from microwaves 101, https: / / www.microwavesl01.com / encyclopedias / monapulse -antennas).

[0006] Monopulse assumes the origin of a thermal noise model and relies on calibration of the antenna of the radar system. Therefore, preliminary calibration of the antenna before it is installed on a platform is required, e.g. as well as an ongoing maintenance of the antenna, to avoid deviations due to calibration inaccuracies occurring over time or platform electromagnetic disruptions.

[0007] It is therefore desired to determine an angular direction of a target in an accurate manner, without being dependent on calibration of the antenna, while reducing environmental effects on the radar system. Also, it is desired to determine accurately an angular direction of a target, using the current channels of antenna, without the need to increase their number. Friendly targets usually report their location. It is convenient and advantageous to use this information for maintenances during operation.

[0008] GENERAL DESCRIPTION

[0009] In an object detection system, radar systems use radio waves to determine the range, altitude, direction, or speed of targets. Information on spatial angles of targets can be obtained using Monopulse Radar systems. The Monopulse Radar systems uses four antennas of the radar system, or quadrants of a single antenna. For example, the antennas can be horns, or sections of a flat plate array of radiators, or, alternatively, subarrays of an active electrically scanned antenna (AESA) phased array. In Monopulse Radar systems, the target is illuminated by all four quadrants equally. A comparator network is used to "calculate" four return signals (also to be referred to as "channels"). A sum signal has the same pattern in receive mode as transmit mode. The sum signal may be used to track target distance and sometimes velocity. An elevation difference signal is formed by subtracting the two upper quadrants from the two lower quadrants, and may be used to calculate the target's position relative to the horizon. An azimuth difference signal is formed by subtracting the left quadrants from the right quadrants and is used to calculate the target's position to the left or right. A fourth, additional, signal, sometimes called a "Q difference" or a "delta signal" is the diagonal difference of the quadrants. This signal is often not being used, as in practice, the value of the diagonal difference of the quadrants equals zero. Therefore, typically, monopulse receiver uses only three channels to provide information on the angular direction of a target (see "Monopulse Antennas" from microwaves 101 as above).

[0010] The accuracy of the obtained angular direction from a radar system, when using a monopulse algorithm, is insufficient. A lack of accuracy or a reduced accuracy may be a result of various factors, including among others, calibration inaccuracies of the antenna of the radar system, that occur over time or occurred as the antenna was calibrated in an antenna range, but laterwas installed on a platform, such that it affected the calibration of the antenna. Also, the accuracy of the obtained angular direction may be affected by platform electromagnetic disruptions.

[0011] In order to achieve improved accuracy in determining an angular direction of a target, according to the presently disclosed subject matter, an artificial intelligence (Al) model is used. In some examples, the raw data obtained from the radar system, which is commonly used in a monopulse algorithm, is used, instead of or in addition to, a trained Al model. The Al model is configured to determine the angular direction of the target based on the raw data.

[0012] Training and using an Al model with the raw data obtained from the radar system may increase the accuracy in determining the angular direction of a target as compared to determining the angular direction using a monopulse algorithm as detailed above, as the raw data used by the trained Al model, which usually includes some deviation or noise, is used in an Al model, in an optimal way, thus achieving improved performance. For example, as explained further below, the return diagonal difference may be inputted in the Al model; thus, overall an additional input is provided in the determination process, compared to the known monopulse algorithm. Also, in some examples, the Al model can be used in an antenna which was not calibrated, or which was initially calibrated but now reflects some calibration inaccuracies over time; nevertheless the Al model achieves accurate determination of the angular direction.

[0013] According to a first aspect of the presently disclosed subject matter, there is provided a computer-implemented method for determining an angular direction of a target comprising: obtaining radar data from a radar system comprising an antenna with at least four subarrays; and using the radar data in a trained artificial intelligence (Al) model to determine the angular direction of the target.

[0014] In addition to the above features, the computer implemented method according to this aspect of the presently disclosed subject matter can optionally comprise in some examples one or more of features (i) to (x) below, in any technically possible combination or permutation:

[0015] (i). Wherein the radar system is installed on a mobile platform.

[0016] (ii). Wherein the antenna is configured to be installed on the mobile platform, and wherein the model is configured to determine the angular direction, regardless of where the antenna is installed on the mobile platform.

[0017] (iii). Wherein the radar data comprises at least a diagonal difference signal, and the method comprises using the diagonal difference signal to determine the angular direction.

[0018] (iv). Wherein the model is configured to determine the angular direction and compensate for deviation due to calibration inaccuracy over time or mobile platform electromagnetic disruption.

[0019] (v). The method further comprising: executing a monopulse algorithm using the radar data; and using at least an output of the Al model and an output of the monopulse algorithm in combination with at least one criterion to determine the angular direction.

[0020] (vi). Wherein the at least one criterion includes at least one environmental condition.

[0021] (vii). Wherein the at least one criterion includes at least one condition pertaining to the platform.

[0022] (viii). The method further comprising: (a) obtaining multiple samples of radar data, each sample being obtained from illumination directed to respective first targets by the radar system, each first target being located at a known angular direction;

[0023] (b) maintaining the Al model by mapping the multiple samples to the known angular direction for each sample;

[0024] (c) repeatedly executing (a) and (b) over a period of time during which calibration of the antenna is prone to deviation from an initial calibration; whereby using the Al model allows determination of the angular direction of the target, despite periodic deviation in calibration of the antenna.

[0025] (ix). The method further comprising: determining that a second trained Al model should be used; pre-processing the radar data to obtain a respective monopulse ratio; and using the pre-processed radar data in a second trained Al model to determine the angular direction of the target.

[0026] (x). The method further comprising: based o the radar data, selecting a trained Al model from a plurality of trained Al models; and using the selected Al model to determine the angular direction of the target.

[0027] The presently disclosed subject matter further comprises a computer system comprising a processing circuitry that comprises at least one processor and a computer memory, the processing circuity is configured to execute a method as described above with reference to the first aspect and may optionally further comprise one or more of the features (i) to (x) listed above, mutatis mutandis, in any technically possible combination or permutation. The presently disclosed subject matter further comprises a non-transitory computer readable storage medium tangibly embodying a program of instructions that, when executed by a computer, cause the computer to perform a method as described above with reference the first aspect, and may optionally further comprise one or more of the features (i) to (x) listed above, mutatis mutandis, in any technically possible combination or permutation.

[0028] According to a second aspect of the presently disclosed subject matter, there is yet further provided a computer-implemented method for training an artificial intelligence (Al) model configured to determine the angular direction of a target, the method comprising: obtaining multiple samples of radar data, each sample being obtained from illumination directed to respective first targets by a radar system comprising an antenna with at least four subarrays, each first target being located at a known angular direction; and maintaining an Al model by mapping the multiple samples to the known angular direction for each sample, whereby the Al model allows determination of an unknown angular direction of a second target, from received radar data obtained from illumination directed to the second target by a second radar system comprising an antenna with at least four subarrays.

[0029] In addition to the above features, the method according to this aspect of the presently disclosed subject matter can comprise features (i) and / or (ii) listed below, in any desired combination or permutation which is technically possible:

[0030] (i). Wherein each sample is pre-processed to obtain a respective monopulse ratio, and the method further comprising: maintaining a second Al model by mapping the respective monopulse ratios to the known angular direction for each sample, whereby the second Al model allows the determination of the unknown angular direction of the second target.

[0031] (ii). Wherein each sample is pre-processed to obtain a respective monopulse ratio, and maintaining the Al model by mapping the multiple pre-processed samples to the known angular direction for each sample, whereby the Al model allows determination of an unknown angular direction of a second target, from received radar data obtained from illumination directed to the second target by a second radar system comprising an antenna with at least four subarrays, where the received radar data is pre-processed to obtain a respective monopulse ratio.

[0032] The presently disclosed subject matter further comprises a computer system comprising a processing circuitry that comprises at least one processor and a computer memory, the processing circuity is configured to execute a method as described above with reference the second aspect and may optionally further comprise one or more of the features (i) and / or (ii) listed above, mutatis mutandis, in any technically possible combination or permutation.

[0033] The presently disclosed subject matter further comprises a non-transitory computer readable storage medium tangibly embodying a program of instructions that, when executed by a computer, cause the computer to perform a method as described above with reference the second aspect, and may optionally further comprise one or more of the features (i) and / or (ii) listed above, mutatis mutandis, in any technically possible combination or permutation.

[0034] According to a third aspect of the presently disclosed subject matter, there is yet further provided a computerized system for determining an angular direction of a target comprising: a radar system comprising an antenna with at least four subarrays, configured to receive radar data; and a processing circuitry comprising at least one processer and computer memory, the processing circuitry being configured to use the radar data in a trained artificial intelligence (Al) model to determine the angular direction of the target.

[0035] In addition to the above features, the system according to this aspect of the presently disclosed subject matter can comprise one or more of features (i) to (vii) listed below, in any desired combination or permutation which is technically possible:

[0036] (i). Wherein the radar system is installed on a mobile platform.

[0037] (ii). Wherein the antenna is installed on the mobile platform, and wherein the model is configured to determine the angular direction regardless of where the antenna is installed on the mobile platform.

[0038] (iii). Wherein the model is configured to determine the angular direction and compensate for deviation in calibration of the antenna.

[0039] (iv). herein the processing circuitry being configured to execute a monopulse algorithm using the radar data and to use at least an output of the Al model and an output of the monopulse algorithm in combination with at least one criterion to determine the angular direction.

[0040] (v). herein the at least one criterion includes at least one environmental condition.

[0041] (vi). Wherein the at least one criterion includes at least one condition pertaining to the platform.

[0042] (vii). Wherein the processing circuitry being configured to: a) obtain multiple samples of radar data, each sample was obtained from illumination directed to respective first targets by the radar system, each first target being located at a known angular direction; b) maintain the Al model by mapping the multiple samples to the known angular direction for each sample; c) repeatedly execute (a) and (b) over a period of time during which calibration of the antenna is prone to deviation from an initial calibration; whereby using the Al model allows determination of the angular direction of the target, despite periodic deviation in calibration of the antenna.

[0043] BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to understand the invention and to see how it can be carried out in practice, embodiments will be described, by way of non-limiting examples, with reference to the accompanying drawings, in which:

[0045] Fig. 1 illustrates a functional block diagram of a determination system, in accordance with certain embodiments of the presently disclosed subject matter;

[0046] Fig. 2 illustrates a generalized flow-chart of a method for determining an angular direction of a target, in accordance with certain embodiments of the presently disclosed subject matter;

[0047] Fig. 3 illustrates a generalized flow-chart method for training the Al model over time period, in accordance with certain embodiments of the presently disclosed subject matter; and

[0048] Fig. 4 illustrates a generalized flow-chart method for training an artificial intelligence (Al) model in accordance with certain embodiments of the presently disclosed subject matter.

[0049] DETAILED DESCRIPTION

[0050] In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be understood by those skilled in the art that the presently disclosed subject matter may be practiced without these specific details. In other instances, well-known methods, procedures, components and circuits have not been described in detail so as not to obscure the presently disclosed subject matter.

[0051] Unless specifically stated otherwise, as apparent from the following discussions, it is appreciated that throughout the specification discussions utilizing terms such as "obtaining", "using", "determining", "executing", "maintaining", "preprocessing", or the like, refer to the action(s) and / or process(es) of a computer that manipulate and / or transform data into other data, said data represented as physical, such as electronic, quantities and / or said data representing the physical objects. The term "computer" should be expansively construed to cover any kind of hardware-based electronic device with data processing capabilities including a personal computer, a server, a computing system, a communication device, a processor or processing unit (e.g. digital signal processor (DSP), a microcontroller, a microprocessor, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), etc.), and any other electronic computing device, including, by way of non-limiting example, computerized systems or devices such determination system 110, disclosed in the present application.

[0052] The terms "non-transitory memory" and "non-transitory storage medium" used herein should be expansively construed to cover any volatile or non-volatile computer memory suitable to the presently disclosed subject matter.

[0053] The operations in accordance with the teachings herein may be performed by a computer specially constructed for the desired purposes or by a general-purpose computer specially configured for the desired purpose by a computer program stored in a non-transitory computer-readable storage medium.

[0054] Embodiments of the presently disclosed subject matter are not described with reference to any particular programming language. It will be appreciated that a variety of programming languages may be used to implement the teachings of the presently disclosed subject matter as described herein.

[0055] Usage of conditional language, such as "may", "might", or variants thereof, should be construed as conveying that one or more examples of the subject matter may include, while one or more other examples of the subject matter may not necessarily include, certain methods, procedures, components and features. Thus, such conditional language is not generally intended to imply that a particular described method, procedure, component or circuit is necessarily included in all examples of the subject matter. Moreover, the usage of non-conditional language does not necessarily imply that a particular described method, procedure, component or circuit is necessarily included in all examples of the subject matter. Also, reference in the specification to "one case", "some cases", "other cases", or variants thereof, means that a particular feature, structure or characteristic described in connection with the embodiment(s) is included in at least one embodiment of the presently disclosed subject matter. Thus the appearance of the phrase "one case", "some cases", "other cases" or variants thereof does not necessarily refer to the same embodiment(s).

[0056] It is appreciated that certain features of the presently disclosed subject matter, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the presently disclosed subject matter, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable sub-combination.

[0057] Bearing this in mind, attention is drawn to Fig. 1 illustrating a functional block diagram of a determination system 110 , in accordance with certain embodiments of the presently disclosed subject matter. System 110 may include a computer that comprises several components which operatively communicate with each other, and that is configured to determine an angular direction of a target. System 110 comprises a processor and memory circuitry (PMC) 120 comprising a processor 130 and a memory 140. System 110 may further comprise a radar system 150 comprising an antenna 152 configured to illuminate a target and receive return signals. System 110 further comprises a communication interface 180, enabling system 110 to operatively communicate with external devices such as environmental sensors 160, and / or with external storages, which may store angle samples 170. System 110 is configured to be installed on a platform, either static such as a static ground platform, or on a mobile platform 100 illustrated in Fig. 1, such as a vehicle or an airborne platform.

[0058] Radar system 150 can be any radar system configured to provide radar data suitable for executing monopulse algorithm. For example, radar system 150 can comprise an antenna 152, such as phased array antenna, comprising a plurality of antenna elements, which can be subdivided into subarrays. For example, antenna 152 can have at least four subarrays. Radar system 150 is configured to illuminate a target and receive a single measurement, e.g. a single received pulse. The signal received at the antenna elements is processed to generate a sum signal and three difference signals. The sum (2) may be used to track target distance and sometimes velocity. The azimuth difference signal (AAZ) may be used to calculate the target's position to the left or right. The elevation difference signal (AEL) may be used to calculate the target's position relative to the horizon, and a diagonal difference signal (Q) is calculated from the diagonal difference of the quadrants of the antenna. If the antenna has a symmetric structure, then the diagonal difference signal usually equals zero, and hence, in practice, is not used. All four return signals are to be referred to herein and below as "sum and difference signals". In some examples, the signal used for generating the sum and difference signals can be measured on two or more separate antenna or antenna elements. In some examples, the signal can be measured from displaced phase centers of an array antenna (see Wikipedia "Phase-comparison monopulse").

[0059] It should be noted that for purpose of illustration only, the following description is provided for a radar system 150 comprising an antenna 152 with at least four subarrays. Those skilled in the art will readily appreciate that the teachings of the presently disclosed subject matter are, likewise, applicable to any radar system that is configured to provide radar data suitable for executing monopulse algorithm, with any suitable arrangement of an antenna. Other types are well known, including the array type, in which the beamformer generates the desired beams directly, and these other types of antennas may be used in the system 110 according to the claimed subject matter, so long as they are arranged to produce at least the sum and difference signals.

[0060] The processor 130 is configured to execute several functional modules in accordance with computer-readable instructions implemented on a non-transitory computer-readable storage medium such as memory 140. Such functional modules are referred to hereinafter as comprised in the processor 130. The processor 130 can comprise an obtaining module 132, determining module 134, monopulse module 136 and training module 138. According to certain embodiments of the presently disclosed subject matter, in orderto determine the angular direction of a target, obtaining module 132 is configured to obtain radar data from a radar system 150 (not shown in Fig. 1). The radar data comprises sum and difference signals that were generated by a comparator component included in the radar system 150, using known methods, from raw data received by the radar system 150. Determining module 134 is configured to use the radar data obtained by obtaining module 132 in a trained artificial intelligence (Al) model to determine the angular direction of the target. In some examples, the input may not require any pre-processing and may include the raw data from the signal received at the antenna 152. In some examples, the input may be pre-processed, as described further below to calculate a monopulse ratio, i.e. the result of the divisions between the different channels, and then may be inputted into the Al model. Optionally, the Al model may receive addition input data such as beam direction, or geo location for the platform, such as platform 100. Alternatively, or additionally, the addition input data may be used to choose between different Al models to execute. The output of the Al model may include the deviation angle from the center of the beam that was illuminated by antenna 152, used for determining an angular direction of a target. For example, an artificial neural network (ANN) of some depth can be used. The ANN may receive sampled signal from all four channels and provide the deviation angle from the center of the beam. The trained Al model can be stored in Al models memory 142 in the memory 140.

[0061] The Al models memory 142 stored in the memory 140 can store one or more Al models configured to determine the angular direction of the target. Determining module 134 can retrieve one or more of the plurality of stored Al models to determine the angular direction. The memory can further store internal angle samples 144 used for training the Al models memory 142 stored in the memory 140 and selection criteria 146 used for determining the weight that should be given to the output of the Al model compared to executed monopulse algorithm or other Al models. Usage of both internal angle samples 144 and selection criteria 146 is described further below with respect to Fig. 4.

[0062] Determination system 110 is configured to communicate with external elements such as environmental sensors 160, e.g. sensors installed on the platform such as GPS (geo location and velocities), accelerometers (platform orientation - yaw, pitch and roll) or other communication sensors for receiving location from friendly / non friendly targets, or with an external memory such as angle samples 170 which may also be used for training the Al models memory 142 stored in the memory 140.

[0063] It is noted that the teachings of the presently disclosed subject matter are not bound by the determination system 110 described with reference to Fig. 1. Equivalent and / or modified functionality can be consolidated or divided in another manner and can be implemented by any appropriate combination of software with firmware and / or hardware and executed on a suitable device. For example, the obtaining module 132 can be comprised in the radar system 150, and environmental sensors 160 can be comprised in the determination system 110. Those skilled in the art will also readily appreciate that the data repositories, such as memory 140, Al models memory 142, Internal angle samples 144, or angle samples 170 can be consolidated or divided in other manner; databases can be shared with other systems or be provided by other systems, including third party equipment.

[0064] Referring to Fig. 2, there is illustrated a generalized flow-chart 200 of a method for determining an angular direction of a target, in accordance with certain embodiments of the presently disclosed subject matter. The following flowchart operations are described with reference to elements of determination system 110 and the PMC 120 including processor 130 described in Fig. 1. However, this is by no means binding, and the operations can be performed by elements other than those described herein.

[0065] In some cases, radar data is obtained from a radar system, such as radar system 150 (block 210). The radar system 150 can comprise an antenna with at least four subarrays. The radar data can be obtained e.g., by obtaining module 121 included in processor 130. The radar data can be used e.g. by determining module 134 in a trained artificial intelligence (Al) model to determine the angular direction of the target (block 220). The input to the Al model can be as described above, e.g. the raw data from the radar system or a pre-processed signal including monopulse ratio, while the output of the Al model may include the deviation angle from the center of the beam.

[0066] Using an Al model, as opposed to using the monopulse algorithm as is done in known systems, is advantageous, as it may achieve accurate determination of angular direction of a target irrespective of whether the antenna on the radar system is calibrated or not. In some cases, even if the antenna is initially calibrated, once it is installed on a platform, such as an airborne platform, there is a certain reduction in calibration of the antenna, e.g. since the calibration process did not consider interferences from the fuselage that disrupts the beam, or other mobile platform electromagnetic disruptions. In other cases, even if the antenna is initially calibrated, over time, the accuracy of the calibration is reduced. Therefore, using the Al model may achieve accurate determination of angular direction of a target, since the Al model can be trained after the antenna is already installed on the platform, and can, optionally, be repeatedly trained and maintained over a period of time during which the calibration accuracy is reduced. Therefore, the Al model may be configured to determine the angular direction and compensate for deviation due to calibration inaccuracy over time or mobile platform electromagnetic disruption.

[0067] In some examples, the radar system 150 is installed on a static platform such as ground station. Yet, in some examples, the radar system 150 is installed on a mobile platform, e.g. a vehicle or an airborne platform. Since the Al model is trained when installed on the platform, the Al model is configured to determine the angular direction regardless of where the antenna is installed on the mobile platform. For example, the antenna can be installed at various angles and locations of the airborne platform, such as fighter aircraft, airborne early warning and control (AEW&C). Yet, the data obtained from the radar system 150 can be used by the trained Al model to accurately determine the angular direction of the target.

[0068] Upon receipt of a single at the radar system 150, sum and difference signals may be generated, using known methods. The difference signals include, among others, the delta signal which is the diagonal difference of the quadrants. As explained above, this signal is often not used as, in practice, the antenna has a symmetric structure, such that the value of the diagonal difference of the quadrants equals zero. Therefore, typically, the known monopulse algorithm uses only three channels to provide information on the angular direction of a target. However, in some cases, such as where there are interferences of the airborne platform on the antenna, the symmetric structure is not perfect, resulting in a non-zero value of the diagonal difference. This non-zero value of the diagonal difference is insignificant when inputted into the monopulse algorithm, yet, due to the nature of operation of Al models, this input may improve the accuracy of the determined angular direction of the target, when used in an Al model. Therefore, in some examples, the determining module 134 can use the diagonal difference signal from the radar data to determine the angular direction of the target (block 230), while this value would be ignored or omitted in the known monopulse algorithm.

[0069] In some examples, in addition to the execution of the Al model, a monopulse algorithm can be executed using the same radar data obtained from the radar system 150. Determining the angular direction of a target can be based on a combination of the outputs of the two algorithms, the Al model and the monopulse. Hence, monopulse module 136 can execute a monopulse algorithm using the radar data (block 240). Then, the determining module 134 can use at least the output of the Al model and the output of the monopulse algorithm to determine the angular direction (block 250). The two outputs can be used in combination with at least one criterion. In some examples, the criterion can include a weight given to each output in a combined output of the two outputs, e.g. equal weight to each of the outputs such that an average output is generated based on the two outputs of the algorithms, or, increased or zero weight may be given to one of the outputs. The criterion can also include at least one environmental condition, such as weather conditions, etc. Alternatively or additionally, the criterion can include at least one condition pertaining to the platform, such as the current velocity of the mobile platform, angles of the antenna, etc. For example, different elevations of an airborne platform may be associated with different weather conditions, which may affect the operation of the antenna. In different elevations, different weights may be given to each of the outputs of the Al model and the monopulse model. Optional criteria can be stored and may be updated in e.g. selection criteria 146, used for determining the weight that should be given to the output of the Al model compared to executed monopulse algorithm or other Al models.

[0070] In some cases, a second Al model may be maintained and used, in addition, or instead, to the first Al model. For example, it may be determined that for certain radar systems, such as Fire Control Radar (FCR), Airborne Early Warning (AEW), or for certain angular direction of targets, a different or an additional mapping of a second Al model should be used, as its determination may be more accurate than that which is used for the other angular direction of targets.

[0071] In some examples, a selection algorithm may be used to select from a plurality of available Al models the current Al model to use for determining the angular direction. The selection algorithm may consider several factors, such as the specific radar system orthe direction of the beam for selecting the Al model. In such cases where the direction of the beam is considered, the direction of the beam may be provided as an input to the selection algorithm, e.g. by executing a monopulse algorithm or by applying a default Al model, and then providing the angular direction to the selection algorithm for choosing a second Al model for receiving a more accurate results. This may be advantageous, e.g. in cases where a certain Al model is superior over other Al models in terms of accuracy at a certain angle or conditions. In some cases, the radar data may be pre-processed to obtain monopulse ratio and using this pre-processed radar data to apply a second Al model to determine the angular direction of the target. Training an Al model over monopulse ratio as an input should require less data since the mapping of the Al model may be simpler.

[0072] Referring back to Fig. 2, the radar data may be pre-processed, e.g. by obtaining module 138, to obtain a respective monopulse ratio (block 260). The pre-processed radar data may be used in a second trained Al model to determine the angular direction of the target (block 220).

[0073] Note that for purpose of illustration only, only one additional Al model is described. However, those skilled in the art will readily appreciate that the teachings of the presently disclosed subject matter are, likewise, applicable to more than one additional Al models which are trained and / or maintained for particular cases where their mapping is likely to provide a more accurate determination. One or more Al models can be selected to be executed to determine the angular direction of a target. Usually, over time, the calibration accuracy of an antenna decreases such that the calibration of the antenna is prone to deviation from an initial calibration. To maintain the Al model to provide accurate determination of an angular direction of a target, despite deviation in calibration, the Al model can be repeatedly trained, to keep mapping of radar data obtained from the radar system to accurate angular direction of targets, despite the deviation in calibration.

[0074] Referring to Fig. 3, there is illustrated a generalized flow-chart method 300 for training the Al model, in accordance with certain embodiments of the presently disclosed subject matter. The operations of Fig. 3 are described with reference to elements of determination system 110 and the PMC 120 including processor 130 described in Fig. 1. However, this is by no means binding, and the operations can be performed by elements other than those described herein.

[0075] Operations 210 and 220 in Fig. 3 of obtaining the radar data and using the radar data in a trained Al model are similar to those described with respect to Fig. 2. In some examples, in order to maintain the Al model such that using of the Al model allows determination of the angular direction of the target, despite periodic deviation in calibration of the antenna, training module 138 included in processor 130 is configured to obtain multiple samples of radar data (block 310). Samples can be stored in e.g. internal angle samples 144 stored in memory 130 or can be samples obtained from an external database, such as angle samples 170. Each sample may be obtained from illumination directed to targets by the radar system 150. For example, the sample may include the return signal, received at the radar system 150, when the radar system 150 illuminated to a certain target, referred to as a first target. The angular direction of the first target is known. These return signals of first targets and their respective known angular direction may be used as a training set to maintain the Al model, by training it to map the multiple samples to the known angular direction for each sample (block 320). The process of obtaining training sets and maintaining the Al model with the training sets may be executed repeatedly over a period of time (shown by the arrow 330). Each training set comprising samples of radar data obtained over a period of time e.g. from the initial calibration of the antenna, already embodies some deviation of the antenna. Thus, training over time, may be advantageous, since the Al model is maintained to determine accurate angular direction of targets, despite any periodic deviation in calibration of the antenna.

[0076] Moreover, usually in known systems, an antenna must be periodically calibrated. Calibration is performed in lab conditions, e.g. in an antenna range, usually, by removing the antenna from the platform on which it is installed, calibrating the antenna, and then reinstalling the antenna on the platform. This process entails several problems. First, the platform must halt its operation to enable the calibration process. Halting the operation is even more problematic in large airborne platforms. Second, calibrating the antenna in lab conditions, and then reinstalling the antenna on the platform already causes some deviation in the calibration of the antenna. Hence, maintaining the Al model to accurately determine the angular direction of the target, despite the periodic deviation in calibration of the antenna, is advantageous.

[0077] Referring to Fig. 4, there is illustrated a generalized flow-chart 400 of a method for determining an angular direction of a target, in accordance with certain embodiments of the presently disclosed subject matter. The following flowchart operations are described with reference to elements of determination system 110 and the PMC 120 including processor 130 described in Fig. 1. However, this is by no means binding, and the operations can be performed by elements other than those described herein.

[0078] The process illustrated in Fig. 3 referred to repeatedly training an Al model over a period of time, where the Al model was already trained and used. Training over time the Al model, as illustrated in Fig. 3, achieves at least the advantages described above. The method illustrated in Fig. 4 refers to the initial training of the Al model. Training an Al model may be performed on one platform and then be used on a different platform. If the antenna on the first platform is calibrated, then the Al model is trained with data sets obtained from calibrated antenna, and thus, may be used on different, calibrated, platforms. In cases where the antenna is not calibrated on the first platform, then the data sets used for training the Al model, which are obtained from the uncalibrated antenna, may already embody the uncalibrated measurements of the antenna. The trained Al model will thus be able to determine an angular direction of a new target, despite the antenna being uncalibrated, when the antenna directs illumination to the new targe, and the data was obtained from the same radar system.

[0079] As illustrated in Fig. 4, in some cases, in order to train an artificial intelligence (Al) model to determine the angular direction of a target, obtaining module 132 may obtain multiple samples of radar data (block 410). Each sample may be obtained from illumination directed to targets by a radar system comprising an antenna with at least four subarrays, such as radar system 150. Each sample may include the return signal, received at the radar system 150, when the radar system 150 directs illumination to a certain target, referred to as a first target, whose angular direction is known. These samples and their respective known angular directions may be used e.g. by training module 138, as a training set to maintain an Al model, by training it to map the multiple samples to the known angular direction for each sample (block 420). The trained Al model can be stored in Al models memory 142. A different, second radar system, comprising an antenna with at least four subarrays, such as radar system 150 (which may be different than the one used for training) may obtain radar data from illumination directed to a different, second target. The trained Al model may allow determination of the unknown angular direction of the second target.

[0080] As described above, in some cases, a second Al model may be maintained, in addition to the first Al model, e.g. for use in certain radar systems or in certain angular direction of targets. A second Al model may be trained using monopulse ratio calculated based on the radar system. A second Al model can be used in cases where its mapping is likely to provide a more accurate determination than that which is provided by the first IA model, such as for certain radar systems, such as Fire Control Radar (FCR), Airborne Early Warning (AEW), or for certain angular direction of targets, as described above. To train a second Al model each sample may be pre-processed, e.g. by training module 138, to obtain a respective monopulse ratio (block 430). A second Al model may be maintained, e.g. by training module 138, by mapping the respective monopulse ratios to the known angular direction for each sample (block 440). The second Al model may allow the determination of the unknown angular direction of a second target, where the radar data obtained with respect to the second target is pre-processed to obtain a respective monopulse ratio. The second Al model can be stored in e.g. Al models memory 142. Usage of the second Al model instead of or in addition to the first Al model may be determined, e.g. by determining module 134, in accordance with the selection algorithm described above.

[0081] It is noted that the teachings of the presently disclosed subject matter are not bound by the flow chart illustrated in Figs. 2-4, the illustrated operations can occur out of the illustrated order. For example, operations 220 and 240 shown in succession can be executed substantially concurrently or in the reverse order.

[0082] It should be noted that the term "criterion" as used herein should be expansively construed to include any compound criterion, including, for example, several criteria and / or their logical combinations. Also, the specific examples of criteria should not be considered as limiting, and those skilled in the art will readily appreciate that the teachings of the presently disclosed subject matter are, likewise, applicable to other criteria.

[0083] It is to be understood that the invention is not limited in its application to the details set forth in the description contained herein or illustrated in the drawings. The invention is capable of other embodiments and of being practiced and carried out in various ways. Hence, it is to be understood that the phraseology and terminology employed herein are for the purpose of description and should not be regarded as limiting. As such, those skilled in the art will appreciate that the conception upon which this disclosure is based may readily be utilized as a basis for designing other structures, methods, and systems for carrying out the several purposes of the presently disclosed subject matter. It will also be understood that the system according to the invention may be, at least partly, implemented on a suitably programmed computer. Likewise, the invention contemplates a computer program being readable by a computer for executing the method of the invention. The invention further contemplates a non- transitory computer-readable memory tangibly embodying a program of instructions executable by the computer for executing the method of the invention.

[0084] Those skilled in the art will readily appreciate that various modifications and changes can be applied to the embodiments of the invention as hereinbefore described without departing from its scope, defined in and by the appended claims.

Claims

CLAIMS:

1. A computer-implemented method for determining an angular direction of a target comprising: obtaining radar data from a radar system comprising an antenna with at least four subarrays; and using the radar data in a trained artificial intelligence (Al) model to determine the angular direction of the target.

2. The method of claim 1, wherein the radar system is installed on a mobile platform.

3. The method of claim 2, wherein the antenna is configured to be installed on the mobile platform, and wherein the model is configured to determine the angular direction regardless of where the antenna is installed on the mobile platform.

4. The method of any one of the preceding claims, wherein the radar data comprises at least a diagonal difference signal, and the method comprises using the diagonal difference signal to determine the angular direction.

5. The method of any one of the preceding claims, wherein the model is configured to determine the angular direction and compensate for deviation due to calibration inaccuracy over time or mobile platform electromagnetic disruption.

6. The method of any one of the preceding claims, further comprising: executing a monopulse algorithm using the radar data; and using at least an output of the Al model and an output of the monopulse algorithm in combination with at least one criterion to determine the angular direction.

7. The method of claim 6, wherein the at least one criterion includes at least one environmental condition.

8. The method of claim 6 or 7, wherein the at least one criterion includes at least one condition pertaining to a platform on which the radar system is installed.

9. The method of any one of the preceding claims, further comprising:(a) obtaining multiple samples of radar data, each sample being obtained from illumination directed to respective first targets by the radar system, each first target being located at a known angular direction;(b) maintaining the Al model by mapping the multiple samples to the known angular direction for each sample;(c) repeatedly executing (a) and (b) over a period of time during which calibration of the antenna is prone to deviation from an initial calibration; whereby using the Al model allows determination of the angular direction of the target, despite periodic deviation in calibration of the antenna.

10. The method of any one of the preceding claims further comprising: determining that a second trained Al model should be used; pre-processing the radar data to obtain a respective monopulse ratio; and using the pre-processed radar data in a second trained Al model to determine the angular direction of the target.

11. The method of any one of the preceding claims, further comprising: based o the radar data, selecting a trained Al model from a plurality of trained Al models; and using the selected Al model to determine the angular direction of the target.

12. A computer-implemented method for training an artificial intelligence (Al) model configured to determine an angular direction of a target, the method comprising: obtaining multiple samples of radar data, each sample being obtained from illumination directed to respective first targets by a radar system comprising an antennawith at least four subarrays, each first target being located at a known angular direction; and maintaining an Al model by mapping the multiple samples to the known angular direction for each sample, whereby the Al model allows determination of an unknown angular direction of a second target, from received radar data obtained from illumination directed to the second target by a second radar system comprising an antenna with at least four subarrays.

13. The method of claim 12, wherein each sample is pre-processed to obtain a respective monopulse ratio, and the method further comprising: maintaining a second Al model by mapping the respective monopulse ratios to the known angular direction for each sample, whereby the second Al model allows the determination of the unknown angular direction of the second target.

14. The method of claim 12 or 13, wherein each sample is pre-processed to obtain a respective monopulse ratio, and maintaining the Al model by mapping the multiple pre-processed samples to the known angular direction for each sample, whereby the Al model allows determination of an unknown angular direction of a second target, from received radar data obtained from illumination directed to the second target by a second radar system comprising an antenna with at least four subarrays, where the received radar data is pre-processed to obtain a respective monopulse ratio.

15. A computerized system for determining an angular direction of a target comprising: a radar system comprising an antenna with at least four subarrays, configured to receive radar data; and a processing circuitry comprising at least one processer and computer memory, the processing circuitry being configured to use the radar data in a trained artificial intelligence (Al) model to determine the angular direction of the target.

16. The system of claim 15 wherein the radar system is installed on a mobile platform.

17. The system of claim 16 the antenna is installed on the mobile platform, and wherein the model is configured to determine the angular direction regardless of where the antenna is installed on the mobile platform.

18. The system of any one of claims 15 to 17, wherein the model is configured to determine the angular direction and compensate for deviation in calibration of the antenna.

19. The system of any one of claims 15 to 17, wherein the processing circuitry being configured to execute a monopulse algorithm using the radar data and to use at least an output of the Al model and an output of the monopulse algorithm in combination with at least one criterion to determine the angular direction.

20. The system of claim 19, wherein the at least one criterion includes at least one environmental condition.

21. The system of claim 19 or 20, wherein the at least one criterion includes at least one condition pertaining to a platform on which the radar system is installed.

22. The system of any one of claims 15 to 21, wherein the processing circuitry being configured to: a) obtain multiple samples of radar data, each sample was obtained from illumination directed to respective first targets by the radar system, each first target being located at a known angular direction; b) maintain the Al model by mapping the multiple samples to the known angular direction for each sample; c) repeatedly execute (a) and (b) over a period of time during which calibration of the antenna is prone to deviation from an initial calibration;whereby using the Al model allows determination of the angular direction of the target, despite periodic deviation in calibration of the antenna.

23. A computer system for determining an angular direction of a target, the system comprising a processing circuitry comprising at least one processer and computer memory, the processing circuitry being configured to execute the method as defined by any one of claims 1 to 11.

24. A non-transitory computer readable storage medium tangibly embodying a program of instructions that, when executed by a computer, cause the computer to perform a method for determining an angular direction of a target as defined by any one of claims 1 to 11.

25. A computer system for determining an angular direction of a target, the system comprising a processing circuitry comprising at least one processer and computer memory, the processing circuitry being configured to execute the method as defined by any one of claims 12 to 14.

26. A non-transitory computer readable storage medium tangibly embodying a program of instructions that, when executed by a computer, cause the computer to perform a method for determining an angular direction of a target as defined by any one of claims 12 to 14.

Images