Aircraft positioning method, program product, electronic device and storage medium
By combining the image classification model and weight processing of inertial navigation data, the problem of insufficient vehicle positioning accuracy in the prior art is solved, and long-term accurate positioning is achieved without relying on GPS information.
Patent Information
- Application Number
- CN202411003171.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-25
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-25
AI Technical Summary
The existing aircraft positioning methods that do not rely on GPS information cannot achieve accurate positioning for a long time, mainly because the inertial navigation system has low accuracy and large cumulative errors.
By inputting the target image collected by the aircraft's image acquisition device into the trained image classification model, the image classification results are obtained, and combined with the inertial data provided by the inertial navigation unit, the target acquisition position is determined and the aircraft's positioning information is updated. This method considers the weight processing of image classification results and inertial data to improve the accuracy of positioning information.
Without relying on GPS information, the aircraft is accurately positioned for a long time and improves the update accuracy of positioning information.
Smart Images

Figure CN118936450A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of navigation, guidance and control technology, and in particular to an aircraft positioning method, a program product, an electronic device and a storage medium. Background Art
[0002] In traditional aircraft positioning solutions, GPS information is usually used as the source of positioning information. When GPS information is interfered or the accuracy is insufficient, the flight path of the aircraft is easily disturbed. Therefore, research on positioning solutions that do not rely on GPS information is very important.
[0003] Common positioning methods that do not rely on GPS information include inertial navigation positioning methods. Inertial navigation positioning methods rely on the inertial measurement unit of the micro inertial navigation system. However, the inertial measurement unit has low accuracy and large cumulative errors, and cannot achieve long-term accurate positioning. Summary of the invention
[0004] In view of this, the purpose of the embodiments of the present application is to provide an aircraft positioning method, a program product, an electronic device and a storage medium to solve the technical problem that the existing positioning methods that do not rely on GPS information cannot achieve long-term accurate positioning.
[0005] In a first aspect, an embodiment of the present application provides an aircraft positioning method, the method comprising:
[0006] Inputting the target image captured by the image acquisition device of the aircraft into the trained image classification model to obtain the image classification result output by the image classification model;
[0007] Determining a target acquisition position of the target image according to the image classification result and the inertial data provided by the inertial navigation unit;
[0008] The positioning information of the aircraft is updated based on the target acquisition position.
[0009] In the above implementation process, the aircraft positioning method inputs the target image captured by the image acquisition device of the aircraft into the trained image classification model, and obtains the image classification result output by the image classification model; determines the target acquisition position of the target image according to the image classification result and the inertial data provided by the inertial navigation unit; and then updates the positioning information of the aircraft based on the target acquisition position. In the process of updating the aircraft positioning information, the aircraft positioning method simultaneously considers the classification result of the target image captured by the image acquisition device and the inertial data provided by the inertial navigation unit, jointly determines the target acquisition position based on the inertial data based on the classification result of the target image, and determines the positioning information of the aircraft based on the target acquisition image, thereby improving the accuracy of each updated positioning information, and thus being able to achieve long-term accurate positioning of the aircraft without relying on GPS information. The technical problem that the existing positioning method that does not rely on GPS information cannot achieve long-term accurate positioning is solved.
[0010] Optionally, in an embodiment of the present application, the image classification result includes: a plurality of category images and a classification similarity between the target image and each of the category images; each of the category images corresponds to an image classification position;
[0011] Determining the target acquisition position of the target image based on the image classification result and the inertial data provided by the inertial navigation unit includes: determining the inertial positioning position of the aircraft based on the inertial data; determining the weight processing result of the category image based on the positioning distance between the image classification position corresponding to each category image and the inertial positioning position and the classification similarity; and determining the target acquisition position based on the weight processing result.
[0012] In the above implementation process, by performing weight processing on the category images according to the positioning distance between the image classification position and the inertial positioning position of each category of images and the classification similarity, a weight processing result that more accurately characterizes the target acquisition position information can be obtained; then the target acquisition position is determined based on the weight processing result, thereby further improving the accuracy of the positioning information updated according to the target acquisition position.
[0013] Optionally, in an embodiment of the present application, the weight processing result of the category image is determined based on the positioning distance between the image classification position and the inertial positioning position corresponding to each category image and the classification similarity, including: determining the distance score of the category image based on the positioning distance; determining the weight processing result of each category image based on the distance score, a first weight ratio corresponding to the distance score, the classification similarity and a second weight ratio corresponding to the classification similarity.
[0014] In the above implementation process, the distance score reflecting the accuracy of the image processing result can be determined by the positioning distance between the image classification position and the inertial positioning position; and then the weight processing result that can characterize the target acquisition position information is determined based on the distance score, the first weight ratio, the classification similarity and the second weight ratio.
[0015] Optionally, in an embodiment of the present application, the sum of the first weight ratio and the second weight ratio is a fixed value, and the first weight ratio is positively correlated with the positioning distance.
[0016] In the above implementation process, since the first weight ratio is positively correlated with the positioning distance, when the positioning distance between the image classification position and the inertial positioning position is relatively close, the first weight ratio is smaller and the second weight ratio is larger. At this time, the influence of the image classification result represented by the classification similarity on the target acquisition position is more considered; and when the positioning distance between the image classification position and the inertial positioning position is relatively far, the influence of the positioning distance represented by the distance score on the target acquisition position is more considered to improve the accuracy of the updated positioning information.
[0017] Optionally, in an embodiment of the present application, the weight processing result includes a weight score for each of the category images; determining the target acquisition position based on the weight processing result includes: determining, based on the weight processing result, a weight category image with the highest weight score among the category images and satisfying the preset score condition; determining an image classification range of the target image based on a preset flight speed of the aircraft, an image acquisition interval of the image acquisition device, and a previous image acquisition position; wherein the previous image acquisition position includes aircraft position information when the image acquisition device acquired a previous image of the target image; when the weight category image is within the image classification range, determining the target acquisition position based on the image classification position corresponding to the weight category image; and when the weight category image is outside the image classification range, determining the target acquisition position based on the inertial data.
[0018] In the above implementation process, the image classification range of the target image is determined according to the preset flight speed of the aircraft, the image acquisition interval of the image acquisition device and the previous image acquisition position; and the determined weight category image is verified based on the determined image classification range. When the weight category image is within the image classification range, the target acquisition position is determined according to the image classification position corresponding to the weight category image; and when the weight category image is not within the image classification range, the target acquisition position is determined according to relatively accurate inertial data; so as to further improve the accuracy of the updated positioning information.
[0019] Optionally, in an embodiment of the present application, the target image captured by the image acquisition device of the aircraft is input into a trained image classification model to obtain an image classification result output by the image classification model, including: determining an image classification range of the target image based on a preset flight speed of the aircraft, an image acquisition interval of the image acquisition device, and a previous image acquisition position; wherein the previous image acquisition position includes the position information of the aircraft when the image acquisition device captured the previous image of the target image; and inputting the image classification range and the target image into the trained image classification model to obtain the image classification result output by the image classification model.
[0020] In the above implementation process, the image classification range and the target image are input into the trained image classification model to narrow the range of the image classification results, thereby reducing the computational complexity of the image classification model and improving the speed of obtaining the image classification results; the target acquisition position can be determined more quickly based on the image classification results and inertial data, and the aircraft positioning information can be updated in a timely manner based on the target acquisition position.
[0021] Optionally, in an embodiment of the present application, the trained image classification model includes: a first-level classification model and a plurality of second-level classification models;
[0022] The method of inputting a target image acquired by an image acquisition device of the aircraft into a trained image classification model to obtain an image classification result output by the image classification model includes: extracting features of the target image to obtain target image features; inputting the target image features into the first-level classification model to obtain a first-level classification result output by the first-level classification model; wherein the first-level classification result includes multiple hierarchical images and the classification similarity between the target image and each of the hierarchical images; each of the hierarchical images includes multiple category images; according to the first-level classification result, determining a second-level target classification model from multiple second-level classification models; and inputting the target image features into the second-level target classification model to obtain the image classification result output by the second-level target classification model.
[0023] In the above implementation process, the image classification model includes a first-level classification model and multiple second-level classification models. Compared with the method of directly using a single classification model for image classification, the multi-level classification method provided by the present application (that is, first using the first-level classification model to perform feature processing on the target image features to obtain the first-level classification results, and then inputting the target image features into the second-level target classification model corresponding to the first-level classification results to obtain the image classification results output by the second-level target classification model) can greatly reduce the calculation amount of the classification model and improve the acquisition speed of the image classification results; thereby, the target acquisition position can be determined more quickly based on the image classification results and inertial data, and the aircraft positioning information can be updated in a timely manner based on the target acquisition position.
[0024] In a second aspect, an embodiment of the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the method described in any one of the first aspects above.
[0025] In a third aspect, an embodiment of the present application further provides an electronic device; the electronic device includes:
[0026] Memory;
[0027] processor;
[0028] The memory stores a computer program executable by the processor, and when the computer program is executed by the processor, the aircraft positioning method described in any one of the first aspects is performed.
[0029] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the aircraft positioning method as described in any one of the first aspects is executed.
[0030] In a fifth aspect, an embodiment of the present application further provides an aircraft positioning device, the device comprising:
[0031] A classification result acquisition module, used to input the target image acquired by the image acquisition device of the aircraft into the trained image classification model, and acquire the image classification result output by the image classification model;
[0032] A target acquisition position determination module, used to determine the target acquisition position of the target image according to the image classification result and the inertial data provided by the inertial navigation unit;
[0033] A positioning information updating module is used to update the positioning information of the aircraft based on the target acquisition position.
[0034] The beneficial effects of the present application are as follows: in the process of updating the positioning information of the aircraft, the aircraft positioning method simultaneously considers the classification results of the target images collected by the image acquisition device and the inertial data provided by the inertial navigation unit, jointly determines the target acquisition position based on the classification results of the target image and the inertial data, and the positioning information of the aircraft based on the target acquisition image, thereby improving the accuracy of each updated positioning information, and thus being able to achieve long-term accurate positioning of the aircraft without relying on GPS information. The technical problem that the existing positioning method that does not rely on GPS information cannot achieve long-term accurate positioning is solved. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0036] Figure 1 A schematic diagram of a flow chart of an aircraft positioning method provided in an embodiment of the present application;
[0037] Figure 2 A schematic diagram of a process for determining a target acquisition position provided in an embodiment of the present application;
[0038] Figure 3 A flowchart of a method for obtaining image classification results provided in an embodiment of the present application;
[0039] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0040] The following embodiments of the technical solution of the present application are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application, and are therefore only used as examples, and cannot be used to limit the scope of protection of the present application.
[0041] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application.
[0042] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.
[0043] See also Figure 1 The flowchart of an aircraft positioning method provided by an embodiment of the present application is shown. The aircraft positioning method may include the following steps:
[0044] S101, inputting a target image acquired by an image acquisition device of an aircraft into a trained image classification model, and obtaining an image classification result output by the image classification model;
[0045] S102, determining a target acquisition position of the target image according to the image classification result and the inertial data provided by the inertial navigation unit;
[0046] S103: Update the positioning information of the aircraft based on the target acquisition position.
[0047] Among them, in step S101, the aircraft can be controlled to leave the ground for space flight, such as an airplane, a drone or an aircraft. The image acquisition device can be specifically implemented by a visible light camera; the visible light camera can use the visible light band to monitor and capture images of the ground in real time from a high altitude. The image acquisition device can also be implemented by other devices that can realize target image acquisition, such as a fisheye camera or a macro camera. The image classification model can be implemented by a support vector machine, a convolutional neural network or a recurrent neural network, etc., and this application does not make specific limitations on this. The image classification result may include the image type to which the target image belongs; it may also include multiple category images and the classification similarity between the target image and each category image, and the category image most similar to the target image can be determined based on the classification similarity.
[0048] Among them, in step S102, the inertial navigation unit can measure the acceleration data and three-axis attitude data of the aircraft. The inertial data may include acceleration data and three-axis attitude data and other data that can determine the position of the aircraft. The target acquisition position refers to "the position of the aircraft when the image acquisition device acquires the target image" determined according to the image classification results and inertial data. Specifically, when the image classification results include multiple category images and the classification similarity between the target image and each category image, the optimal category image that best meets the actual flight situation can be determined according to the image classification results and inertial data, and when the image position corresponding to the optimal category image is within a reasonable position range, the image position corresponding to the optimal category image is determined as the target acquisition position; and when the image position corresponding to the optimal category image is not within a reasonable position range, the target acquisition position is determined according to the inertial position corresponding to the inertial data.
[0049] In step S103, the target acquisition position can be used as the position of the aircraft when acquiring the target image, and the positioning information of the aircraft can be updated according to the target acquisition position. The update frequency of the aircraft positioning information can be determined according to the acquisition interval of the target image, and the acquisition interval of the target image can be determined according to the model of the actual image acquisition device or the set shooting parameters.
[0050] It can be seen that the aircraft positioning method provided by the embodiment of the present application, in the process of updating the aircraft positioning information, takes into account the classification results of the target image collected by the image acquisition device and the inertial data provided by the inertial navigation unit, jointly determines the target acquisition position based on the classification results of the target image and the inertial data, and the positioning information of the aircraft based on the target acquisition image, thereby improving the accuracy of each updated positioning information, and thus being able to achieve long-term accurate positioning of the aircraft without relying on GPS information. The technical problem that the existing positioning method that does not rely on GPS information cannot achieve long-term accurate positioning is solved.
[0051] In some optional embodiments, the image classification result includes: a plurality of category images and a classification similarity between the target image and each of the category images; each of the category images corresponds to an image classification position;
[0052] S102, determining the target acquisition position of the target image according to the image classification result and the inertial data provided by the inertial navigation unit, including: determining the inertial positioning position of the aircraft according to the inertial data; determining the weight processing result of the category image according to the positioning distance between the image classification position corresponding to each category image and the inertial positioning position and the classification similarity; determining the target acquisition position according to the weight processing result.
[0053] Among them, the prescribed flight area of the aircraft can be divided to obtain multiple divided areas, and each divided area corresponds to a category image. The category image can be an image collected by the aircraft in any divided area in the prescribed flight area when the GPS information is accurate. Since each divided area corresponds to a category image, according to the image classification position corresponding to the category image, it is possible to determine "the position of the aircraft when the image acquisition device collects the target image", that is, the target acquisition position. The real-time flight speed of the aircraft can be determined according to the initial speed and acceleration data of the aircraft, the real-time flight attitude of the aircraft can be determined according to the three-axis attitude data, and then the inertial positioning position of the aircraft can be determined according to the real-time flight attitude and flight speed of the aircraft. The weight processing result can be determined according to the positioning distance, the preset distance range, the classification similarity and the preset similarity threshold; specifically, the weight processing result can be determined according to the category image whose positioning distance is within the preset distance range and whose classification similarity is greater than the preset similarity threshold; and the image classification position corresponding to the category image with the largest classification similarity in the weight processing result is used to determine the target acquisition position. Alternatively, a distance score corresponding to the positioning distance may be determined first, and then a weight processing result may be determined according to the distance score, a first weight ratio corresponding to the distance score, the classification similarity, and a second weight ratio corresponding to the classification similarity.
[0054] In some optional embodiments, the weight processing result of the category image is determined based on the positioning distance between the image classification position and the inertial positioning position corresponding to each category image and the classification similarity, including: determining the distance score of the category image based on the positioning distance; determining the weight processing result of each category image based on the distance score, a first weight ratio corresponding to the distance score, the classification similarity and a second weight ratio corresponding to the classification similarity.
[0055] Among them, the positioning distance can be divided into multiple distance levels, and different distance levels correspond to different distance scores. According to the distance level corresponding to the positioning distance between the image classification position and the inertial positioning position, the distance score of the category image can be determined. The positioning distance ranges corresponding to different distance levels may be the same or may not be completely different. The range of the distance score may be greater than or equal to 0 and less than or equal to 100; it may also be greater than or equal to 50 and less than or equal to 100. The first weight ratio and the second weight ratio may be determined according to the distance level corresponding to the positioning distance, or may be fixed ratio values. This application does not make specific restrictions on this. The weight processing result may include a weight processing score corresponding to the category image, and the weight processing score may be determined according to the product of the distance score and the first weight ratio and the sum of the product of the classification similarity and the second weight ratio; and the target acquisition position may be determined according to the image classification position corresponding to the category image with the highest weight processing score.
[0056] In some optional embodiments, the sum of the first weight ratio and the second weight ratio is a fixed value, and the first weight ratio is positively correlated with the positioning distance.
[0057] Among them, the weight ratio can be expressed by a decimal or a percentage, and the sum of the first weight ratio and the second weight ratio can be 1 or 100%. Since the first weight ratio is positively correlated with the positioning distance, when the positioning distance between the image classification position and the inertial positioning position is relatively close, the first weight ratio is relatively small and the second weight ratio is relatively large. At this time, the influence of the image classification result represented by the classification similarity on the target acquisition position is more considered; and when the positioning distance between the image classification position and the inertial positioning position is relatively far, the influence of the positioning distance represented by the distance score on the target acquisition position is more considered to improve the accuracy of the updated positioning information.
[0058] Please refer to Figure 2 , Figure 2A flow chart for determining a target acquisition position is provided for an embodiment of the present application. In some optional embodiments, the weight processing result includes a weight score for each of the category images; the determination of the target acquisition position according to the weight processing result includes: S201, determining the weight category image with the highest weight score and satisfying the preset score condition in the category image according to the weight processing result; S202, determining the image classification range of the target image according to the preset flight speed of the aircraft, the image acquisition interval of the image acquisition device, and the previous image acquisition position; wherein the previous image acquisition position includes the aircraft position information when the image acquisition device acquires the previous image of the target image; S203, when the weight category image is within the image classification range, determining the target acquisition position according to the image classification position corresponding to the weight category image; S204, when the weight category image is outside the image classification range, determining the target acquisition position according to the inertial data.
[0059] Among them, the preset score condition may include a preset score threshold; accordingly, the category image with the highest weight score and a weight score greater than the preset score threshold can be determined as a weight category image. The preset score condition may also include a distance score threshold and a similarity threshold; accordingly, the category image with the highest weight score and "the distance score is greater than the distance score threshold, and the classification similarity is greater than the similarity threshold" can be determined as a weight category image. The preset flight speed may include the maximum flight speed of the aircraft or the flight speed range corresponding to each speed level. The maximum flight distance of the aircraft within the image acquisition interval can be determined based on the maximum flight speed of the aircraft or the currently set speed level (different speeds are equal to the corresponding allowable flight speed range is inconsistent). And based on the previous image acquisition position and the maximum flight distance, the possible flight range of the aircraft is determined; the image classification range of the target image is determined based on the possible flight range. The previous image acquisition position can be determined based on GPS positioning information, or it can be determined using the aircraft positioning method provided in this application. Exemplarily, when the acquisition frequency of the image acquisition device is 50Hz, the maximum flight speed of the aircraft is Max (unit: m / s), and the previous image acquisition position is (x0, y0), the maximum flight distance of the aircraft is L = 0.02*Max (unit: m); the corresponding image classification range can be a circular range with (x0, y0) as the center and L as the radius; it can also be a square range with (x0, y0) as the center and 2L as the side length. It should be noted that the present application mainly considers the positioning information of the aircraft in longitude and latitude, and the flight altitude information of the aircraft can be determined by methods such as barometric altimetry, ultrasonic altimetry or laser radar altimetry. The determined weight category image is verified based on the image classification range. When the weight category image is within the image classification range, the target acquisition position is determined according to the image classification position corresponding to the weight category image; and when the weight category image is not within the image classification range, the target acquisition position is determined according to relatively accurate inertial data; the accuracy of the updated positioning information can be further improved.
[0060] In some optional embodiments, S101, inputting the target image captured by the image acquisition device of the aircraft into the trained image classification model to obtain the image classification result output by the image classification model, including: determining the image classification range of the target image according to the preset flight speed of the aircraft, the image acquisition interval of the image acquisition device and the previous image acquisition position; wherein the previous image acquisition position includes the position information of the aircraft when the image acquisition device captured the previous image of the target image; inputting the image classification range and the target image into the trained image classification model to obtain the image classification result output by the image classification model.
[0061] Among them, by inputting the image classification range and the target image into the trained image classification model, the range of the image classification results can be narrowed, thereby reducing the computational complexity of the image classification model and improving the speed of acquiring the image classification results; so as to more quickly determine the target acquisition position based on the image classification results and inertial data, and realize timely update of the aircraft positioning information based on the target acquisition position.
[0062] Please refer to Figure 3 , Figure 3 A flowchart of a method for obtaining image classification results provided in an embodiment of the present application.
[0063] In some optional embodiments, the trained image classification model includes: a first-level classification model and multiple second-level classification models; S101, inputting the target image acquired by the image acquisition device of the aircraft into the trained image classification model, and obtaining the image classification result output by the image classification model, including: S1011, performing feature extraction on the target image to obtain target image features; S1012, inputting the target image features into the first-level classification model, and obtaining the first-level classification result output by the first-level classification model; wherein the first-level classification result includes multiple hierarchical images and the classification similarity between the target image and each of the hierarchical images; each of the hierarchical images includes multiple category images; S1013, determining a second-level target classification model from multiple second-level classification models based on the first-level classification results; S1014, inputting the target image features into the second-level target classification model, and obtaining the image classification result output by the second-level target classification model.
[0064] Among them, when the prescribed flight area of the aircraft is determined, the prescribed flight area can be divided into multiple hierarchical images according to the actual positioning requirement accuracy, and each hierarchical image can be divided into multiple category images; to ensure that the target acquisition position determined according to the category image can meet the actual positioning requirement accuracy. The first-level classification model can be implemented by ResNet or VGGNet model, and the second-level classification model can also be implemented by ResNet or VGGNet model. The first-level classification model and the second-level classification model can be the same or different. When the first-level classification model or the second-level classification model is implemented by ResNet, Bottleneck Block can be used to replace the basic residual block to further reduce the model calculation amount and parameter amount and optimize the model's calculation efficiency. The size of each hierarchical image can be n*g*g, where n represents the number of category images that can be divided into each hierarchical image, and g*g represents the size of each category image. The specified flight area can be imaged according to the actual positioning accuracy requirement, and the multiple images of size n*g*g collected can be input into the first-level classification model to be trained for training, and the multiple images of size g*g collected can be input into the second-level classification model to be trained for training. The specified flight area can be divided into multiple areas of size n*g*g, and image collection can be performed in any area of size n*g*g, and the second-level classification model can be trained using the multiple images of size g*g collected.
[0065] An embodiment of the present application also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the aircraft positioning method as described in any one of the first aspects above.
[0066] Please refer to Figure 4 , Figure 4 The electronic device 300 is a schematic diagram of a structure of an electronic device 300 provided in an embodiment of the present application. The electronic device 300 comprises: a memory 302 and a processor 301; the memory 302 stores a computer program executable by the processor 301, and when the computer program is executed by the processor 301, the aircraft positioning method described in any one of the first aspects is executed.
[0067] The memory 302 and the processor 301 may be interconnected and communicate with each other via a communication bus 303 and / or other forms of connection mechanisms (not shown). The memory 302 stores a computer program executable by the processor 301. When the computer program is executed by the processor 301, the aircraft positioning method described in the first aspect above is executed.
[0068] The embodiment of the present application further provides a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are executed by the processor 301, the aircraft positioning method described in the first aspect above is executed.
[0069] Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.
[0070] In several embodiments provided by the embodiments of the present application, it should be understood that the disclosed device / system and method can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to the multiple embodiments of the embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and a part of the module, program segment or code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.
[0071] In addition, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part.
[0072] The above description is only an optional implementation manner of the embodiments of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the embodiments of the present application, which should be covered within the protection scope of the embodiments of the present application.
Claims
1. A method for positioning an aircraft, characterized in that: The method comprises: Inputting the target image captured by the image acquisition device of the aircraft into the trained image classification model to obtain the image classification result output by the image classification model; Determining a target acquisition position of the target image according to the image classification result and the inertial data provided by the inertial navigation unit; The positioning information of the aircraft is updated based on the target acquisition position.
2. The method according to claim 1, characterized in that in, The image classification result includes: a plurality of category images and the classification similarity between the target image and each of the category images; each of the category images corresponds to an image classification position; Determining the target acquisition position of the target image according to the image classification result and the inertial data provided by the inertial navigation unit includes: Determining the inertial positioning position of the aircraft based on the inertial data; Determining a weight processing result of the category image according to a positioning distance between the image classification position and the inertial positioning position corresponding to each category image and the classification similarity; The target acquisition position is determined according to the weight processing result.
3. The method according to claim 2, characterized in that Determining the weight processing result of the category image according to the positioning distance between the image classification position and the inertial positioning position corresponding to each category image and the classification similarity includes: Determining a distance score of the category image according to the positioning distance; The weight processing result of each of the category images is determined according to the distance score, the first weight ratio corresponding to the distance score, the classification similarity and the second weight ratio corresponding to the classification similarity.
4. The method according to claim 3, characterized in that in, The sum of the first weight ratio and the second weight ratio is a fixed value, and the first weight ratio is positively correlated with the positioning distance.
5. The method according to claim 3, characterized in that: in, The weight processing result includes a weight score of each of the category images; and determining the target acquisition position according to the weight processing result includes: According to the weight processing result, determining a weight category image with the highest weight score and meeting a preset score condition among the category images; Determining the image classification range of the target image according to the preset flight speed of the aircraft, the image acquisition interval of the image acquisition device, and the previous image acquisition position; wherein the previous image acquisition position includes the position information of the aircraft when the image acquisition device acquires the previous image of the target image; In a case where the weighted category image is within the image classification range, determining the target acquisition position according to the image classification position corresponding to the weighted category image; In a case where the weighted category image is outside the image classification range, the target acquisition position is determined according to the inertial data.
6. The method according to claim 1, characterized in that The step of inputting the target image acquired by the image acquisition device of the aircraft into the trained image classification model to obtain the image classification result output by the image classification model comprises: Determining the image classification range of the target image according to the preset flight speed of the aircraft, the image acquisition interval of the image acquisition device, and the previous image acquisition position; wherein the previous image acquisition position includes the position information of the aircraft when the image acquisition device acquires the previous image of the target image; The image classification range and the target image are input into the trained image classification model to obtain the image classification result output by the image classification model.
7. The method according to any one of claims 2 to 5, characterized in that: in, The trained image classification model includes: a first-level classification model and a plurality of second-level classification models; The step of inputting the target image acquired by the image acquisition device of the aircraft into the trained image classification model to obtain the image classification result output by the image classification model comprises: Performing feature extraction on the target image to obtain target image features; Input the target image features into the first-level classification model to obtain a first-level classification result output by the first-level classification model; wherein the first-level classification result includes a plurality of hierarchical images and a classification similarity between the target image and each of the hierarchical images; each of the hierarchical images includes a plurality of the category images; Determining a second-level target classification model from a plurality of the second-level classification models according to the first-level classification results; The target image features are input into the second-level target classification model to obtain the image classification result output by the second-level target classification model.
8. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.
9. An electronic device, characterized in that: The electronic device comprises: Memory; processor; The memory stores a computer program executable by the processor, and when the computer program is executed by the processor, the method according to any one of claims 1 to 7 is performed.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is executed.
Citation Information
Patent Citations
Unmanned aerial vehicle pose determination method and device, computer equipment and storage medium
CN110047108A
Method and device for identifying action categories, electronic equipment and storage medium
CN110222551A
Viaduct area identification method and device in vehicle navigation
CN113819910A
Target image screening method and device, electronic equipment and storage medium
CN114612696A
Vehicle control method and device, storage medium and vehicle
CN114802258A
