CN118999591A

CN118999591A - Aviation track navigation method and device based on geomagnetic signals and electronic equipment

Info

Publication number: CN118999591A
Application number: CN202411503849.0A
Authority: CN
Inventors: 兰昆艳; 刘洋; 马嘉; 窦宝成; 施航
Original assignee: Zhejiang Lab
Current assignee: Zhejiang Lab
Priority date: 2024-10-25
Filing date: 2024-10-25
Publication date: 2024-11-22
Anticipated expiration: 2044-10-25
Also published as: CN118999591B

Abstract

The specification discloses an aviation track navigation method and device based on geomagnetic signals and electronic equipment, and the method specifically comprises the following steps: and determining a geomagnetic track heat map and a two-dimensional geomagnetic tensor corresponding to the target object according to the historical track geomagnetic sequence of the target object. And determining the motion time characteristics of the target object according to the two-dimensional geomagnetic tensor, inputting a geomagnetic track heat map into a spatial characteristic extraction model, and determining the motion spatial characteristics of the target object. And fusing the motion time characteristics and the motion space characteristics to determine the predicted position information for carrying out track navigation on the target object. By the method, time and space characteristics can be effectively combined, accuracy of a predicted position is effectively improved, track deviation caused by accumulated error of an inertial navigation system is greatly corrected, the problems of application limitation of geomagnetic navigation technology based on a mode identification means and data transmission errors caused by long-distance transmission are solved, task execution efficiency is effectively improved, and meanwhile low cost consumption is also considered.

Description

Aviation track navigation method and device based on geomagnetic signals and electronic equipment

Technical Field

The present disclosure relates to the field of geomagnetic navigation, and in particular, to an aviation track navigation method, device and electronic equipment based on geomagnetic signals.

Background

With the continuous development and progress of the scientific technology at the present stage, various technologies related to aviation and flying in the aerospace field are mature. The wide spread and application of aviation flight technology brings great convenience and numerous benefits to the fields of daily life, economic trade, humane communication and the like of people.

The advantages of aviation flight technology are extremely remarkable, and how to successfully ensure the accuracy of the flight track of the flight equipment is very important. The related technology commonly used in the geomagnetic navigation field at the present stage has a more traditional rigid transformation method, geomagnetic signals are converted into position information through a linear technology, but the problem of poor nonlinear mapping and positioning effects exists in the technology. The neural network related technology based on pattern recognition can effectively solve the problem, but related recognition positioning technologies such as similar object scene matching positioning technology, regional geomagnetic field characteristic classification recognition positioning technology and the like in the prior art often need to consume higher resources and cost to perform regional modeling and labeling. Meanwhile, most of the technology belongs to indoor technologies with obvious magnetic field local distribution characteristic change, obvious numerical value change is often needed to detect change when geomagnetic signals are processed, so that position prediction accuracy is low, accuracy of flight navigation is possibly affected due to transmission error interference, and overall flight efficiency is affected.

Therefore, how to ensure that the flight equipment can efficiently complete the flight task according to the expected flight trajectory with high accuracy and low cost is a critical issue.

Disclosure of Invention

The specification provides an aviation track navigation method and device based on geomagnetic signals and electronic equipment, so as to partially solve the problems existing in the prior art.

The technical scheme adopted in the specification is as follows:

The specification provides an aviation track navigation method based on geomagnetic signals, which comprises the following steps:

Acquiring a historical track geomagnetic sequence of a target object, wherein the historical track geomagnetic sequence is used for representing a geomagnetic vector sequence acquired on a historical motion track of the target object aiming at a time point of the target object, which is divided into a preset time interval in a preset historical time length;

Generating a geomagnetic track heat map corresponding to the historical motion track of the target object and determining a two-dimensional geomagnetic tensor corresponding to the historical track geomagnetic sequence according to geomagnetic vector values corresponding to all time points in the determined historical track geomagnetic sequence;

Determining a motion time characteristic corresponding to the target object according to the historical track geomagnetic sequence and the two-dimensional geomagnetic tensor, and inputting the geomagnetic track heat map into a preset spatial characteristic extraction model, so that the spatial characteristic extraction model determines a motion spatial characteristic corresponding to the target object according to the geomagnetic track heat map;

and determining predicted position information corresponding to the target object at the next time point according to the motion time characteristics and the motion space characteristics, and performing track navigation on the target object based on the predicted position information.

Optionally, determining geomagnetic vector values corresponding to time points in the geomagnetic sequence of the historical track specifically includes:

Noise filtering is carried out on the historical track geomagnetic sequence, so that a noise-filtered historical track geomagnetic sequence is obtained;

Determining a magnetic field intensity change gradient corresponding to the noise-filtered historical track geomagnetic sequence according to the noise-filtered historical track geomagnetic sequence, and determining a historical track geomagnetic gradient sequence corresponding to the noise-filtered historical track geomagnetic sequence according to the magnetic field intensity change gradient and the noise-filtered historical track geomagnetic sequence;

And taking geomagnetic vector values corresponding to all time points in the geomagnetic gradient sequence of the historical track as geomagnetic vector values corresponding to all time points in the determined geomagnetic gradient sequence of the historical track.

Optionally, generating a geomagnetic track heat map corresponding to the historical motion track of the target object according to geomagnetic vector values corresponding to time points in the geomagnetic sequence of the historical track specifically includes:

acquiring a motion trail line of the target object on a historical motion trail;

Based on a preset regional geomagnetic heat map, determining a geomagnetic track heat map corresponding to the historical track geomagnetic sequence according to geomagnetic vector values corresponding to time points in the historical track geomagnetic sequence and the motion track line.

Optionally, according to geomagnetic vector values corresponding to time points in the historical track geomagnetic sequence, the two-dimensional geomagnetic tensor corresponding to the historical track geomagnetic sequence specifically includes:

Performing fast Fourier transform processing on the historical track geomagnetic sequence based on the time dimension of the historical track geomagnetic sequence, and determining corresponding frequency components and corresponding frequency intensities of the frequency components in the historical track geomagnetic sequence;

determining a target frequency component from the frequency components according to a preset frequency intensity threshold value, and determining a period corresponding to the target frequency component according to the frequency corresponding to the target frequency component as a geomagnetic variation period corresponding to the geomagnetic sequence of the historical track;

And carrying out sequence folding processing on the historical track geomagnetic sequence based on the geomagnetic change period and the frequency corresponding to the geomagnetic change period, and determining a two-dimensional geomagnetic tensor corresponding to the historical track geomagnetic sequence.

Optionally, determining the motion time feature corresponding to the target object according to the historical track geomagnetic sequence and the two-dimensional geomagnetic tensor specifically includes:

Inputting the two-dimensional geomagnetic tensor into a preset time feature extraction model, so that the time feature extraction model determines two-dimensional time features corresponding to the historical track geomagnetic sequence according to the two-dimensional geomagnetic tensor;

And performing dimension reduction processing on the two-dimensional time features to obtain one-dimensional time features corresponding to the two-dimensional time features, and performing weighted summation on the one-dimensional time features according to the frequency intensity corresponding to the target frequency component to obtain weighted one-dimensional time features serving as motion time features corresponding to the target object.

Optionally, determining predicted position information corresponding to the target object at a next time point according to the motion time feature and the motion space feature, and performing track navigation on the target object based on the predicted position information specifically includes:

performing feature stitching fusion on the motion time features and the motion space features to obtain space-time feature vectors corresponding to the target objects;

And inputting the space-time feature vector into a pre-trained position prediction model, so that the position prediction model predicts the position of the target object according to the space-time feature vector, obtains coordinate position information corresponding to the target object at the next time point, uses the coordinate position information as the predicted position information, and performs track navigation on the target object based on the predicted position information.

Optionally, training the position prediction model specifically includes:

Acquiring expected track information and sample space-time feature vectors corresponding to the target object;

Inputting the expected track information and the sample space-time feature vector into a position prediction model to be trained, so that the position prediction model to be trained predicts the position of the target object according to the sample space-time feature vector to obtain sample prediction position information;

And determining a loss value aiming at the position prediction model to be trained according to the sample prediction position information and the expected track position information corresponding to the sample space-time feature vector, and training the position prediction model to be trained according to the loss value, wherein the magnitude of the loss value and the similarity between the sample prediction position information and the expected track position information are in a negative correlation.

The present specification provides an aviation track navigation device based on geomagnetic signals, comprising:

the acquisition module is used for acquiring a historical track geomagnetic sequence of a target object, wherein the historical track geomagnetic sequence is used for representing a geomagnetic vector sequence acquired on a historical motion track of the target object according to time points divided by preset time intervals in preset historical time length aiming at the target object;

the generation module is used for generating a geomagnetic track heat map corresponding to the historical motion track of the target object and determining a two-dimensional geomagnetic tensor corresponding to the historical track geomagnetic sequence according to geomagnetic vector values corresponding to all time points in the determined historical track geomagnetic sequence;

The feature determining module is used for determining the motion time feature corresponding to the target object according to the historical track geomagnetic sequence and the two-dimensional geomagnetic tensor, and inputting the geomagnetic track heat map into a pre-trained spatial feature extraction model so that the spatial feature extraction model determines the motion spatial feature corresponding to the target object according to the geomagnetic track heat map;

And the prediction module is used for determining the predicted position information corresponding to the target object at the next time point according to the motion time characteristics and the motion space characteristics, and carrying out track navigation on the target object based on the predicted position information.

The present specification provides a computer readable storage medium storing a computer program which when executed by a processor implements the above-described geomagnetic signal-based aviation track navigation method.

The present disclosure provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the above-mentioned geomagnetic signal-based aviation track navigation method when executing the program.

The above-mentioned at least one technical scheme that this specification adopted can reach following beneficial effect:

According to the method, in the aviation track navigation method based on the geomagnetic signal, the geomagnetic track heat map corresponding to the historical movement track of the target object and the two-dimensional geomagnetic tensor corresponding to the historical track geomagnetic sequence can be determined according to the obtained historical track geomagnetic sequence of the target object. Then, the motion time characteristics corresponding to the target object can be determined according to the two-dimensional geomagnetic tensor, and the geomagnetic track heat map is input into a pre-trained spatial characteristic extraction model, so that the spatial characteristic extraction model determines the motion spatial characteristics corresponding to the target object according to the received geomagnetic track heat map. And finally, according to the motion time characteristics and the motion space characteristics, determining the predicted position information corresponding to the target object at the next time point, and performing track navigation on the target object based on the predicted position information.

From the above, it can be seen that, according to the geomagnetic signal-based aviation track navigation method provided in the present specification, according to the obtained geomagnetic sequence of the historical track of the target object, the motion time feature and the motion space feature of the target object on the historical motion track can be determined, and further, according to the motion time feature and the motion space feature, the position information of the target object at the next moment is predicted, so as to guide the target object to execute the motion task according to the expected track. The method in the specification can effectively and accurately predict the future position information of the target object, can effectively improve the accuracy of the predicted position information in a mode of combining time features and space features, and effectively avoids the track deviation problem caused by the accumulated error problem of the inertial navigation system and the data transmission error caused by the indoor technology long-distance transmission in the prior art. The predicted position information determined by the method can accurately navigate the target object to efficiently complete the task according to the expected track, resources and cost consumed in the process are saved compared with modeling construction aiming at geomagnetic features in the area, task execution efficiency is effectively improved, and meanwhile low cost consumption is also considered.

Drawings

The accompanying drawings, which are included to provide a further understanding of the specification, illustrate and explain the exemplary embodiments of the present specification and their description, are not intended to limit the specification unduly. In the drawings:

Fig. 1 is a schematic flow chart of an aviation track navigation method based on geomagnetic signals provided in the present specification;

fig. 2a, 2b, and 2c are schematic diagrams illustrating an geomagnetic track heat map determining procedure provided in the present specification;

Fig. 3 is a schematic overall flow structure of an aviation track navigation method based on geomagnetic signals provided in the present specification;

Fig. 4 is a schematic diagram of an aviation track navigation device based on geomagnetic signals provided in the present specification;

fig. 5 is a schematic structural diagram of an electronic device corresponding to fig. 1 provided in the present specification.

Detailed Description

For the purposes of making the objects, technical solutions and advantages of the present specification more apparent, the technical solutions of the present specification will be clearly and completely described below with reference to specific embodiments of the present specification and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present specification. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are intended to be within the scope of the present disclosure.

The following describes in detail the technical solutions provided by the embodiments of the present specification with reference to the accompanying drawings.

Fig. 1 is a schematic flow chart of an aviation track navigation method based on geomagnetic signals, which is provided in the specification, and includes the following steps:

S101: and acquiring a historical track geomagnetic sequence of the target object.

Currently, the day-to-day mature and widely popularized aerospace technology brings great practicality and convenience to various fields. The precise flying of the flying device according to the expected track is extremely important in the field of aerospace, the conventional technology at the present stage mainly corrects the inertial navigation track deviating from the expected track through rigid transformation, expected track information needs to be predicted in the positioning process, and the application effect is poor when the device faces nonlinear mapping scenes such as complex terrain.

The related technologies such as the similar regional geomagnetic field characteristic classification recognition positioning technology based on the neural network can well overcome the defects, but the technology needs to model the magnetic field characteristics in the positioning region in advance, so that resources and time are extremely consumed, and the technology is mostly used for indoor monitoring and positioning with obvious magnetic field local distribution characteristic change, and has poor navigation effect precision when facing outdoor and large-scale navigation positioning scenes. Therefore, how to ensure high accuracy and realize low cost consumption to enable the flying equipment to complete the flying task according to the preset track, thereby realizing accurate navigation of the flying equipment is an important problem.

For this reason, the present specification provides an aviation track navigation method based on geomagnetic signals, where the execution subject adopted in the method provided in the present specification may be a terminal device having a geomagnetic signal collection function, such as a desktop computer, a notebook computer, or the like, or may be a server, and in addition, the execution subject in the present specification may also be a subject in the form of software, such as a client installed in the terminal device having a geomagnetic signal collection function, or the like. For convenience of explanation, the description below uses only the terminal device as the execution subject, and describes the provided navigation method of the aviation track based on geomagnetic signals.

Based on the above, the terminal device applying the geomagnetic signal-based aviation track navigation method provided by the specification can acquire a historical track geomagnetic sequence acquired on a historical motion track aiming at a target object. According to the geomagnetic sequence of the historical track, the terminal equipment can determine the motion space characteristic and the motion time characteristic of the target object on the historical motion track, and predict the next position information of the target object according to the motion space characteristic and the motion time characteristic so as to guide the target object to fly.

The application scenario of the track navigation task executed by the terminal device aiming at the target object can be determined according to the actual application requirement. For example, for a flight device like a civil passenger plane, a helicopter, etc., which is manually and actively controlled, the terminal device can determine the corresponding spatial features and temporal features of the flight device on the historical motion track according to the geomagnetic sequence of the historical track collected by the flight device, predict the next motion position of the flight device according to the spatio-temporal features, and display the predicted position result to the relevant crews. The crew can operate the flying device to fly according to the predicted position result, or adjust the running state and parameters of the current flying device according to the predicted position result, so as to ensure that the deviation between the inertial navigation track of the flying device and the expected track is reduced as much as possible.

For another example, for unmanned aerial vehicle that need not to manually operate and can carry out the task automatically in the unmanned aerial vehicle field, terminal equipment can confirm the space feature and the time feature that unmanned aerial vehicle corresponds according to the historical orbit geomagnetic sequence that gathers through unmanned aerial vehicle to according to the fusion vector of space feature and time feature, predict unmanned aerial vehicle's positional information at next moment. The unmanned aerial vehicle can execute the flight task according to the prediction result, and meanwhile, the error of the inertial navigation system is corrected, so that deviation between the actual flight track and the expected track is effectively avoided, and the task execution efficiency corresponding to the target task is effectively improved while the target task is accurately executed according to the flight track.

In the present specification, the terminal device may acquire a historical track geomagnetic sequence of the target object, where the terminal device applying the method in the present specification may be set up on the target object, so as to conveniently acquire the historical track geomagnetic sequence of the target object in real time in the motion process of the target object.

In addition, the terminal device may also obtain preset expected track information corresponding to the target object, for example, by requesting GPS data. The expected track information is actively planned by a flight task executed by a person aiming at the target object, and can be particularly used for representing the corresponding flight track of the target object under the optimal condition when the flight task is executed. In the aviation track navigation process of the specification, expected track information can not directly participate in subsequent position prediction based on space-time fusion characteristics, and can be only used for reference comparison and evaluation of final predicted position information.

The historical track geomagnetic sequence of the target object can be specifically used for representing that a plurality of acquisition time points are divided according to preset time intervals in preset historical time duration, data acquisition is carried out on geomagnetic vectors corresponding to each time point of the target object on the historical motion track, and the geomagnetic vectors corresponding to the time points jointly form the historical track geomagnetic sequence corresponding to the target object. The geomagnetic vector collected for each time point or position may specifically have three components, which respectively correspond to components of the geomagnetic field of the position where the target object is located in the north direction, the east direction and the vertical direction, that is, geomagnetic components in the directions of the x axis, the y axis and the z axis.

Specifically, it is assumed that a time interval preset for the historical track geomagnetic sequence of the target object is 1s, the total duration of data collection is the last 10s of the historical motion process, and the preset number of the instant points is 10. The terminal device can obtain geomagnetic vectors of the positions of the corresponding time point target objects according to preset time intervals and corresponding time point numbers, and the specific data form can be { about}, Each of which isThe geomagnetic vector used for representing the geographic location of the target object at the current time point may specifically include the above-mentioned geomagnetic components in the north, east and vertical directions of the geomagnetic field corresponding to the location of the target object.

S102: and generating a geomagnetic track heat map corresponding to the historical motion track of the target object according to the geomagnetic vector values corresponding to the time points in the determined geomagnetic sequence of the historical track, and determining a two-dimensional geomagnetic tensor corresponding to the geomagnetic sequence of the historical track.

In the present specification, the terminal device may determine, according to the obtained geomagnetic sequence of the historical track corresponding to the target object, a geomagnetic track heat map corresponding to the historical motion track of the target object, and a two-dimensional geomagnetic tensor corresponding to the geomagnetic sequence of the historical track.

It should be noted in advance that, in the present description, noise problems or device heterogeneity problems may exist in the geomagnetic vector data obtained directly, which may further cause an influence on the accuracy of the subsequent position prediction result. Therefore, after the terminal device applying the method in the specification acquires the historical track geomagnetic sequence corresponding to the target object, noise filtering processing is performed on the historical track geomagnetic sequence, so that the historical track geomagnetic sequence after noise filtering is obtained.

Then, the terminal device can determine the corresponding magnetic field intensity change gradient according to the noise-filtered historical track geomagnetic sequence, so that the historical track geomagnetic gradient sequence corresponding to the noise-filtered historical track geomagnetic sequence is determined according to the noise-filtered historical track geomagnetic sequence and the determined magnetic field intensity change gradient. The terminal device may use geomagnetic vector values corresponding to each time point in the geomagnetic gradient sequence of the history track as geomagnetic vector values corresponding to each time point in the geomagnetic gradient sequence of the history track.

In the above-mentioned geomagnetic trajectory heat map determining process, specifically, the terminal device may determine and obtain, in advance, a geomagnetic heat map of a region corresponding to a large-range region corresponding to a location where the target object is located, and obtain, based on a historical motion trajectory of the target object, a motion trajectory line of the target object on the historical motion trajectory.

Then, the terminal device can determine geomagnetic measurement sequence values corresponding to historical motion track sampling points of the target object according to positions corresponding to time points in the historical motion track sequence on the motion track line based on the regional geomagnetic heat map. Then, the terminal device may fold the geomagnetic measurement sequence value corresponding to the historical motion track with a fixed length, and determine a geomagnetic track heat map of the geomagnetic sequence of the historical track.

The regional geomagnetic heat map may specifically be an RGB color geomagnetic heat map pre-constructed according to different geomagnetic signal intensities in the region, where three color components corresponding to each pixel point in the map are: red (R), green (G), and blue (B) correspond to geomagnetic components of x-axis, y-axis, and z-axis, respectively, of a position where the pixel point is located in the region. The terminal equipment can determine the geomagnetic track heat map corresponding to the geomagnetic sequence of the historical track according to the movement track line of the target object and the geomagnetic heat map of the corresponding area. The geomagnetic track heat map can be specifically used for representing the geomagnetic signal change process of geomagnetic vectors corresponding to all time points in the historical track geomagnetic sequence, wherein the geomagnetic signal change process is displayed in the form of RGB color images.

In order to facilitate understanding of the specific expression form of the geomagnetic track heat map, an explanation will be given below with reference to a schematic diagram of an example of a determination process of the geomagnetic track heat map corresponding to the geomagnetic sequence of the historical track, and specific reference may be made to fig. 2a, 2b, and 2c.

Fig. 2a, 2b, and 2c are schematic diagrams illustrating an geomagnetic trajectory heat map determining procedure provided in the present specification.

In fig. 2a, the gray scale map representation corresponding to the geomagnetic chart of the region corresponding to the position of the target object is shown, and for convenience of understanding, it is assumed that the target object moves twice in the region corresponding to the geomagnetic chart of the region according to different trajectories, and the corresponding motion trajectory lines are respectively traj1 and traj2 shown in fig. 2 a. The terminal device may determine a geomagnetic track heat map of the historical track geomagnetic sequence according to the track trend of the motion track line in the local geomagnetic heat map, as shown in fig. 2b and 2 c.

The gray map corresponding to the geomagnetic trajectory heat map of the target object when moving along the movement trajectory line traj1 is shown in fig. 2b, and the gray map corresponding to the geomagnetic trajectory heat map of the target object when moving along the movement trajectory line traj2 is shown in fig. 2 c. As can be seen from the examples in the figure, no matter how the motion trail actually changes, the motion trail can show the geomagnetic signal change process between all positions where the target object passes through in the corresponding motion trail in the form of image color change, and in practical application, the change of the image color RGB value of the geomagnetic heat map can well show the change of geomagnetic vector corresponding intensity between all time points, and the spatial characteristics of the target object on the historical motion trail are shown to a certain extent.

In addition, while determining the geomagnetic track heat map corresponding to the historical motion track of the target object according to the historical track geomagnetic sequence, the terminal equipment can also determine the two-dimensional geomagnetic tensor corresponding to the historical track geomagnetic sequence according to the historical track geomagnetic sequence.

Specifically, the terminal device may perform fast fourier transform on the historical track geomagnetic sequence according to a time dimension in the historical track geomagnetic sequence, so as to determine each frequency component corresponding to the historical track geomagnetic sequence and a frequency intensity corresponding to each frequency component.

Then, the terminal device may determine a single or multiple target frequency components from the frequency components according to a preset frequency intensity threshold, and determine a period corresponding to the target frequency components according to frequencies corresponding to the target frequency components. The terminal device may use the period corresponding to the target frequency component as the geomagnetic variation period corresponding to the geomagnetic sequence of the history track.

The above-mentioned geomagnetic variation period determination process may specifically refer to the following formula:

Wherein, In particular for representing a geomagnetic sequence of a historical track (a particular data form may be e.g. { as mentioned in the above examples}），Is aimed atPerforming fast Fourier transform) And calculating the frequency intensity corresponding to each frequency component in the determined geomagnetic sequence of the historical track according to the equation. And the second row formula is used for passingSelecting the intensity arrangement in each frequency component corresponding to the geomagnetic sequence of the history track to be in frontThe frequency components of the individual ones of the plurality of frequency components,Then is used for representing the frontThe frequency components correspond to the frequency magnitudes, respectively.For historical track geomagnetic sequencesThe corresponding length of time is set to be,Then corresponds to the frontThe period of the frequency components. The above formula can be simplified to the following formula:

Through the operation of the formula, the terminal equipment can be based on the acquired geomagnetic sequence of the historical track Determining the geomagnetic sequence of historical trackIntensity ranking in corresponding frequency componentsPeriods corresponding to the frequency componentsAnd takes it as a history track geomagnetic sequenceCorresponding geomagnetic variation period. Through the period calculation process, the magnetic field characteristics of the historical track geomagnetic sequence changing along with time can be effectively displayed in the subsequent time characteristic extraction process, so that the motion time characteristics are more obvious and are more convenient to extract to a certain extent, and the accuracy of final predicted position information is effectively improved.

Then, the terminal device can determine the corresponding movement time characteristic of the target object when moving on the historical movement track according to the historical track geomagnetic sequence of the target object and the geomagnetic change period determined through the process.

Specifically, the terminal device may perform sequence folding processing on the historical track geomagnetic sequence based on a geomagnetic variation period and a corresponding frequency corresponding to the historical track geomagnetic sequence, so as to determine two-dimensional geomagnetic tensor data corresponding to the historical track geomagnetic sequence. Regarding the specific operational flow of the folding process, the following formula may be referred to:

,

Wherein, similar to the above formula, In particular for representing a geomagnetic sequence of a historical track,For indicating the firstThe frequencies corresponding to the individual target frequency components,Then is used to represent the firstThe period corresponding to each target frequency component. In the above formulaIn particular for geomagnetic sequences in historic trajectories0 Is appended to the last bit of (c) so that in executionIn operation, may be divided by the period of the frequency componentAnd (5) integer division.The method is particularly used for representing two-dimensional geomagnetic tensor data obtained after sequence folding processing of the target frequency component. The two-dimensional geomagnetic tensor set shown as the next row can be obtained by the first row formula in the above formula,,。

And performing sequence folding processing on the historical track geomagnetic sequence through the formula, so that two-dimensional geomagnetic tensor data corresponding to the historical track geomagnetic sequence can be obtained. In the two-dimensional geomagnetic tensor, each column of tensor and each row of tensor respectively correspond to adjacent time and adjacent period in the geomagnetic sequence of the historical track. Similar time sequence changes can be reflected in the aspect of data characteristics, so that obvious two-dimensional local characteristics are formed, time characteristics in a geomagnetic sequence of a historical track can be more obviously reflected through a two-dimensional geomagnetic tensor, and the accuracy of a follow-up predicted movement position is favorably ensured.

S103: and determining the motion time characteristics corresponding to the target object according to the historical track geomagnetic sequence and the two-dimensional geomagnetic tensor, and inputting the geomagnetic track heat map into a preset spatial characteristic extraction model so that the spatial characteristic extraction model determines the motion spatial characteristics corresponding to the target object according to the geomagnetic track heat map.

In the present specification, the terminal device may determine, according to the historical track geomagnetic sequence of the target object and the two-dimensional geomagnetic tensor determined by the above steps, a motion time feature corresponding to the target object when the target object moves on the historical motion track.

Specifically, the terminal device may input the two-dimensional geomagnetic tensor determined through the above steps into a preset time feature extraction model, so that the preset time feature extraction model may extract the two-dimensional time feature corresponding to the geomagnetic sequence of the history track according to the received two-dimensional geomagnetic tensor. It should be noted that the above-mentioned time feature extraction model may be used in the present specification as a Swin Transformer model, and this type of mathematical model has the advantages of high accuracy, high timeliness and focusing on global and local information at the same time, and this design directly benefits the timing analysis task from an advanced visual backbone network.

The two-dimensional time characteristics corresponding to the geomagnetic sequence of the historical track can be extracted by a Swin transducer model, and the following formula can be specifically referred to:

Wherein, In particular for representing two-dimensional geomagnetic tensor data corresponding to a geomagnetic sequence of a historical trackThen the two-dimensional time characteristics corresponding to the geomagnetic sequences of the historical tracks extracted by the Swin transducer model.

Then, in order to ensure that the data dimensions in the subsequent position prediction are uniform in the present specification, the terminal device may perform dimension reduction processing on the two-dimensional time feature obtained through the above process, so as to obtain a one-dimensional time feature corresponding to the two-dimensional time feature. And then, the terminal equipment can carry out weighted summation on the one-dimensional time features according to the frequency intensity corresponding to the target frequency component determined in the step, so as to obtain the weighted and summed one-dimensional time features corresponding to the geomagnetic sequence of the historical track, and the weighted and summed one-dimensional time features are used as the movement time features corresponding to the target object.

Regarding the above-described dimension reduction processing performed for the two-dimensional time feature, the following formula may be referred to:

Wherein, The method is particularly used for representing the two-dimensional time characteristics corresponding to the geomagnetic sequences of the historical tracks, has the same meaning as the formulas in the above description,For indicating the firstThe frequencies corresponding to the individual target frequency components,Then is used to represent the firstThe period corresponding to the target frequency component, and in the formulaFor use in determining two-dimensional geomagnetic tensor dataThe complement 0 resulting from the operation is removed. Through the calculation process corresponding to the formula, the terminal equipment can reduce the two-dimensional time characteristicsCorresponding tensor dimension, thereby obtaining one-dimensional time characteristics corresponding to the geomagnetic sequence of the historical track。

And the following formula can be referred to for a specific process of the one-dimensional time feature according to the frequency intensity:

Wherein, For representing the frequency intensity corresponding to each target frequency component, the first formula is used for passingBased on the frequency intensity corresponding to each target frequency componentDetermining the weight corresponding to each target frequency component, namely in the first row formula. Then through the weighted summation of the secondary line formulas, the terminal equipment can determine the weighted and summed one-dimensional time characteristics corresponding to the geomagnetic sequences of the historical tracks, and further take the weighted and summed one-dimensional time characteristics as the motion time characteristics corresponding to the target objects, namely the motion time characteristics in the secondary line formulas。

It should be noted that, the whole process of determining the motion time characteristic corresponding to the target object based on the geomagnetic variation period can be actually executed through a TimesNet mathematical model, so that the problem of insufficient modeling capability due to long-term dependence of an LSTM network can be overcome. The TimesNet model comprises a stacked TimesBlock structure, and through feature extraction of the multilayer TimesBlock structure, deeper time features in a geomagnetic sequence of a historical track can be better extracted, and specific model types can be flexibly adjusted according to actual application scenes.

In addition, in the present specification, while determining the motion time feature corresponding to the target object through the above steps, the terminal device may further input the geomagnetic track heat map corresponding to the geomagnetic sequence of the historical track determined through the above steps into the pre-trained spatial feature extraction model, so that the spatial feature extraction model may determine the motion spatial feature corresponding to the target object in the geomagnetic sequence of the historical track according to the received geomagnetic track heat map.

The spatial feature extraction model can adopt an M-Darknet-PAFPN structural model, darknet is a deep convolution neural network supporting a YOLO (You Only Look Once) target detection algorithm, PAFPN is a path aggregation feature pyramid network, english is fully called Path Aggregation Feature Pyramid Network, and the model can better fuse multi-scale features in data when being applied, so that the detection capability of the whole model on objects or objects with different sizes is improved, and the spatial feature extraction model mentioned in the specification can be formed by combining the two neural networks, and is similar to a YOLOX (You Only Look Once with eXcitement) model commonly used in the prior art, but has obvious differences.

In this specification, considering that in the current technology, the computing model supporting the YOLO target detection algorithm mostly performs target detection on natural images, and the geomagnetic track heat map of the historical track geomagnetic sequence is more special compared with natural images in the ordinary technical field, in order to ensure the accuracy of the extracted features of the model and keep the uniform data dimension with the motion time features obtained in the above process, in this specification, the pre-measuring head part of the YOLOX (You Only Look Once with eXcitement) model can be removed in the final feature output stage and replaced by n CBL (Convolution-Batch Normalization-Leaky ReLU) modules with preset numbers, so that the final output features of the whole model can be mapped into low-dimensional feature vectors, and further the subsequent space-time fusion and position prediction processes are conveniently performed.

For example, the spatial feature extraction model of the M-Darknet-PAFPN structure presented in this specification differs from the YOLOX model in that n CBL modules are provided before final feature generation for reducing feature tensor dimensions. In addition, when the normal YOLOX model outputs the target feature data to the final prediction header portion, there are often different feature data corresponding to multiple dimensions, and if the terminal device inputs the geomagnetic track heatmap with the format 640×640 corresponding to the geomagnetic track geomagnetic sequence to the YOLOX model, before the feature data is input to the prediction header portion, the feature data includes feature data with multiple dimensions such as 1×256×80×80, 1×512×80×80, 1×1024×40×40, and the like, and the normal calculation flow combines the feature data with multiple dimensions, so that the calculation amount is huge, and the overall processing efficiency of the model may be affected.

In the spatial feature extraction model of the M-Darknet-PAFPN structure in the present specification, when the final feature generation is processed, only feature data of 1×256×80×80 dimensionality level may be selected and input into a plurality of preset CBL modules, the feature data is subjected to low-dimensional mapping and normalization processing by the CBL modules, and finally a linear activation process is completed through a linear activation function, so as to obtain a motion spatial feature corresponding to the target object in the geomagnetic sequence of the historical track.

For a specific process of the above CBL module for performing data processing on feature data, the following formula may be referred to:

Wherein, For a high-dimensional feature tensor in a selected dimension (e.g. feature data at a dimension level of 1 x 256 x 80 in the example above),To pass through high-dimensional feature tensorsAnd the low-dimensional characteristic tensor is obtained after the dimension mapping calculation of the top row formula. The second row of formulas is used entirely for low-dimensional feature tensorsIs used for the normalization calculation process of (1),Is normalized vector data calculated by a formula. The third row of formulas is a linear activation function, modified linear correction unit (Leaky ReLU) pairs in the CBL moduleA linear activation process is performed. Finally, the terminal equipment can obtain the motion space characteristics corresponding to the target object in the geomagnetic sequence of the historical track through the space characteristic extraction modelIn the above formula，，，，，Are all optimized learning parameters in the training process aiming at the space extraction feature modelAndMean calculation and standard deviation calculation are represented respectively.

S104: and determining predicted position information corresponding to the target object at the next time point according to the motion time characteristics and the motion space characteristics, and performing track navigation on the target object based on the predicted position information.

In the present specification, the terminal device may predict the position information corresponding to the target object at the next time point according to the motion time feature and the motion space feature determined through the above steps. Then, the terminal device can conduct track navigation on the target object based on the generated predicted position information, so that the target object can complete tasks according to the normal expected track.

Specifically, since the vector dimensions corresponding to the motion time feature and the motion space feature determined through the steps are the same as the one-dimensional feature data, the terminal device can perform vector splicing and fusion processing on the motion time feature and the motion space feature corresponding to the target object on the historical motion track so as to obtain the space-time feature vector corresponding to the target object. Then, the terminal device can predict the position of the target object according to the space-time feature vector corresponding to the target object, determine the coordinate position information corresponding to the target object at the next time point, and serve as the predicted position information.

In this specification, the specific method of position prediction may be a position regression processing operation, and may specifically be implemented by using a preset full connection layer (Fully Connected layer, FC) module as a position prediction model. The terminal equipment can input the space-time feature vector into a pre-trained position prediction model, so that the position prediction model can perform position prediction on the target object according to the space-time feature vector to obtain coordinate position information corresponding to the target object at the next time point, and the coordinate position information is used as prediction position information.

It should be noted that, the terminal device may perform model training on the fully-connected layer module according to the position prediction model to be trained, which is preset based on the preset expected track information, so that the fully-connected layer module may effectively memorize the mapping relationship between the expected track information and the time-space fusion feature. And further, when the position prediction is carried out, the predicted position information of the target object can be determined accurately according to the space-time characteristic vector of the target object, so that the motion track of the target object is adjusted, and the deviation error between the inertial navigation track and the expected track of the actual motion is close to zero. In addition, the specific application method of the position prediction operation can be flexibly adjusted according to the actual application scene and the requirement, and is not strictly limited in the specification.

In the training process of the position prediction model, the terminal equipment can input the obtained expected track information and sample space-time feature vector corresponding to the target object into the position prediction model to be trained, so that the position prediction model to be trained can perform position prediction on the target object according to the received sample space-time feature vector, and sample prediction position information is obtained. Then, the terminal equipment can determine a loss value of the position prediction model to be trained in the training process according to the obtained sample prediction position information and expected track position information corresponding to the sample space-time feature vector, train the position prediction model to be trained according to the loss value, and adjust the network parameter of the position prediction model to be trained so that the prediction result of the position prediction model can trend to the expected track information more. The magnitude of the loss value and the similarity between the sample predicted position information and the expected track position information are in a negative correlation.

In addition, the time interval between the predicted position information and the current position obtained through the above process can be adjusted according to the requirement, and the time interval between the predicted position information and the current position is not strictly limited to the time interval between each time point in the geomagnetic sequence of the historical track in the specification. For example, assuming that the geomagnetic sequence of the historical track is a geomagnetic sequence with a total duration of 10s at a time interval of 2s, the predicted position information when predicting the position information of the target object through the above steps may be the predicted position information of the target position after 2s corresponding to the time of the last sampling point. The time length can be flexibly set according to the actual requirement, for example, the prediction information after 5s or 1s of the last sampling point can be also used.

It should be noted that, normally, the position corresponding to the last sampling point may not coincide with the current position of the target object when the predicted position information is generated, and when the moving speed of the target object is slow or the position predicting time is very short, the position corresponding to the last sampling point may be similar to the current position of the target object when the predicted position information is generated.

In addition, generally, the position of the last sampling time point in the geomagnetic sequence of the historical track often has a certain distance from the actual position where the target object is located when the predicted position information is determined, and in practical application, the calculation time of the position prediction process is often less than 1s. Therefore, the historical track sampling interval is reasonably set, so that the predicted position is closer to the actual position. In a word, the time interval between the two can be flexibly set according to the requirement, and the numerical limitation in the strict sense is avoided.

Further, after the terminal device determines the predicted position information for the target object, the predicted position information can be fed back to the target object to guide or prompt the target object to move according to the predicted position information, so that an accurate motion navigation function is achieved on the target object, and the deviation influence of the accumulated error problem commonly existing in the inertial navigation track on the actual motion track is effectively reduced. For example, in a flight scene with active control by personnel, the terminal device can determine the predicted position information of the target object at the next moment through the steps according to the historical track geomagnetic sequence corresponding to the target object, and feed back the predicted position information to the control personnel of the target object so as to prompt the related personnel of the future flight track of the current target object, ensure that the flight task of the target object can be normally carried out according to the predicted track, and avoid the problem of deviation from the predicted track due to the accumulated error problem of the inertial navigation system (Inertial Navigation System, INS).

In order to facilitate understanding of the above-described overall process from the history track geomagnetic sequence to the generation of predicted position information of the target object, an explanation will be given below with an overall flow structure diagram of an aviation track navigation method based on geomagnetic signals, as shown in fig. 3.

Fig. 3 is a schematic overall flow structure of an aviation track navigation method based on geomagnetic signals provided in the present specification.

As shown in fig. 3, the terminal device may perform data preprocessing on the acquired geomagnetic sequence of the historical track, that is, noise filtering and gradient processing in the above steps. Then, the terminal device can determine the motion time characteristic and the motion space characteristic of the target object on the historical motion track respectively based on geomagnetic vector values corresponding to sampling time points in the processed historical track geomagnetic sequence. Then, the terminal device can perform data splicing and fusion on the motion time features and the motion space features which are in the form of one-dimensional tensor data, input the fused space-time feature vectors into a position prediction model which is obtained by training based on expected track information of the target object in advance, and perform position regression processing according to the space-time feature vectors through the position prediction model so as to obtain the predicted position information of the target object.

In addition to the above, in the present specification, the model accuracy of the overall trajectory prediction network model may be adjusted by performing overall training based on the trajectory prediction network model including the feature extraction process and the position prediction process, and optimizing the error loss value of the finally obtained predicted position. The specific training process of the track prediction network model of the track navigation method based on geomagnetic signals provided in the specification can be realized, and the following formula can be referred to:

Wherein, For representing the positioning error of the trajectory prediction network model during the training process,For representing predicted position information predicted by a trajectory prediction network model to be trained based on sample dataStandard predicted location information for representing the correspondence of the sample data,For representing the number of prediction points. By the above formula, the positioning error is minimizedModel parameters of each part in the track prediction network model are adjusted (such as the CBL module in the motion space feature extraction process)，，，，And) And optimizing the whole track prediction network model, so that the accuracy of the final predicted position information in the actual application process is improved.

In addition, the EMA (Exponential Moving Average) weight updating method in the prior art can effectively smooth weight distribution in the model in the training process, so that generalization capability and stability of the model are improved. The cosine learning rate scheduling (CosineLRScheduler) is a strategy for dynamically adjusting the learning rate, and the method can adjust the model learning rate according to the periodic change of the cosine function, help the model jump out of a local optimal solution and promote better convergence. Therefore, in order to improve the training effect of the overall track prediction network model, the training methods such as EMA weight updating and cosine learning rate scheduling can be integrated into the training process of the track prediction network model, so that the track prediction network model can obtain more accurate and efficient position prediction capability through the training process.

From the above, it can be seen that, according to the geomagnetic signal-based aviation track navigation method provided in the present specification, according to the obtained geomagnetic sequence of the historical track of the target object, the motion time feature and the motion space feature of the target object on the historical motion track can be determined, and further, according to the motion time feature and the motion space feature, the position information of the target object at the next moment is predicted, so as to guide the target object to execute the motion task according to the expected track.

The method in the specification can effectively and accurately predict the future position information of the target object, and can effectively improve the accuracy of the predicted position by combining the time characteristics and the space characteristics, so that the problem of track deviation caused by the accumulated error problem of the inertial navigation system in the prior art is avoided to a great extent, and the target object can be accurately navigated to efficiently complete the task according to the expected track. Meanwhile, compared with modeling construction aiming at geomagnetic features in areas, the method has the advantages of greatly reducing resources and cost, along with low cost and high efficiency.

The above is a method implemented by one or more embodiments of the present disclosure, and based on the same concept, the present disclosure further provides a corresponding geomagnetic signal-based aviation track navigation apparatus, as shown in fig. 4.

Fig. 4 is a schematic diagram of an aviation track navigation device based on geomagnetic signals provided in the present specification, including:

The obtaining module 401 is configured to obtain a historical track geomagnetic sequence of a target object, where the historical track geomagnetic sequence is used to represent a geomagnetic vector sequence acquired by dividing a time point of the target object in a preset historical time period according to a preset time interval on a historical motion track of the target object;

a generating module 402, configured to generate a geomagnetic track heat map corresponding to a historical motion track of the target object according to geomagnetic vector values corresponding to time points in the determined historical track geomagnetic sequence, and determine a two-dimensional geomagnetic tensor corresponding to the historical track geomagnetic sequence;

The feature determining module 403 is configured to determine a motion time feature corresponding to the target object according to the historical track geomagnetic sequence and the two-dimensional geomagnetic tensor, and input the geomagnetic track heat map to a pre-trained spatial feature extraction model, so that the spatial feature extraction model determines a motion spatial feature corresponding to the target object according to the geomagnetic track heat map;

And the prediction module 404 is configured to determine predicted position information corresponding to the target object at a next time point according to the motion time feature and the motion space feature, and perform trajectory navigation on the target object based on the predicted position information.

Optionally, the generating module 402 is specifically configured to perform noise filtering on the historical track geomagnetic sequence to obtain a historical track geomagnetic sequence after noise filtering; determining a magnetic field intensity change gradient corresponding to the noise-filtered historical track geomagnetic sequence according to the noise-filtered historical track geomagnetic sequence, and determining a historical track geomagnetic gradient sequence corresponding to the noise-filtered historical track geomagnetic sequence according to the magnetic field intensity change gradient and the noise-filtered historical track geomagnetic sequence; and taking geomagnetic vector values corresponding to all time points in the geomagnetic gradient sequence of the historical track as geomagnetic vector values corresponding to all time points in the determined geomagnetic gradient sequence of the historical track.

Optionally, the generating module 402 is specifically configured to obtain a motion track line of the target object on the historical motion track; based on a preset regional geomagnetic heat map, determining a geomagnetic track heat map corresponding to the historical track geomagnetic sequence according to geomagnetic vector values corresponding to time points in the historical track geomagnetic sequence and the motion track line.

Optionally, the generating module 402 is specifically configured to perform a fast fourier transform process on the historical track geomagnetic sequence based on a time dimension of the historical track geomagnetic sequence, and determine each frequency component corresponding to the historical track geomagnetic sequence and a frequency intensity corresponding to each frequency component; determining a target frequency component from the frequency components according to a preset frequency intensity threshold value, and determining a period corresponding to the target frequency component according to the frequency corresponding to the target frequency component as a geomagnetic variation period corresponding to the geomagnetic sequence of the historical track; and carrying out sequence folding processing on the historical track geomagnetic sequence based on the geomagnetic change period and the frequency corresponding to the geomagnetic change period, and determining a two-dimensional geomagnetic tensor corresponding to the historical track geomagnetic sequence.

Optionally, the feature determining module 403 is specifically configured to input the two-dimensional geomagnetic tensor into a preset time feature extraction model, so that the time feature extraction model determines, according to the two-dimensional geomagnetic tensor, a two-dimensional time feature corresponding to the geomagnetic sequence of the historical track; and performing dimension reduction processing on the two-dimensional time features to obtain one-dimensional time features corresponding to the two-dimensional time features, and performing weighted summation on the one-dimensional time features according to the frequency intensity corresponding to the target frequency component to obtain weighted one-dimensional time features serving as motion time features corresponding to the target object.

Optionally, the prediction module 404 is specifically configured to perform feature stitching fusion on the motion time feature and the motion space feature to obtain a space-time feature vector corresponding to the target object; and inputting the space-time feature vector into a pre-trained position prediction model, so that the position prediction model predicts the position of the target object according to the space-time feature vector, obtains coordinate position information corresponding to the target object at the next time point, uses the coordinate position information as the predicted position information, and performs track navigation on the target object based on the predicted position information.

Optionally, the prediction module 404 is specifically configured to obtain expected track information and a sample space-time feature vector corresponding to the target object; inputting the expected track information and the sample space-time feature vector into a position prediction model to be trained, so that the position prediction model to be trained predicts the position of the target object according to the sample space-time feature vector to obtain sample prediction position information; and determining a loss value aiming at the position prediction model to be trained according to the sample prediction position information and the expected track position information corresponding to the sample space-time feature vector, and training the position prediction model to be trained according to the loss value, wherein the magnitude of the loss value and the similarity between the sample prediction position information and the expected track position information are in a negative correlation.

The present specification also provides a computer-readable storage medium storing a computer program operable to perform the above-described geomagnetic signal-based aviation track navigation method provided by fig. 1.

The present specification also provides a schematic structural diagram of an electronic device corresponding to fig. 1 shown in fig. 5. At the hardware level, as shown in fig. 5, the electronic device includes a processor, an internal bus, a network interface, a memory, and a nonvolatile storage, and may of course include hardware required by other services. The processor reads the corresponding computer program from the nonvolatile memory to the memory and then runs the computer program to realize the aviation track navigation method based on geomagnetic signals as shown in the figure 1.

In the 90 s of the 20 th century, improvements to one technology could clearly be distinguished as improvements in hardware (e.g., improvements to circuit structures such as diodes, transistors, switches, etc.) or software (improvements to the process flow). However, with the development of technology, many improvements of the current method flows can be regarded as direct improvements of hardware circuit structures. Designers almost always obtain corresponding hardware circuit structures by programming improved method flows into hardware circuits. Therefore, an improvement of a method flow cannot be said to be realized by a hardware entity module. For example, a programmable logic device (Programmable Logic Device, PLD) (e.g., field programmable gate array (Field Programmable GATE ARRAY, FPGA)) is an integrated circuit whose logic functions are determined by user programming of the device. A designer programs to "integrate" a digital system onto a PLD without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Moreover, nowadays, instead of manually manufacturing integrated circuit chips, such programming is mostly implemented with "logic compiler (logic compiler)" software, which is similar to the software compiler used in program development and writing, and the original code before being compiled is also written in a specific programming language, which is called hardware description language (Hardware Description Language, HDL), but HDL is not just one, but a plurality of kinds, such as ABEL（Advanced Boolean Expression Language）、AHDL（Altera Hardware Description Language）、Confluence、CUPL（Cornell University Programming Language）、HDCal、JHDL（Java Hardware Description Language）、Lava、Lola、MyHDL、PALASM、RHDL（Ruby Hardware Description Language）, and VHDL (Very-High-SPEED INTEGRATED Circuit Hardware Description Language) and Verilog are currently most commonly used. It will also be apparent to those skilled in the art that a hardware circuit implementing the logic method flow can be readily obtained by merely slightly programming the method flow into an integrated circuit using several of the hardware description languages described above.

The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, application SPECIFIC INTEGRATED Circuits (ASICs), programmable logic controllers, and embedded microcontrollers, examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, atmel AT91SAM, microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic of the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller in a pure computer readable program code, it is well possible to implement the same functionality by logically programming the method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Such a controller may thus be regarded as a kind of hardware component, and means for performing various functions included therein may also be regarded as structures within the hardware component. Or even means for achieving the various functions may be regarded as either software modules implementing the methods or structures within hardware components.

The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. One typical implementation is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being functionally divided into various units, respectively. Of course, the functions of each element may be implemented in one or more software and/or hardware elements when implemented in the present specification.

It will be appreciated by those skilled in the art that embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the present specification may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present description can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present description is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the specification. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.

It will be appreciated by those skilled in the art that embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the present specification may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present description can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.

The foregoing is merely exemplary of the present disclosure and is not intended to limit the disclosure. Various modifications and alterations to this specification will become apparent to those skilled in the art. Any modifications, equivalent substitutions, improvements, or the like, which are within the spirit and principles of the present description, are intended to be included within the scope of the claims of the present description.

Claims

1. An aviation trajectory navigation method based on geomagnetic signals, characterized by comprising:

Acquire a historical trajectory geomagnetic sequence of the target object, where the historical trajectory geomagnetic sequence is used to represent a sequence of geomagnetic vectors collected on the historical motion trajectory of the target object at time points divided according to preset time intervals within a preset historical duration;

Generate a geomagnetic trajectory heat map corresponding to the historical motion trajectory of the target object and determine a two-dimensional geomagnetic tensor corresponding to the historical trajectory geomagnetic sequence according to the determined geomagnetic vector value corresponding to each time point in the historical trajectory geomagnetic sequence;

Determine the motion time characteristics corresponding to the target object according to the historical trajectory geomagnetic sequence and the two-dimensional geomagnetic tensor, and input the geomagnetic trajectory heat map into a preset spatial feature extraction model, so that the spatial feature extraction model determines the motion spatial characteristics corresponding to the target object according to the geomagnetic trajectory heat map;

According to the motion time feature and the motion space feature, the predicted position information corresponding to the target object at the next time point is determined, and trajectory navigation is performed on the target object based on the predicted position information.

2. The method according to claim 1, wherein determining the geomagnetic vector value corresponding to each time point in the historical trajectory geomagnetic sequence specifically comprises:

Performing noise filtering on the historical trajectory geomagnetic sequence to obtain a noise-filtered historical trajectory geomagnetic sequence;

According to the noise-filtered historical trajectory geomagnetic sequence, determining the magnetic field intensity change gradient corresponding to the noise-filtered historical trajectory geomagnetic sequence, and according to the magnetic field intensity change gradient and the noise-filtered historical trajectory geomagnetic sequence, determining the historical trajectory geomagnetic gradient sequence corresponding to the noise-filtered historical trajectory geomagnetic sequence;

The geomagnetic vector value corresponding to each time point in the historical trajectory geomagnetic gradient sequence is used as the determined geomagnetic vector value corresponding to each time point in the historical trajectory geomagnetic sequence.

3. The method according to claim 1, characterized in that generating a geomagnetic trajectory heat map corresponding to the historical motion trajectory of the target object according to the geomagnetic vector values corresponding to each time point in the historical trajectory geomagnetic sequence specifically comprises:

Obtaining a motion trajectory of the target object on a historical motion trajectory;

Based on a preset regional geomagnetic heat map, a geomagnetic trajectory heat map corresponding to the historical trajectory geomagnetic sequence is determined according to the geomagnetic vector values corresponding to each time point in the historical trajectory geomagnetic sequence and the motion trajectory line.

4. The method according to claim 1, characterized in that determining the two-dimensional geomagnetic tensor corresponding to the historical trajectory geomagnetic sequence according to the geomagnetic vector value corresponding to each time point in the historical trajectory geomagnetic sequence specifically comprises:

Based on the time dimension of the historical trajectory geomagnetic sequence, the historical trajectory geomagnetic sequence is subjected to fast Fourier transform processing to determine the corresponding frequency components in the historical trajectory geomagnetic sequence and the frequency intensity corresponding to the frequency components;

According to a preset frequency intensity threshold, a target frequency component is determined from the frequency components, and according to the frequency corresponding to the target frequency component, a period corresponding to the target frequency component is determined as the geomagnetic change period corresponding to the historical trajectory geomagnetic sequence;

Based on the geomagnetic variation cycle and the frequency corresponding to the geomagnetic variation cycle, the historical trajectory geomagnetic sequence is subjected to sequence folding processing to determine the two-dimensional geomagnetic tensor corresponding to the historical trajectory geomagnetic sequence.

5. The method according to claim 4, characterized in that the motion time characteristics corresponding to the target object are determined according to the historical trajectory geomagnetic sequence and the two-dimensional geomagnetic tensor, specifically comprising:

Inputting the two-dimensional geomagnetic tensor into a preset time feature extraction model, so that the time feature extraction model determines the two-dimensional time feature corresponding to the historical trajectory geomagnetic sequence according to the two-dimensional geomagnetic tensor;

The two-dimensional time feature is subjected to dimensionality reduction processing to obtain a one-dimensional time feature corresponding to the two-dimensional time feature, and the one-dimensional time feature is weighted and summed according to the frequency intensity corresponding to the target frequency component to obtain a weighted one-dimensional time feature as the motion time feature corresponding to the target object.

6. The method according to claim 1, characterized in that, according to the motion time feature and the motion space feature, determining the predicted position information corresponding to the target object at the next time point, and performing trajectory navigation on the target object based on the predicted position information, specifically comprises:

The motion time feature and the motion space feature are combined and fused to obtain a spatiotemporal feature vector corresponding to the target object;

The spatiotemporal feature vector is input into a pre-trained position prediction model so that the position prediction model predicts the position of the target object according to the spatiotemporal feature vector, obtains the coordinate position information corresponding to the target object at the next time point as the predicted position information, and performs trajectory navigation on the target object based on the predicted position information.

7. The method according to claim 6, wherein training the location prediction model specifically comprises:

Obtaining expected trajectory information and sample spatiotemporal feature vectors corresponding to the target object;

Inputting the expected trajectory information and the sample spatiotemporal feature vector into a position prediction model to be trained, so that the position prediction model to be trained predicts the position of the target object according to the sample spatiotemporal feature vector to obtain sample predicted position information;

According to the sample predicted position information and the expected trajectory position information corresponding to the sample spatiotemporal feature vector, a loss value for the position prediction model to be trained is determined, and according to the loss value, the position prediction model to be trained is trained, wherein the size of the loss value is negatively correlated with the degree of similarity between the sample predicted position information and the expected trajectory position information.

8. An aviation trajectory navigation device based on geomagnetic signals, characterized by comprising:

An acquisition module, used to acquire a historical trajectory geomagnetic sequence of a target object, wherein the historical trajectory geomagnetic sequence is used to represent a sequence of geomagnetic vectors collected on a historical motion trajectory of the target object at time points divided according to preset time intervals within a preset historical duration;

A generation module, for generating a geomagnetic trajectory heat map corresponding to the historical motion trajectory of the target object and determining a two-dimensional geomagnetic tensor corresponding to the historical trajectory geomagnetic sequence according to the determined geomagnetic vector values corresponding to each time point in the historical trajectory geomagnetic sequence;

A feature determination module, for determining the motion time feature corresponding to the target object according to the historical trajectory geomagnetic sequence and the two-dimensional geomagnetic tensor, and inputting the geomagnetic trajectory heat map into a pre-trained spatial feature extraction model, so that the spatial feature extraction model determines the motion spatial feature corresponding to the target object according to the geomagnetic trajectory heat map;

A prediction module is used to determine the predicted position information corresponding to the target object at the next time point according to the motion time characteristics and the motion space characteristics, and perform trajectory navigation on the target object based on the predicted position information.

9. A computer-readable storage medium, characterized in that the storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the program.