Track feature modeling method and system, and electronic device for track retrieval

Abstract

The present invention belongs to the technical field of aircraft trajectory modeling. In order to solve at least one technical problem in the prior art that track retrieval has insufficient accuracy and the processing speed cannot meet the real-time requirements, the present invention provides a track feature modeling method and system, an electronic device, and a non-volatile storage medium for track retrieval. A multi-level earth position dictionary is established and used to design track features; the multi-level earth position dictionary is then used to quickly find similar tracks; because the earth position coding adopts a multi-level algorithm, a set of tracks that meet the conditions can be preliminarily screened and found through a step-by-step refinement method, and then the refined regional dictionary is gradually used to accurately locate the most similar track; the search efficiency of the track is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft trajectory modeling, and in particular to a track feature modeling method and system, an electronic device, and a non-volatile storage medium for track retrieval. Background Art

[0002] Aircraft tracks are important data for target identification, target warning, and target behavior intention analysis. In most cases, the target's travel activities are relatively regular, so we can use similar track retrieval technology to achieve target identification, warning, behavior analysis, data completion, etc. However, in actual situations, the real-time track obtained is very unstable, and there are often problems such as missing data, abnormal data, and data sampling without a cycle or inconsistent cycles.

[0003] For example, the Chinese patent application with the Chinese patent application number CN201510084399.0 provides a processing method using the Douglas-Peucker algorithm to compress the track and then use the longest common substring to process in order to solve the technical problem that the "INS + GPS" integrated navigation system is susceptible to interference and the "INS + terrain matching" integrated navigation system can only be used in areas with specific terrain undulations. However, these algorithms still exist, either the accuracy is not enough or the speed cannot meet the real-time requirements.

[0004] Therefore, how to design a method for quickly and accurately retrieving similar tracks under these difficult conditions has become a technical problem that needs to be solved urgently by those skilled in the art. Summary of the invention

[0005] In order to solve at least one of the technical problems in the prior art that track retrieval has insufficient accuracy and processing speed cannot meet real-time requirements, the present invention provides a track feature modeling method and system, an electronic device, and a non-volatile storage medium for track retrieval. Through a multi-level coding algorithm for earth positions, all positions on the earth are mapped to a certain area code, and then the codes of the areas passed by the aircraft track are strung together, and then a predetermined algorithm is used to find similar tracks.

[0006] On one hand, the present invention provides a track feature modeling method for track retrieval, characterized by comprising:

[0007] Create a coordinate system for the earth's surface and determine different levels of earth area division according to the different precisions of the division areas;

[0008] According to the different precisions of the divided regions, the earth position dictionary code libraries of different levels of precision are established respectively; wherein the earth position dictionary code library of each level of precision sets a number for each region in the earth region according to the corresponding regional precision;

[0009] According to the flight track of the aircraft, from the flight starting point to the flight end point, and connecting each point of the passing area, a track area chain is formed; and the track area chain includes: based on different levels of earth position dictionary code libraries, combining the flight starting point, each point of the passing area, and the flight end point, to form a corresponding flight track;

[0010] When the flight track of the aircraft needs to be queried, in the track area chain, the corresponding flight track is searched from the earth position dictionary code library with low accuracy to the earth position dictionary code library with high accuracy.

[0011] In a preferred implementation manner of an embodiment of the present invention, a flight history track area chain is formed based on the flight history of the aircraft; when an aircraft appears in the area to be monitored, based on the flight track of the aircraft, a method of gradually refining the level of the earth position dictionary code library is used in the flight history track area chain to search whether there is an identical or similar track in the flight history track area chain.

[0012] In a preferred implementation of the embodiment of the present invention, different earth region division levels are determined according to different precisions of the division regions, specifically including:

[0013] Determine the initialization area parameters of the earth region based on different sizes of longitude and latitude;

[0014] The initialized regions are merged according to the maximum weighted population and the maximum area of ​​the region.

[0015] In a further preferred implementation of the embodiment of the present invention, determining the initialization region parameters of the earth region includes:

[0016] Each region is represented by a triplet To express,

[0017] Among them, {(j 1 ,w 1 ),(j 2 ,w 2 )}, indicating the location of the region;

[0018] represents the weighted population of the region;

[0019] And {(j 1 ,w 1 ),(j 2 ,w 2 )} The area calculation of the area corresponding to the region includes:

[0020] Calculate the average latitude of the area as:

[0021] Calculate the area aspect ratio of the region:

[0022] Calculate the actual area of ​​the region: s = |w 1 -w 2 |*|j 1 -j 2 |*cos((w 1 +w 2 ) / 2).

[0023] In a further preferred implementation of the embodiment of the present invention, the initialization regions are merged according to the maximum weighted population of the region and the maximum area of ​​the region, specifically including:

[0024] Determine the world population density based on the number of people per unit land area on the global land surface, and determine the population density corresponding to each area in the earth position dictionary coding library with different levels of accuracy;

[0025] In the track area chain, the population density of the area where the flight starting point and the flight end point of each track are located is increased by a first predetermined value, and the population density of each area traversed by the remaining tracks is increased by a second predetermined value;

[0026] Calculate the weighted area of ​​the region = actual area of ​​the region × (1 + α × log (1 + population density increase predetermined value / regional population));

[0027] Through the square principle and the center-of-gravity centering algorithm, the initialization area of ​​the earth region is merged to obtain the merged earth region.

[0028] In a preferred implementation manner of an embodiment of the present invention, the method for establishing the earth position dictionary code library includes: setting a number for each area, and then applying a general red-black tree algorithm to establish an index to form an earth position dictionary; so that any location in the earth area can directly query the area code through the dictionary index, and based on the area code, the area information can also be directly queried through the dictionary.

[0029] In a preferred implementation of an embodiment of the present invention, the track features include a track area chain, a flight time period, a flight duration, and a flight altitude. The track blockchain includes a flight starting point, a flight end point, the earth area passed by each point, and the area passed by a line connecting two adjacent points; the flight time period is used to record the start time and the end time, the flight duration is used to record the flight duration, and the flight altitude is used to record the maximum altitude, the highest altitude, and the average altitude; the flight speed is used to record the maximum speed, the fastest speed, and the average speed; the track features are used to perform the same or similar track search, and the track retrieval includes using a low-precision area chain for initial screening, and then using a high-precision area chain, combined with the track area chain, flight time period, flight duration, and flight altitude, to determine a similar track.

[0030] The second aspect of the present invention also provides a track feature modeling system for track retrieval, characterized in that it includes:

[0031] The earth region division module is used to create a coordinate system for the earth's surface and determine different earth region division levels according to the different precisions of the division areas;

[0032] The module for establishing the earth position dictionary code library establishes earth position dictionary code libraries of different levels of accuracy according to the different precisions of the divided regions; wherein the earth position dictionary code library of each level of accuracy sets a number for each region in the earth region according to the corresponding regional precision;

[0033] The track area chain forming module is used to form a track area chain from the flight starting point to the flight end point and the point-by-point connection of the passing area according to the flight track of the aircraft; and the track area chain includes: combining the flight starting point, the points in the passing area and the flight end point in the earth position dictionary code library of different levels to form a corresponding flight track;

[0034] When the flight track of the aircraft needs to be queried, in the track area chain, the corresponding flight track is searched from the earth position dictionary code library with low accuracy to the earth position dictionary code library with high accuracy.

[0035] The third aspect of the present invention further provides an electronic device, characterized in that it includes:

[0036] Memory,

[0037] processor, and

[0038] Computer programs;

[0039] Wherein, the computer program is stored in the memory and is configured to be executed by the processor to implement the track feature modeling method for track retrieval as described in any one of the items provided in the first aspect.

[0040] The fourth aspect of the present invention further provides a non-volatile storage medium, characterized in that a computer program is stored thereon; the computer program is executed by a processor to implement any one of the track feature modeling methods for track retrieval provided in the first aspect.

[0041] Therefore, in the above technical solution provided by the present application, all the positions of the earth are mapped to a certain area code through the multi-level coding algorithm of the earth position, and then the codes of the areas through which the aircraft track passes are strung together to form a track area chain that is easy to identify. In addition, in the process of track identification, since the earth position coding adopts a multi-level algorithm, similar tracks can be found through a gradually refined method, which greatly improves the retrieval efficiency; for example, the most coarsely divided regional dictionary is first used to find the qualified track set through preliminary screening, and then the refined regional dictionary is gradually used to finally accurately locate the most similar track.

[0042] Other features and advantages of the invention will be described in the following description, and partly become apparent from the description, or understood by implementing the technical solution of the invention. The purpose and other advantages of the invention can be achieved and obtained by the structures and / or processes specifically pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flow chart of a track feature modeling method for track retrieval is provided for an embodiment of the present invention.

[0044] Figure 2 A flowchart of a method for encoding a dictionary of earth positions is provided for an embodiment of the present invention.

[0045] Figure 3 A schematic diagram of a method for merging regions in a terrestrial location is provided for an embodiment of the present invention.

[0046] Figure 4 A schematic diagram of track feature sample data is provided for an embodiment of the present invention.

[0047] Figure 5 A structural block diagram of a track feature modeling device for track retrieval is provided for an embodiment of the present invention.

[0048] Figure 6 A structural block diagram of an electronic device is provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will describe the implementation methods of the present invention in detail with reference to the accompanying drawings and embodiments, so that the implementation process of how the present invention applies technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly. It should be noted that these specific descriptions are only to make it easier and clearer for ordinary technicians in the field to understand the present invention, rather than a restrictive interpretation of the present invention; and as long as there is no conflict, the various embodiments of the present invention and the various features in the embodiments can be combined with each other, and the technical solutions formed are all within the protection scope of the present invention.

[0050] Additionally, the steps shown in the flowcharts of the accompanying drawings may be executed in a control system such as a set of controller executable instructions, and although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that shown here.

[0051] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments:

[0052] Example

[0053] In order to solve at least one of the technical problems in the prior art that track retrieval is not accurate enough and the processing speed cannot meet the real-time requirements, the technical solution provided in this embodiment maps all the positions of the earth to a certain area code through a multi-level coding algorithm for the earth position, and then strings together the codes of the areas passed by the aircraft track, and then uses a predetermined algorithm to find similar tracks.

[0054] Among them, the track retrieval provided in this embodiment includes but is not limited to the track retrieval corresponding to civil aviation, unmanned aerial vehicles and / or reconnaissance aircraft; and the places where the track of the aircraft mentioned in this embodiment passes can also be combined with the specific type of aircraft, using different signal acquisition methods, such as GPS, Beidou, ADS-B and other types of signal trackers, and the position of the aircraft is determined based on these signal trackers, and then the position of the aircraft is combined with the track feature modeling method mentioned below to determine the flight track of the aircraft. And further, it can also be combined with the historical flight track of the aircraft or the flight track allowed for each aircraft to set up a track area chain. When the flight track of a new aircraft is monitored, it can be quickly determined whether the new aircraft corresponds to the flight track or the flight track allowed for each aircraft; to achieve effective monitoring of the flight trajectory of the aircraft in the predetermined airspace; and it can also focus on aircraft that do not meet the requirements. If there is a situation that affects the normal flight of other aircraft or appears in a flight restricted area, an early warning will be issued.

[0055] like Figure 1 As shown, this embodiment provides a track feature modeling method for track retrieval, and the track feature modeling method includes:

[0056] S110. Create a coordinate system for the earth's surface, and determine different earth region division levels according to the different precisions of the division regions;

[0057] S120. According to the different precision of the divided regions, the earth position dictionary code library with different levels of precision is established respectively; wherein the earth position dictionary code library with each level of precision is respectively set with a number for the corresponding regional precision;

[0058] S130. According to the flight track of the aircraft, from the flight starting point to the flight end point, and connecting the points of the passing area one by one, a track area chain is formed; and the track area chain includes: based on different levels of earth position dictionary code libraries, combining the flight starting point, the points of the passing area, and the flight end point, to form a corresponding flight track;

[0059] Among them, when it is necessary to query the flight track of the aircraft, in the track area chain, it is searched from the earth position dictionary code library with low accuracy to the earth position dictionary code library with high accuracy to see whether there is a corresponding flight track.

[0060] The earth position coding dictionary mentioned in this embodiment divides all locations on the earth into a certain area through a reasonable area segmentation method. The basis for the division includes factors such as population density, data in the track library, airport coordinates, etc.

[0061] Taking 4-level coding as an example, you can also set other levels of coding according to your needs. The earth's surface area is calculated as 510 million square kilometers. This embodiment uses 4-level coding during verification:

[0062] (1) The first-level code intends to divide the earth into no more than 10,000 regions (the average region size is approximately a square with a side length of 225 km).

[0063] (2) The secondary code intends to divide the earth into no more than 100,000 regions (the average region size is approximately a square with a side length of 71 km).

[0064] (3) The third-level coding intends to divide the earth into no more than 1,000,000 regions (the average region size is approximately a square with a side length of 23 km).

[0065] (4) The four-level coding intends to divide the earth into no more than 10,000,000 regions (the average region size is approximately a square with a side length of 7 km).

[0066] The track feature mentioned in this embodiment is to express a track using a set of measurable character strings, numerical values, etc. The preferred track feature design meets the following two points:

[0067] A: If two tracks are similar, then the features must be similar (the feature distance is close to zero);

[0068] B: If there are similar tracks in the track library, and N tracks are obtained through track feature comparison, then the similar track must be among these N tracks; the higher the ranking of the similar track in the result, the more perfect the track feature design is.

[0069] In a preferred implementation of this embodiment, a flight history track area chain is formed based on the flight history of the aircraft; when an aircraft appears in the area to be monitored, based on the flight track of the aircraft, a method of gradually refining the level of the earth position dictionary code library is used to search in the flight history track area chain to see whether there is an identical or similar track.

[0070] In a preferred implementation of this embodiment, different earth region division levels are determined according to different precisions of the division regions, specifically including:

[0071] Determine the initialization area parameters of the earth region based on different sizes of longitude and latitude;

[0072] The initialized regions are merged according to the maximum weighted population and the maximum area of ​​the region.

[0073] In a further preferred implementation of the embodiment of the present invention, determining the initialization region parameters of the earth region includes:

[0074] Each region is represented by a triplet To express,

[0075] Among them, {(j 1 ,w 1 ),(j 2 ,w 2 )}, indicating the location of the region;

[0076] represents the weighted population of the region;

[0077] And {(j 1 ,w 1 ),(j 2 ,w 2 )} The area calculation of the area corresponding to the region includes:

[0078] Calculate the average latitude of the area as:

[0079] Calculate the area aspect ratio of the region:

[0080] Calculate the actual area of ​​the region: s = |w 1 -w2 |*|j 1 -j 2 |*cos((w 1 +w 2 ) / 2).

[0081] In a further preferred implementation of the embodiment of the present invention, the initialization regions are merged according to the maximum weighted population of the region and the maximum area of ​​the region, specifically including:

[0082] Determine the world population density based on the number of people per unit land area on the global land surface, and determine the population density corresponding to each area in the earth position dictionary coding library with different levels of accuracy;

[0083] In the track area chain, the population density of the area where the flight starting point and the flight end point of each track are located is increased by a first predetermined value, and the population density of each area traversed by the remaining tracks is increased by a second predetermined value;

[0084] Calculate the weighted area of ​​the region = actual area of ​​the region × (1 + α × log (1 + population density increase predetermined value / regional population));

[0085] Through the square principle and the center-of-gravity centering algorithm, the initialization area of ​​the earth region is merged to obtain the merged earth region.

[0086] In a preferred implementation of this embodiment, the method for establishing the earth position dictionary code library includes: setting a number for each area, and then applying a general red-black tree algorithm to establish an index to form an earth position dictionary; so that any location in the earth area can directly query the area code through the dictionary index, and based on the area code, the area information can also be directly queried through the dictionary.

[0087] In this embodiment, the specific implementation of the earth position dictionary encoding is as follows: Figure 2 As shown, including:

[0088] S210. Determine parameters according to the position dictionary classification:

[0089] Before starting this work, it is necessary to establish a coordinate system on the surface of the earth. Since the world's most famous satellite positioning systems (GLONASS, GPS, Beidou, etc.) all use longitude and latitude coordinates, this invention also uses longitude and latitude coordinates, with a longitude range of -180° to 180° and a latitude range of -90° to 90°. However, since the earth is a sphere, if the aspect ratio and area formulas under the classical coordinate system are used, the results obtained are wrong, so these formulas must be corrected in a targeted manner. The specific correction formulas are shown in Section 2.3.

[0090] In this coordinate system, the first step is to determine the parameters, mainly the initialization regional parameters of the earth region and the termination conditions of the regional merging.

[0091] (1) Initialize regional parameters of the Earth region

[0092] In the case of first-level encoding, we divide the image into regions every 0.3°. In this case, 1200x600=720,000 regions are initialized.

[0093] In the case of secondary encoding, we divide the image into regions every 0.1°. In this case, 3600x1800=6.48 million regions are initialized.

[0094] In the case of three-level encoding, we divide one region every 0.03°. In this case, 12000x6000=0.72 billion regions are initialized.

[0095] In the case of four-level encoding, we divide one region every 0.01°. In this case, 36000x18000=648 million regions are initialized.

[0096] (2) Termination conditions for regional merger (including the maximum weighted population of the region and the maximum area limit of the region)

[0097] In the case of first-level coding, the actual area of ​​the largest region is 250,000 square kilometers, and the weighted maximum population is no more than 100 million.

[0098] In the case of secondary coding, the actual area of ​​the largest region is 25,000 square kilometers, and the weighted maximum population is no more than 10 million.

[0099] In the case of three-level coding, the actual area of ​​the largest region is 2,500 square kilometers, and the weighted maximum population is no more than 1 million.

[0100] In the case of four-level coding, the actual area of ​​the largest region is 250 square kilometers, and the weighted maximum population is no more than 10.

[0101] S220. Initialization of Earth Region Division:

[0102] Each area uses Such a triple is represented by {(j 1 ,w 1 ),(j 2 ,w 2 )} indicates the location of the region, Represents the weighted population of the area.

[0103] 1. Calculation of actual area of ​​the region

[0104] The calculation method of regional area. Since the earth is spherical, we can intuitively feel that each area near the equator is larger, and the area near the north and south poles is smaller. Suppose the coordinates of a region are {(j 1 ,w 1 ),(j 2 ,w 2 )},So

[0105] Mean latitude

[0106] Area aspect ratio It can be seen that near the pole, Close to 90°, the aspect ratio is close to 0; near the equator, As it approaches 0°, the latitude and longitude ratio weighting approaches 1, that is, the area initialized near the equator is closest to a square.

[0107] The actual area of ​​the region s=|w 1 -w 2 |*|j 1 -j 2 |*cos((w 1 +w 2 ) / 2)

[0108] 2. About the calculation of weighted population

[0109] The calculation of the weighted population of a region mainly considers two factors. One is the actual population number, which is simulated by multiplying the world population density by the area. We downloaded a database of 78,707 major cities in the world, including latitude and longitude values ​​and population data; the other is combined with the application of the present invention to integrate historical track data into the population number.

[0110] World population density is the number of people per unit area of ​​the global land surface. The unit usually used is people / km 2 Or people / hectare. Using the world population density as an indicator can simply and clearly reflect the differences in population distribution between regions. For example, in 1990, the world average population density was 39 people / km 2 , of which Europe exceeds 10.1 people / km 2 , Oceania is only 3 people / km 2 From the perspective of the entire earth, the world's population density is dominated by low-density areas. A British geographer calculated that without taking into account the polar ice caps, the population density is less than 2 people / km2. 2 The area of ​​Europe and Asia accounts for 60% of the world's land area. From the perspective of continents, the proportion of high-density population in Eurasia is significantly higher than that in other continents, while the proportion of high-density population in Oceania is particularly low, and the population distribution in Europe is the most balanced.

[0111] Combined with the technical solution provided in this embodiment, the historical track data is integrated into the population: for each track, the population density of the area where the starting track and the end track are located is increased by 5 people / km 2 ; For the remaining areas crossed by the track, the population density of each area increases by 1 / km 2 .

[0112] The starting point and the end point are most likely the coordinates of an airport, etc., but sometimes they may be a mobile airport (such as an aircraft carrier).

[0113] 3. Regional weighted area calculation:

[0114] The weighted area of ​​a region takes into account both the actual area of ​​the region and the weighted population of the region.

[0115] For example: Regional weighted area = actual area of ​​the region * (1 + α * log (1 + regional weighted population / 10000))

[0116] In this embodiment, α is 10, that is:

[0117] If the regional population is 1,000, the regional weighted area increases by approximately 40%;

[0118] If the regional population is 10,000 people, the regional weighted area increases by about 3 times;

[0119] If the regional population is 100,000, the regional weighted area increases approximately 10 times;

[0120] If the regional population is 1 million, the weighted area of ​​the region increases 20 times;

[0121] Other quantities are also calculated using the above formula.

[0122] S230. Location area merging:

[0123] Location area merging includes but is not limited to the following three factors:

[0124] 1. Regarding which areas need to be merged, in the technical solution provided in this embodiment, the merged area selection first solves the first problem. The areas that need to be merged most have the following characteristics:

[0125] A. The smaller the weighted area of ​​a region, the more it needs to be merged;

[0126] B. The larger the aspect ratio, the more it needs to be merged;

[0127] C. The more mergeable areas there are around, the more they need to be merged.

[0128] From the above three points, it is not difficult to conclude that the areas near the North and South Poles are merged first, because these areas meet almost all of the above conditions:

[0129] The area is the smallest, the population is small, and there are few aircraft passing through, so the weighted population is also small, and the final weighted area is also the smallest;

[0130] The aspect ratio is the largest among all initialization areas;

[0131] Nearby areas are similar, and there are many areas that can be merged;

[0132] In the technical solution provided in this embodiment, the above three conditional quantitative indicators of each initialized area are calculated, and then the indicators are pseudo-randomized (i.e., a random adjustment of ±1% is made based on the finally calculated indicators), and then the indicators are sorted to generate a queue of areas to be merged, and the areas with the highest sorting will be merged first.

[0133] 2. About how to merge:

[0134] If a region can be merged with other regions, but there are multiple merging schemes, how to choose the merging scheme, such as Figure 3 As shown, the main area can merge the four surrounding areas. How to make a decision?

[0135] In the technical solution provided in this embodiment, there are two merging principles:

[0136] (1) Square principle: Calculate the aspect ratio of each solution after merging. The closer to 1, the better.

[0137] (2) Principle of center of gravity: Calculate the center of gravity of each scheme after merging (calculated based on weighted population density). The closer the center of gravity is to the center, the better.

[0138] 3. Under what circumstances will the merger be stopped? The conditions for terminating the merger are as follows: the termination conditions have actually been determined in the above implementation mode; if the actual area of ​​the top ranked region in the queue of regions to be merged is greater than the threshold or the population of the region is greater than the threshold, the merger will be terminated.

[0139] S240. Regional Division Evaluation:

[0140] The rules for division and evaluation are very simple, that is, the fewer divided areas the better, and the smaller the standard deviation of the regional weighted area the better.

[0141] Among them, regional evaluation value = weighted regional area mean - α * weighted regional area standard deviation

[0142] In this embodiment, α is set to 1.

[0143] Since a pseudo-random strategy is used when generating the queue of regions to be merged, the randomness of each merging process can be guaranteed. After executing several random merges, the region with the highest score is selected as the basis for the dictionary encoding of this level.

[0144] S250. Determine whether the division meets the requirements? Based on the result of the regional division evaluation in S240, determine whether the predetermined requirements are met. If yes, execute S260; otherwise, return to S220 to continue to perform the initialization process of the earth regional division on the region.

[0145] S260. Determine the earth position dictionary code: Through the above implementation, the earth position dictionary codes corresponding to different accuracy levels can be obtained; these earth position dictionary codes of different accuracy levels are combined into an earth position dictionary code library. In each earth position dictionary code, each area is given a number, and then a general red-black tree algorithm is applied to establish an index to form an earth position dictionary; in this way, for any location, the area code can be directly queried through the dictionary index; similarly, through the area code, the area information can also be directly queried through the dictionary.

[0146] In a preferred implementation of this embodiment, the track features include a track area chain, a flight time period, a flight duration, and a flight altitude. The track blockchain includes a flight starting point, a flight end point, the earth area passed by each point, and the area passed by a line connecting two adjacent points; the flight time period is used to record the start time and the end time, the flight duration is used to record the flight duration, and the flight altitude is used to record the maximum altitude, the highest altitude, and the average altitude; the flight speed is used to record the maximum speed, the fastest speed, and the average speed; the track features are used to perform searches for identical or similar tracks, and the track retrieval includes using a low-precision area chain for initial screening, and then using a high-precision area chain to combine the track area chain, the flight time period, the flight duration, and the flight altitude to determine similar tracks.

[0147] Specifically, Figure 4 The figure shows part of the sample data of a track, where the first two columns are the longitude and latitude coordinates, the third column is the altitude, the fourth column is the speed, and the fifth column is the data collection time.

[0148] The track features provided in this embodiment mainly consider factors such as track area chain, flight time period, flight duration, flight altitude, flight speed, etc., among which the track area chain is the key factor to be considered.

[0149] Among them, the calculation of the track area chain is from the starting point to the end point, the area passed by each point, and the area passed by the line connecting the adjacent points. The characteristics of the track area chain: ordered and non-repetitive. Ordered means that the track area chain is an ordered set; non-repetitive means that if the aircraft was in area A at the last moment, if the aircraft is still in area A at this time, no operation will be performed. Only when the position of the aircraft changes to area B at a certain moment, B will be added to the track area chain. The track area chain is generated according to dictionaries of different levels. Each dictionary corresponds to a track area chain. The track area chain generated by the first-level position dictionary is referred to as the first-level area chain; similarly, the track area chain generated by the second-level position dictionary is referred to as the second-level area chain.

[0150] Flight time period, mainly records the start time and end time.

[0151] Flight time, mainly records the flight time.

[0152] Flight altitude, mainly records maximum altitude, highest altitude and average altitude.

[0153] Flight speed, mainly records maximum speed, fastest speed and average speed.

[0154] Based on the above preferred implementation, the method for realizing rapid search of similar tracks includes:

[0155] First, use the first-level regional chain for initial screening, then gradually refine it, and finally compare it with multiple indicators such as the fourth-level regional chain and flight time period, flight duration, flight altitude, flight speed, etc. until similar tracks are found.

[0156] Since the first-level regional chain is used first, which is a rough division with better performance, the comparison range can be quickly narrowed; and the last comparison uses the fourth-level regional chain to assist other features for comparison with high accuracy.

[0157] Here, the similarity distance adopts the weighted Euclidean distance, which is a general algorithm, but this embodiment is not limited thereto. Specifically:

[0158] Assuming there are tracks P and Q, calculate the track similarity distance R PQ The process is as follows:

[0159] R PQ =α 1 D link +α 2 D time +α 3 D cost +α 4 D high +α 5 D speed

[0160] R PQ The smaller it is, the stronger the correlation between the two tracks is.

[0161] Among them, α1~α5 are the weights of each calculation dimension, α1+α2+α3+α4+α5=1.

[0162] D link , represents the regional chain correlation;

[0163] D time , indicating the correlation of flight time periods;

[0164] D cost , indicating the correlation of flight duration segments;

[0165] D high , indicating the flight altitude correlation;

[0166] D speed , indicating the flight speed correlation.

[0167] In this embodiment, the factors considered regarding the relevance of the regional chain include:

[0168] 1. The overlap of the coding set in the region chain,

[0169] 2. The longest common substring of the regional chain,

[0170] 3. Edit distance of the region chain.

[0171] In this embodiment, the factors considered for the relevance of the flight time period include:

[0172] 1. Flight start time (hours, minutes, seconds),

[0173] 2. Flight termination time (hours, minutes, seconds),

[0174] 3. Calculation based on flight periodicity (for example, one flight per Monday, one flight on the 1st of each month).

[0175] In this embodiment, the factor considered regarding the correlation of the flight duration segments is the difference in flight duration.

[0176] In this embodiment, the factors considered regarding the flight altitude correlation include:

[0177] 1. Average flight altitude,

[0178] 2. Standard deviation of flight altitude,

[0179] 3. Flight altitude change trend.

[0180] In this embodiment, the factors considered regarding the relevance of flight speed include:

[0181] 1. Mean flight speed,

[0182] 2. Standard deviation of flight speed,

[0183] 3. Flight speed change trend.

[0184] Therefore, in the above technical solution provided by this embodiment, all the positions of the earth are mapped to a certain area code through the multi-level coding algorithm of the earth position, and then the codes of the areas where the aircraft track passes through are strung together to form a track area chain that is easy to identify. In addition, in the process of track identification, since the earth position coding adopts a multi-level algorithm, similar tracks can be found through a gradually refined method, which greatly improves the retrieval efficiency; for example, the most coarsely divided regional dictionary is first used to find the track set that meets the conditions through preliminary screening, and then the refined regional dictionary is gradually used to finally accurately locate the most similar track.

[0185] like Figure 5 As shown, this embodiment also provides a track feature modeling system 100 for track retrieval, and the track feature modeling system 100 for track retrieval includes:

[0186] The earth region division module 110 is used to create a coordinate system of the earth surface and determine different earth region division levels according to different precisions of the division regions;

[0187] The module 120 for establishing the earth position dictionary code library is used to establish earth position dictionary code libraries of different levels of accuracy according to the different precisions of the divided regions; wherein the earth position dictionary code library of each level of accuracy sets a number for each region in the earth region according to the corresponding regional precision;

[0188] The track area chain forming module 130 is used to form a track area chain from the flight start point to the flight end point and the point-by-point connection of the passing area according to the flight track of the aircraft; and the track area chain includes: combining the flight start point, the points in the passing area and the flight end point in the earth position dictionary code library of different levels to form a corresponding flight track;

[0189] Among them, when it is necessary to query the flight track of the aircraft, in the track area chain, it is searched from the earth position dictionary code library with low accuracy to the earth position dictionary code library with high accuracy to see whether there is a corresponding flight track.

[0190] For the specific implementation of each module in the track feature modeling system 100 for track retrieval, please refer to Figure 1-Figure 4 The description of the track feature modeling method for track retrieval mentioned above will not be repeated here.

[0191] like Figure 6As shown, this embodiment further provides an electronic device 200, which includes:

[0192] Memory 210,

[0193] processor 220, and

[0194] Computer programs;

[0195] The computer program is stored in the memory and is configured to be executed by the processor to implement any of the track feature modeling methods for track retrieval provided above.

[0196] In addition, this embodiment further provides a non-volatile storage medium on which a computer program is stored; the computer program is executed by a processor to implement any of the track feature modeling methods for track retrieval provided above.

[0197] It can be understood by those skilled in the art that the above-mentioned method according to the embodiment of the present invention can be implemented in hardware, firmware, or implemented as software or computer code that can be stored in a recording medium (such as a CD ROM, RAM, floppy disk, hard disk or magneto-optical disk), or implemented as computer code originally stored in a remote recording medium or a non-temporary machine-readable medium downloaded through a network and to be stored in a local recording medium, so that the method described herein can be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor or programmable or dedicated hardware (such as an ASIC, FPGA or SoC). It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by a computer, a processor or hardware, the processing method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the processing shown herein, the execution of the code converts the general-purpose computer into a dedicated computer for executing the processing shown herein.

[0198] Those of ordinary skill in the art will appreciate that the units and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the embodiments of the present invention.

[0199] Finally, it should be noted that the above description is only the best embodiment of the present invention and does not limit the present invention in any form. Any technician familiar with the art can make many possible changes and simple replacements to the technical solution of the present invention by using the above disclosed practices and technical contents without departing from the scope of the technical solution of the present invention, which all belong to the scope of protection of the technical solution of the present invention.

Claims

1. A track feature modeling method for track retrieval, characterized in that: include: A coordinate system of the earth's surface is created, wherein the coordinate system uses longitude and latitude coordinates; and different earth region division levels are determined according to different precisions of the division regions, specifically including: Determine the initialization area parameters of the earth region based on different sizes of longitude and latitude; According to the maximum weighted population of the region and the maximum area of ​​the region, the initialized regions are merged, specifically including: determining the world population density according to the population per unit land area of ​​the global land surface, and determining the population density corresponding to each region in the earth position dictionary code library of different levels of accuracy; in the track area chain, the population density of the area where the flight starting point and the flight end point of each track are located is increased by a first predetermined value, and the population density of each area of ​​the area traversed by the remaining tracks is increased by a second predetermined value; calculating the weighted area of ​​the region = the actual area of ​​the region × (1 + α × log (1 + the predetermined value of the increase in population density / the number of regional population)); merging the initialized regions of the earth region through the square principle and the center of gravity centering algorithm to obtain the merged earth region; According to the different precisions of the divided regions, the earth position dictionary code libraries of different levels of precision are established respectively; wherein the earth position dictionary code library of each level of precision sets a number for each region in the earth region according to the corresponding regional precision; According to the flight track of the aircraft, from the flight starting point to the flight end point, and connecting each point of the passing area, a track area chain is formed; and the track area chain includes: in different levels of earth position dictionary code libraries, respectively based on combining the flight starting point, each point of the passing area, and the flight end point, a corresponding flight track is formed; the characteristics of the flight track include the track area chain, the flight time period, the flight duration, and the flight altitude; When the flight track of the aircraft needs to be queried, in the track area chain, the corresponding flight track is searched from the earth position dictionary code library with low accuracy to the earth position dictionary code library with high accuracy.

2. The track feature modeling method according to claim 1, characterized in that: Also includes: Based on the flight history of the aircraft, a flight history track area chain is formed; When an aircraft appears in the area to be monitored, based on the flight track of the aircraft, in the flight history track area chain, a method of gradually refining the level of the earth position dictionary code library is used to search whether there is an identical or similar track in the flight history track area chain.

3. The track feature modeling method according to claim 1, characterized in that: Determine the earth region initialization region parameters include: Each region is represented by a triple, To express, Among them, {(j1,w1),(j2,w2)}, represents the location of the region; represents the weighted population of the region; And the area calculation of the region corresponding to {(j1,w1),(j2,w2)} includes: Calculate the average latitude of the area as: Calculate the area aspect ratio of the region: Calculate the actual area of ​​the region: s = |w1-w2|*|j1-j2|*cos((w1+w2) / 2).

4. The track feature modeling method according to claim 1, characterized in that: The method for establishing the earth position dictionary code library includes: setting a number for each area, and then applying a general red-black tree algorithm to establish an index to form an earth position dictionary; so that any location in the earth area can directly query the area code through the dictionary index, and based on the area code, the area information can also be directly queried through the dictionary.

5. The track feature modeling method according to any one of claims 1 to 4, characterized in that: The track area chain includes the flight starting point, the flight end point, the earth area passed by each point, and the area passed by the line connecting two adjacent points; the flight time period is used to record the start time and the end time, the flight duration is used to record the flight duration, the flight altitude is used to record the maximum altitude, the highest altitude and the average altitude; the flight speed is used to record the maximum speed, the fastest speed and the average speed; the track characteristics are used to perform the same or similar track search, and the track retrieval includes using the low-precision area chain for initial screening, and then using the high-precision area chain to combine the track area chain, flight time period, flight duration, and flight altitude to determine similar tracks.

6. A track feature modeling system for track retrieval, characterized in that: include: The earth region division module is used to create a coordinate system on the earth surface, wherein the coordinate system adopts longitude and latitude coordinates; According to the different precisions of the divided regions, different earth region division levels are determined, specifically including: determining the parameters of the earth region initialization region based on different sizes of longitude and latitude; merging the initialization regions according to the maximum weighted population of the region and the maximum area of ​​the region, specifically including: determining the world population density according to the population per unit land area of ​​the global land surface, and determining the population density corresponding to each region in the earth position dictionary code library of different levels of precision; in the track region chain, the population density of the region where the flight starting point and the flight end point of each track are located is increased by a first predetermined value, and the population density of each region of the region traversed by the remaining tracks is increased by a second predetermined value; calculating the weighted area of ​​the region = the actual area of ​​the region × (1 + α × log (1 + the predetermined value of the population density increase / the number of regional population)); merging the earth region initialization regions through the square principle and the center of gravity centering algorithm to obtain the merged earth region; The module for establishing the earth position dictionary code library establishes earth position dictionary code libraries of different levels of accuracy according to the different precisions of the divided regions; wherein the earth position dictionary code library of each level of accuracy sets a number for each region in the earth region according to the corresponding regional precision; The track area chain forming module is used to form a track area chain from the flight starting point to the flight end point and the point-by-point connection of the passing area according to the flight track of the aircraft; and the track area chain includes: based on the earth position dictionary code library of different levels, combining the flight starting point, the point-by-point of the passing area, and the flight end point, to form a corresponding flight track; the characteristics of the flight track include the track area chain, the flight time period, the flight duration, and the flight altitude; When the flight track of the aircraft needs to be queried, in the track area chain, the corresponding flight track is searched from the earth position dictionary code library with low accuracy to the earth position dictionary code library with high accuracy.

7. An electronic device, characterized in that: include: Memory, processor, and Computer programs; Wherein, the computer program is stored in the memory and is configured to be executed by the processor to implement the track feature modeling method for track retrieval as described in any one of claims 1-5.

8. A non-volatile storage medium, characterized in that: A computer program is stored thereon; the computer program is executed by a processor to implement a track feature modeling method for track retrieval as claimed in any one of claims 1 to 5.

Images