Aircraft taxiing state identification and aircraft type inversion method

Abstract

The invention relates to an aircraft taxiing state recognition and model inversion method, and belongs to the field of airport intelligent perception, and the aircraft taxiing state recognition comprises the following steps: collecting original wheel track data of aircraft taxiing by using a laser wheel track instrument; key information in the original wheel mark data is extracted, data segmentation is carried out, and a wheel mark data array is formed; and clustering the wheel track data array by adopting a clustering algorithm, and judging the taxiing state of the airplane according to the wheel track data of a clustering center point. And the aircraft type inversion is to calculate the main undercarriage spacing according to the clustering result, and judge the aircraft type according to the main undercarriage spacing. Compared with the prior art, accurate and effective aircraft taxiing state identification and aircraft type inversion can be realized only by adopting one laser sensor.

Description

Technical Field

[0001] The present invention relates to the field of airport intelligent perception, and in particular to a method for aircraft taxiing state recognition and aircraft model inversion. Background Art

[0002] The lateral distribution of wheel tracks refers to the fact that when the aircraft is taxiing on the runway, the wheel tracks of the aircraft always swing left and right within a certain range near the center line of the cross section and are distributed on the cross section of the runway at a certain frequency. The standard deviation of the lateral distribution of wheel tracks is an important parameter in the design of airport runways. The study of the lateral distribution of wheel tracks is of great significance for the analysis of the response, design and evaluation of airport pavement structures. The number of take-offs and landings, taxiing speed, lateral position offset distance and other information of different aircraft types are important bases for airport operation and maintenance. Referring to the aircraft parameter table for pavement design in the "Civil Airport Cement Concrete Pavement Design Specification MHT5004-2010", the maximum take-off weight of the aircraft is greater than the maximum landing weight. Therefore, the study of the taxiing state of the take-off aircraft is more important for the operation and maintenance of airport runways.

[0003] Domestic and foreign technologies mainly use infrared, embedded sensors, video testing technology and laser sensing technology to measure the taxiing trajectory. Among them, infrared technology relies on infrared beams to detect the position of the wheels, but is easily interfered by external heat sources and obstacles; embedded vibration sensors are based on optical fiber technology and are suitable for point-like monitoring of large areas, but they need to be buried in the road surface, which limits their application in service runways, and are costly, short in service life, and low in survival rate; the method of drawing scales on the pavement and video recording interferes with the normal operation of the aircraft.

[0004] Laser testing technology is widely used in airport wheel track data collection due to its advantages of strong anti-interference, high acquisition frequency and no impact on the normal operation of the airport. CN109444907A discloses a test system and method for the lateral distribution change law of aircraft wheel tracks, which arranges laser ranging sensors, time synchronization and data processing units, power supply units, data storage units and computers on the runway. Among them: the laser ranging sensor mainly measures the distance between the near side tire and the laser testing unit when the aircraft is taxiing; the time synchronization and data processing unit is used to screen the test data of the laser ranging sensor, and all the screened valid data are marked with a unified time parameter; the power supply unit provides the laser ranging sensor and the video unit with the power required for the test; the data storage unit is used to store the data processed by the time synchronization and data processing unit; the computer is used to perform later analysis and processing on the data stored in the data storage unit at a specific time. However, this method requires the use of multiple laser ranging sensors to be implemented, which not only increases the sensor layout cost, but also puts forward more requirements on the layout position of the laser sensor. Summary of the invention

[0005] The purpose of the present invention is to provide an aircraft taxiing state recognition and aircraft type inversion method, which uses only one laser sensor to achieve accurate aircraft taxiing state and aircraft type inversion.

[0006] The purpose of the present invention can be achieved by the following technical solutions: A method for identifying an aircraft taxiing state comprises the following steps: A laser wheel track meter is used to collect the original wheel track data of the aircraft taxiing; Extract key information from the original wheel track data and perform data segmentation to form a wheel track data array; The wheel track data array is clustered using a clustering algorithm, and the aircraft taxiing state is determined based on the wheel track data at the cluster center.

[0007] The laser wheel track meter is arranged at the monitoring end section position of the runway.

[0008] The laser wheel track meter emits a high-frequency laser signal. When the aircraft passes through the section where the laser wheel track meter is located at a high speed, the laser signal is cut. The distance from the laser wheel track meter to the aircraft wheel is calculated based on the laser signal to obtain the original wheel track data.

[0009] The calculation method of the distance from the laser wheel tracker to the wheel is: , ,

[0010] in, is the time from when the laser signal is sent out to when it returns. is the time it takes for the laser signal to reach the wheel; v is the laser propagation speed; is the distance between the wheel and the laser wheel track meter, that is, the original wheel track data.

[0011] The method for acquiring the wheel track data array is as follows: extracting three decimal places after the identifier in the original wheel track data output by the laser wheel track meter as valid original wheel track data; subtracting the distance from the laser wheel track meter to the edge of the runway from the valid original wheel track data to obtain the wheel track data; extracting the time data in the brackets in the original wheel track data output by the laser wheel track meter, and saving the time data after matching the time data with the wheel track data, wherein one time data corresponds to a plurality of wheel track data; and segmenting the saved data according to the time data at a preset time interval, dividing the data within a preset time interval into wheel track data of one flight, and each wheel track data of one flight constitutes a wheel track data array.

[0012] The landing gear of the aircraft is of the front three-point type, the left and right wheels of the main landing gear are symmetrically arranged at a certain distance behind the center of mass of the aircraft, and the nose wheel is arranged below the head of the aircraft; when the aircraft passes through the section where the laser wheel track meter is located during the take-off phase, the nose wheel and the main landing gear are both in a grounded state, and the lateral positions of the two are successively captured by the laser wheel track meter, that is, the nose wheel, the left main landing gear wheel and the right main landing gear wheel of the take-off aircraft are captured; during the landing phase, the aircraft has a certain elevation angle, and the main landing gear touches the ground before the nose wheel, that is, the left main landing gear wheel and the right main landing gear wheel of the landing aircraft are captured.

[0013] The clustering algorithm is used to cluster the wheel track data array, and the aircraft taxiing state is determined according to the wheel track data of the cluster center point, specifically: The wheel track data array is input into the K-means clustering algorithm, and each wheel track data array is clustered into three categories to obtain three cluster center points. The wheel track data of the three cluster center points are recorded in order from small to large as the first wheel track data, the second wheel track data and the third wheel track data, and it is determined whether the wheel track data of the three cluster center points meet the first condition. If so, it is determined that the taxiing state of the aircraft is the take-off state; if not, it is determined that the taxiing state of the aircraft is the landing state, wherein the first condition is: the difference between the second wheel track data and the first wheel track data is greater than a preset value and the difference between the third wheel track data and the second wheel track data is also greater than the preset value.

[0014] When the first condition is not met, it is further determined whether the wheel track data of the three cluster center points meet the second condition. If the second condition is met, each wheel track data array is clustered into two categories to obtain two cluster center points, and the wheel track data of the two cluster center points are recorded as the fourth wheel track data and the fifth wheel track data from small to large. If the second condition is not met, the wheel track data array is deleted, wherein the second condition is that the difference between the second wheel track data and the first wheel track data is greater than a preset value or the difference between the third wheel track data and the second wheel track data is greater than a preset value.

[0015] An aircraft model inversion method comprises the following steps: The above method is used to cluster the wheel track data array; Calculate the main landing gear spacing based on the wheel track data of the cluster center point; Determine the aircraft model based on the main landing gear spacing.

[0016] The main landing gear spacing is the difference between the third wheel track data and the first wheel track data; when the second condition is met, the main landing gear spacing is the difference between the fifth wheel track data and the fourth wheel track data.

[0017] Compared with the prior art, the present invention has the following beneficial effects: (1) Low cost: The present invention can use only one laser sensor (laser wheel track meter) to collect data, and realize aircraft taxiing state recognition and aircraft model inversion by analyzing laser signals. Not only does it reduce the number of laser sensors required to one, greatly reducing the required cost, but it also has low requirements for the layout of the laser sensors. It only needs to set the laser wheel track meter at the monitoring end section of the runway. It is not limited by the requirements of different airport designs or power supply locations, and there is no need to consider the influence of the relative positions of multiple laser wheel track meters on the recognition accuracy.

[0018] (2) High recognition accuracy: The present invention can effectively obtain the lateral position and distribution pattern of the aircraft wheel track, as well as the aircraft take-off and landing status and model by performing data segmentation and clustering processing on the collected aircraft wheel track data, and can infer the taxiing speed from this, providing scientific theoretical support for the design of the unit pavement structure thickness of the airport design. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flow chart of the method of the present invention; Figure 2 This is a schematic diagram of the testing principle of the laser wheel track meter; Figure 3 This is a schematic diagram of a civil aircraft landing gear; Figure 4 This is a schematic diagram of the original wheel track data collected by the laser wheel track meter; Figure 5 It is a schematic diagram of the original data extraction and segmentation process of the laser wheel track meter of the present invention; Figure 6 A flow chart of aircraft taxiing state identification according to the present invention; Figure 7 A flowchart for aircraft model inversion of the present invention; Figure 8 A lateral distribution diagram of wheel tracks of takeoffs in an embodiment; Fig. 9 It is a lateral distribution diagram of wheel tracks of landing sorties in one embodiment. DETAILED DESCRIPTION

[0020] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0021] Example 1

[0022] This embodiment provides a method for identifying the taxiing state of an aircraft, using a laser sensor to test the taxiing state of the aircraft. The reason for this is that the aircraft has a deviation angle with the runway centerline when taxiing, making it possible to use a laser sensor to carry out the test. Specifically, Figure 1 As shown in the upper part of , the method comprises the following steps: S1, a laser wheel track meter is used to collect the original wheel track data of the aircraft taxiing.

[0023] In this embodiment, a non-contact laser sensor test device (referred to as a laser wheel track meter) is set up on one side of the runway monitoring end section to collect wheel track data. The laser wheel track meter is a type of laser sensor. It emits a high-frequency laser signal. When the aircraft passes through the section where the laser wheel track meter is located at high speed, the laser signal is cut. Based on the laser signal, the distance from the laser wheel track meter to the aircraft wheel can be calculated to obtain the original wheel track data. Its principle is as follows Figure 2 As shown, the calculation method of the distance from the laser wheel tracker to the wheel is: , ,

[0024] in, is the time from when the laser signal is sent out to when it returns. is the time it takes for the laser signal to reach the wheel; v is the laser propagation speed; is the distance between the wheel and the laser wheel track meter, that is, the original wheel track data.

[0025] The landing gear of modern civil aircraft is mainly of the front three-point type. The left and right wheels of the main landing gear are arranged symmetrically at a certain distance behind the center of mass of the aircraft, and the nose wheel is arranged below the head of the aircraft. Figure 3 As shown in (3a). Since the laser wheel track meter section is at the end of the runway, when the aircraft passes through the laser wheel track meter section during the take-off phase, the nose wheel and the main landing gear are both in the grounded state, and the lateral positions of the two will be captured by the laser wheel track meter in succession, that is, the take-off aircraft will have three types of data captured, namely the nose wheel, the left main landing gear wheel and the right main landing gear wheel; and during the landing phase, the aircraft has a certain pitch angle, and the main landing gear will touch the ground before the nose wheel, that is, the landing aircraft will have two types of data captured, namely the left main landing gear wheel and the right main landing gear wheel, as shown in Figure 3 as shown in (3b).

[0026] S2, extracts key information from the original wheel track data and performs data segmentation to form a wheel track data array.

[0027] The original wheel track data collected by the laser wheel track meter is as follows: Figure 4 As shown. Figure 5As shown in the figure, the formation process of the wheel track data number is as follows: The three decimal places after the identifier "$BM," are valid data. Extract this part of the data as the valid original wheel track data d. Subtract the distance from the laser wheel track instrument to the runway edge from the valid original wheel track data d to obtain the wheel track data x. Extract the time data t within the square brackets [] in the original wheel track data output by the laser wheel track instrument, and save it corresponding to the wheel track data x, where one time data corresponds to multiple wheel track data. According to the time data t, divide the saved data at a preset time interval (3s). Divide the data within one preset time interval into the wheel track data of one flight, and the wheel track data of each flight constitutes a wheel track data array [t, x].

[0028] S3. Use the clustering algorithm to cluster the wheel track data array, and judge the aircraft taxiing state according to the wheel track data of the clustering center point.

[0029] As Figure 6 shown, S3 specifically includes the following steps: Input the wheel track data array [t, x] into the K-means clustering algorithm, cluster each wheel track data array into three categories, and obtain three clustering center points (t1, x1), (t2, x2), (t3, x3). In this embodiment, it is assumed that x1 < x2 < x3. Judge whether x2 - x1 > 2 and x3 - x2 > 2 are satisfied. If satisfied, judge that the taxiing state of this flight of the aircraft is the takeoff state; if not satisfied, judge that the taxiing state of this flight of the aircraft is the landing state. Further judge whether the wheel track data of the three clustering center points satisfy x2 - x1 > 2 or x3 - x2 > 2. If satisfied, cluster each wheel track data array into two categories to obtain two clustering center points (t4, x4), (t5, x5). In this embodiment, it is assumed that x4 < x5; if not satisfied, delete this wheel track data array.

[0030] This embodiment also provides an aircraft model inversion method. As Figure 1 shown in the lower half of the figure, it includes the following steps: A1. Use the above method to cluster the wheel track data array; A2. Calculate the main landing gear spacing based on the wheel track data of the clustering center point; A3. Judge the aircraft model based on the main landing gear spacing.

[0031] As Figure 7 shown, when x2 - x1 > 2 and x3 - x2 > 2 are satisfied, the main landing gear spacing ∆x = x3 - x1, and the time ∆t of this flight passing through the laser wheel track instrument is ∆t = t max -t min; When \(x2 - x1>2\) and \(x3 - x2>2\) are not satisfied, but \(x2 - x1>2\) or \(x3 - x2>2\) is satisfied, the main landing gear spacing \(\Delta x=x5 - x4\).

[0032] After processing the data for each flight, the lateral position distribution law of the aircraft wheel tracks on the airport pavement can be analyzed. Referring to the aircraft parameter table for pavement design in the "Code for Design of Cement Concrete Pavement of Civil Airports MHT5004 - 2010", based on the calculated main landing gear spacing \(\Delta x\), the aircraft type of each flight can be obtained.

[0033] In one embodiment, the distance \(l\) from the nose wheel of the aircraft to the main landing gear can also be obtained according to the aircraft type information. Combining the time \(\Delta t=t\) max -t min that the aircraft passes through the laser wheel track instrument calculated above, the take - off speed \(v = l / \Delta t\) is calculated.

[0034] Embodiment 2 Take the taxiing state and aircraft type identification of a certain actual aircraft during take - off as an example.

[0035] Input the original wheel track data \((t, x)\) of a certain actual aircraft collected by the laser wheel track instrument, where the time data is \(t\) and the wheel track data is \(x\). Using the K - means algorithm, the original wheel track data is clustered into three categories, and three clustering center points \((t1, x1)\), \((t2, x2)\), \((t3, x3)\) (\(x1 < x2 < x3\)) are obtained. For this flight, \(x2 - x1>2\) and \(x3 - x2>2\), it is determined that this flight is an aircraft in the take - off state. The three categories of data are the nose wheel, the left main landing gear, and the right main landing gear of the aircraft respectively. The time \(\Delta t=t\) max -t min that the aircraft passes through the laser wheel track instrument, and the main landing gear spacing \(\Delta x=x3 - x1\). The lateral distribution diagram of the wheel tracks of this take - off flight is as Figure 8 shown. For this flight, the main landing gear spacing \(\Delta x = 5.8m\approx5.72m\), \(\Delta t = 0.38s\), the aircraft type may be B737 - 800 / B737 - 700, and the distance from the nose wheel to the main landing gear, that is, the wheelbase \(l\) is 17.17m, then the take - off speed \(v = l / \Delta t = 45.18m / s\).

[0036] Embodiment 3 Take the taxiing state and aircraft type identification of a certain actual aircraft during landing as an example.

[0037] Input the original wheel track data (t, x) of a certain actual aircraft collected by a laser wheel tracker. The time data is t, and the wheel track data is x. Using the K-means algorithm, the original wheel track data is clustered into three categories, obtaining three cluster centers (t1, x1), (t2, x2), (t3, x3), (x1 < x2 < x3). If this flight does not satisfy x2 - x1 > 2 and x3 - x2 > 2, it is determined that this flight is an aircraft in the landing state. If it satisfies x2 - x1 > 2 or x3 - x2 > 2, the original wheel track data is clustered into two categories using the K-means algorithm, obtaining two cluster centers (t4, x4), (t5, x5), (x4 < x5). The two categories of data are the left main landing gear wheel and the right main landing gear wheel of the aircraft respectively. The distance between the main landing gears of this flight is ∆x = x5 - x4. The lateral distribution diagram of the wheel tracks of this landing flight is as shown in Fig. 9 Figure. The distance between the main landing gears of this flight is ∆x = 9.6m ≈ 9.8m, and the aircraft model may be B787-9.

[0038] The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those of ordinary skill in the art can make many modifications and variations based on the concept of the present invention without creative labor. Therefore, all technical solutions that can be obtained by those skilled in the art in the technical field based on the concept of the present invention through logical analysis, reasoning, or limited experiments on the basis of the prior art shall fall within the protection scope determined by the claims.

Claims

1. A method for identifying an aircraft taxiing state, characterized in that: The following steps are involved: A laser wheel track meter is used to collect the original wheel track data of the aircraft taxiing; Extract key information from the original wheel track data and perform data segmentation to form a wheel track data array; The wheel track data array is clustered using a clustering algorithm, and the aircraft taxiing state is determined based on the wheel track data at the cluster center.

2. The method for identifying an aircraft taxiing state according to claim 1, characterized in that: The laser wheel track meter is arranged at the monitoring end section position of the runway.

3. The method for identifying an aircraft taxiing state according to claim 2, characterized in that: The laser wheel track meter emits a high-frequency laser signal. When the aircraft passes through the section where the laser wheel track meter is located at a high speed, the laser signal is cut. The distance from the laser wheel track meter to the aircraft wheel is calculated based on the laser signal to obtain the original wheel track data.

4. The method for identifying an aircraft taxiing state according to claim 3, characterized in that: The calculation method of the distance from the laser wheel tracker to the wheel is: , , in, is the time from when the laser signal is sent out to when it returns. is the time it takes for the laser signal to reach the wheel; v is the laser propagation speed; is the distance between the wheel and the laser wheel track meter, that is, the original wheel track data.

5. The method for identifying an aircraft taxiing state according to claim 1, characterized in that: The method for acquiring the wheel track data array is as follows: extracting three decimal places after the identifier in the original wheel track data output by the laser wheel track meter as valid original wheel track data; subtracting the distance from the laser wheel track meter to the edge of the runway from the valid original wheel track data to obtain the wheel track data; extracting the time data in the brackets in the original wheel track data output by the laser wheel track meter, and saving the time data after matching the time data with the wheel track data, wherein one time data corresponds to a plurality of wheel track data; and segmenting the saved data according to the time data at a preset time interval, dividing the data within a preset time interval into wheel track data of one flight, and each wheel track data of one flight constitutes a wheel track data array.

6. The method for identifying an aircraft taxiing state according to claim 1, characterized in that: The landing gear of the aircraft is of the front three-point type, the left and right wheels of the main landing gear are symmetrically arranged at a certain distance behind the center of mass of the aircraft, and the nose wheel is arranged below the head of the aircraft; when the aircraft passes through the section where the laser wheel track meter is located during the take-off phase, the nose wheel and the main landing gear are both in a grounded state, and the lateral positions of the two are successively captured by the laser wheel track meter, that is, the nose wheel, the left main landing gear wheel and the right main landing gear wheel of the take-off aircraft are captured; during the landing phase, the aircraft has a certain elevation angle, and the main landing gear touches the ground before the nose wheel, that is, the left main landing gear wheel and the right main landing gear wheel of the landing aircraft are captured.

7. The method for identifying an aircraft taxiing state according to claim 1, characterized in that: The clustering algorithm is used to cluster the wheel track data array, and the aircraft taxiing state is determined according to the wheel track data of the cluster center point, specifically: The wheel track data array is input into the K-means clustering algorithm, and each wheel track data array is clustered into three categories to obtain three cluster center points. The wheel track data of the three cluster center points are recorded in order from small to large as the first wheel track data, the second wheel track data and the third wheel track data, and it is determined whether the wheel track data of the three cluster center points meet the first condition. If so, it is determined that the taxiing state of the aircraft is the take-off state; if not, it is determined that the taxiing state of the aircraft is the landing state, wherein the first condition is: the difference between the second wheel track data and the first wheel track data is greater than a preset value and the difference between the third wheel track data and the second wheel track data is also greater than the preset value.

8. The method for identifying an aircraft taxiing state according to claim 7, characterized in that: When the first condition is not met, it is further determined whether the wheel track data of the three cluster center points meet the second condition. If the second condition is met, each wheel track data array is clustered into two categories to obtain two cluster center points, and the wheel track data of the two cluster center points are recorded as the fourth wheel track data and the fifth wheel track data from small to large. If the second condition is not met, the wheel track data array is deleted, wherein the second condition is that the difference between the second wheel track data and the first wheel track data is greater than a preset value or the difference between the third wheel track data and the second wheel track data is greater than a preset value.

9. An aircraft model inversion method, characterized in that: The following steps are involved: Clustering the wheel track data array using the method described in any one of claims 1 to 8; Calculate the main landing gear spacing based on the wheel track data of the cluster center point; Determine the aircraft model based on the main landing gear spacing.

10. The aircraft model inversion method according to claim 9, characterized in that: When the first condition is met, the main landing gear spacing is the difference between the third wheel track data and the first wheel track data; when the second condition is met, the main landing gear spacing is the difference between the fifth wheel track data and the fourth wheel track data.

Images