Train vehicle positioning method and device

Abstract

The present invention discloses a method and device for positioning a train vehicle, comprising the following steps: taking photos of the target area in real time to identify the track fasteners that the train vehicle passes through while the train vehicle is traveling in the target area; counting the number of currently identified track fasteners; and calculating the travel distance of the train vehicle based on the predetermined average spacing of the track fasteners and the number of track fasteners counted. The positioning method and device of the present invention can accurately calculate the travel distance of the train vehicle, thereby realizing continuous positioning of the train vehicle in the underground tunnel, and the positioning accuracy can reach ±2m, which greatly improves the positioning accuracy of the train vehicle in the underground tunnel.

Description

Technical Field

[0001] The present invention relates to the field of rail vehicle positioning, and in particular to a train vehicle positioning method and device. Background Art

[0002] The positioning technologies for trains in traditional subway systems usually include the following categories:

[0003] Axle encoder detection and positioning: Axle encoders are usually installed on the axles of train vehicles. The axle encoder detects the number of wheel revolutions to calculate the train's travel distance, thereby obtaining the actual position of the train vehicle. However, factors that affect the axle encoder detection error include wheel idling and wheel sliding, which lead to large errors in the train vehicle positioning results, and the positioning error can even reach 30m.

[0004] Train TCMS network communication positioning: This positioning technology mainly obtains the speed and position information of train vehicles through the train TCMS network to realize the positioning of train vehicles. However, since train vehicles in the subway system are usually in underground tunnels, the network signal needs to be forwarded multiple times, resulting in inaccurate positioning of train vehicles, and even the train TCMS network cannot determine the specific location of train vehicles in underground tunnels.

[0005] From the above analysis, it can be seen that the current positioning technology is difficult to achieve accurate positioning of train vehicles in underground tunnels. In view of this, it is necessary to design a positioning method and device that can effectively improve the positioning accuracy of train vehicles in the subway system. Summary of the invention

[0006] The object of the present invention is to provide a train vehicle positioning method and device, which can effectively improve the positioning accuracy of train vehicles in underground tunnels.

[0007] To achieve this object, the present invention adopts the following technical solutions:

[0008] A train vehicle positioning method comprises the following steps:

[0009] When the train is traveling in the target area, take photos of the target area in real time to identify the track fasteners that the train passes through;

[0010] Count the number of currently identified rail fasteners;

[0011] Based on the predetermined average spacing of the rail fasteners and the number of the counted rail fasteners, the travel distance of the train vehicle is calculated.

[0012] Optionally, the step of taking photos of the target area in real time to identify the track fasteners that the train passes through while the train is traveling in the target area may also include:

[0013] Acquire a plurality of sample rail fastener images for deep learning to form a sample set;

[0014] Using Deep CNN to perform deep learning on the sample set, a rail fastener recognition model for identifying rail fasteners in the target area image is obtained;

[0015] The track fasteners for identifying the train vehicle passing through are specifically:

[0016] The linear light source of the train vehicle shines towards the target area to form a straight line of light;

[0017] The rail fastener recognition model recognizes the rail fastener according to the three-dimensional shape formed by the straight line of light in the target area.

[0018] Optionally, the step of taking photos of the target area in real time to identify the track fasteners that the train passes through while the train is traveling in the target area may also include:

[0019] Count the number of track fasteners in the target area and confirm the distance between the departure station and the stop station of the train vehicle;

[0020] Based on the number of rail fasteners in the route and the target area, the average spacing of the rail fasteners between the departure station and the stop station is calculated; wherein the target area is a strip area connecting the departure station and the stop station.

[0021] Optionally, the target area includes a first strip area on one side of the first track and equipped with a track fastener, and a second strip area on one side or the other side of the second track and equipped with a track fastener, wherein the first strip area and the second strip area are both connected to the departure station and the stop station;

[0022] The step of calculating the travel distance of the train vehicle based on the predetermined average spacing of the track fasteners and the number of the track fasteners counted is specifically as follows:

[0023] Calculate the driving distance using the following formula:

[0024] S = m × (n / 2);

[0025] Wherein, m is the average spacing between rail fasteners, and n is the total number of rail fasteners counted.

[0026] Optionally, it also includes:

[0027] After receiving the arrival information, the train vehicle clears the counted number of the track fasteners;

[0028] After the train vehicle starts again, take photos of the new target area in real time to identify the track fasteners that the train vehicle passes through;

[0029] Count the number of currently identified rail fasteners;

[0030] Based on the predetermined average spacing of the rail fasteners and the number of the counted rail fasteners, the travel distance between the train vehicle and the last stop is calculated.

[0031] A train vehicle positioning device, comprising:

[0032] An image acquisition module disposed at the bottom of the train vehicle is used to take pictures of the target area in real time while the train vehicle is traveling in the target area;

[0033] An image processing module for identifying track fasteners that a train vehicle passes by;

[0034] The image processing module is also used to count the number of currently identified track fasteners, and calculate the travel distance of the train vehicle based on a predetermined average spacing of the track fasteners and the number of counted track fasteners;

[0035] The image processing module is electrically connected to the image acquisition module, and the image processing module and the image acquisition module are used to execute any one of the above train vehicle positioning methods.

[0036] Optionally, it also includes a positioning module for obtaining the train vehicle position, arrival information and departure information, and the positioning module is electrically connected to the image processing module.

[0037] Optionally, the image acquisition module includes a linear array camera and a linear light source arranged side by side along the direction of travel of the train, the linear light source is used to emit a linear laser line toward a target area in a direction perpendicular to the tread of the first rail or the second rail, and the linear light formed by the linear light source in the target area is perpendicular to the direction of travel of the train vehicle.

[0038] Optionally, a rail fastener recognition model is stored in the image processing module, and the rail fastener recognition model is a recognition model formed based on deep learning and used to recognize rail fastener images.

[0039] Optionally, the line light source is an infrared light source, and the wavelength of the infrared laser emitted by the infrared light source is 808nm-950nm.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The train vehicle positioning method and device provided in the embodiment of the present invention can detect the number of track fasteners passed by the train vehicle through image recognition technology during the travel of the train vehicle, and calculate the travel distance of the train vehicle based on the detected number of track fasteners and the average spacing between the track fasteners, thereby realizing continuous positioning of the train vehicle in the underground tunnel, and its positioning accuracy can reach ±2m, greatly improving the positioning accuracy of the train vehicle in the underground tunnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0043] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0044] Figure 1 A flow chart of a train vehicle positioning method provided by an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of the structure of a train vehicle positioning device provided by an embodiment of the present invention;

[0046] Figure 3 A side schematic diagram of a train vehicle positioning device provided by an embodiment of the present invention;

[0047] Figure 4 A schematic diagram of the distribution of target areas provided by an embodiment of the present invention.

[0048] Illustrations: 1. Train vehicle; 2. Target area; 21. First strip area; 22. Second strip area; 3. Track fastener; 4. First track; 5. Second track; 6. Image acquisition module; 61. Light source; 611. Linear laser line; 62. Line array camera; 7. Image processing module. DETAILED DESCRIPTION

[0049] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] In the description of the present invention, it should be understood that the terms "upper", "lower", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. It should be noted that when a component is considered to be "connected" to another component, it may be directly connected to the other component or there may be a centrally arranged component at the same time.

[0051] The technical solution of the present invention is further described below with reference to the accompanying drawings and through specific implementation methods.

[0052] Embodiment 1

[0053] See also Figure 1 - Figure 4 The embodiment of the present invention provides a train vehicle positioning method, comprising the following steps:

[0054] S10, when the train vehicle 1 is traveling in the target area 2, taking photos of the target area 2 in real time to identify the track fasteners 3 that the train vehicle 1 passes through;

[0055] S20, counting the number of currently identified track fasteners 3;

[0056] S30. Calculate the travel distance of the train vehicle based on the predetermined average track fastener spacing and the counted number of track fasteners 3.

[0057] Specifically, the intervals of the track fasteners 3 in the prior art are fixed, and generally the intervals between two adjacent track fasteners 3 are 0.6m, 0.65m, 0.7m or 1.0m, etc. Under the premise that the intervals between two adjacent track fasteners 3 are basically constant, the travel distance of the train vehicle 1 can be calculated by counting the number of track fasteners 3 on the same side of a track that the train vehicle 1 passes, thereby determining the specific position of the train vehicle 1. In addition, it should be known that during the travel of the train vehicle 1, the travel distance of the train vehicle can be calculated by calculating the number of track fasteners 3 detected during the continuous travel time, and continuous and accurate positioning of the train vehicle 1 can be achieved in combination with a map, and the positioning accuracy of the train vehicle is within ±2m.

[0058] Further, in one embodiment of the present invention, before the step of: taking photos of the target area 2 in real time during the travel of the train vehicle 1 in the target area 2 to identify the track fasteners 3 through which the train vehicle 1 passes, the method further includes:

[0059] Acquire a plurality of sample rail fastener images for deep learning to form a sample set;

[0060] Using Deep CNN (deep convolutional neural network) to perform deep learning on the sample set, a rail fastener recognition model for identifying the rail fastener 3 in the target area image is obtained;

[0061] The identification of the track fastener 3 through which the train vehicle 1 passes is specifically:

[0062] The line light source 61 of the train vehicle 1 shines toward the target area to form a line of light;

[0063] The rail fastener recognition model recognizes the rail fastener according to the three-dimensional shape formed by the straight line of light in the target area.

[0064] Specifically, Deep CNN is used to perform deep learning on the sample set to obtain a track fastener recognition model for identifying the track fastener 3 image in the target area image, thereby improving the accuracy of the image processing module 7 in identifying the track fastener 3, avoiding the image processing module 7 from missing the track fastener 3, and ensuring the positioning accuracy of the continuous positioning of the train vehicle 1. The trained track fastener recognition model not only recognizes the track fastener 3 in the target area image, but also accumulates the number of track fasteners 3 it recognizes, and calculates the travel distance of the train vehicle 1 by the number of track fasteners 3 counted.

[0065] Specifically, the linear light emitted by the line light source 61 will form a linear "light spot" (linear light) in the target area where there is no rail fastener 3. When the linear "light spot" sweeps over the rail fastener, the shape of the "light spot" will change (a plane straight line becomes a three-dimensional curve), and the rail fastener recognition model recognizes the rail fastener based on the shape of the light spot.

[0066] Furthermore, in one embodiment of the present invention, the step of taking photos of the target area in real time and identifying the track fasteners passed by the train vehicle during the travel of the train vehicle in the target area also includes:

[0067] Counting the number of track fasteners 3 in the target area 2 to confirm the distance between the departure station and the stop station of the train vehicle 1;

[0068] The average fastener spacing between the departure station and the stop station is calculated based on the distance and the number of fasteners in the target area 2. The target area 2 is a strip area connecting the departure station and the stop station. That is, during the travel of the train vehicle 1, the train vehicle 1 continuously takes images of the strip area below it to detect the track fasteners 3 in the target area 2 passed by the train vehicle, and counts the number of the detected track fasteners 3.

[0069] Further, in one embodiment of the present invention, the target area 2 includes a first strip area 21 on one side of the first track 4 and installed with a track fastener 3, and a second strip area 22 on one side or the other side of the second track 5 and installed with a track fastener 3, wherein the first strip area 21 and the second strip area 22 are both connected to the departure station and the stop station;

[0070] The step of calculating the travel distance of the train vehicle based on the predetermined average spacing of the track fasteners and the number of the track fasteners counted is specifically as follows:

[0071] Calculate the driving distance using the following formula:

[0072] S = m × (n / 2);

[0073] Among them, m is the average fastener spacing, n is the total number of track fasteners 3 counted, that is, n is the sum of the number X of track fasteners in the first strip area and the number Y of track fasteners in the second strip area identified by the image processing module 7.

[0074] It should be known that the running route of the train vehicle may include multiple curves. If only the area on one side of the first track 4 or the second track 5 where the track fasteners 3 are installed is set as the target area 2, the positioning accuracy of the train vehicle 1 will decrease when the train vehicle 1 passes through the curve. Therefore, the area on one side of the first track 4 where the track fasteners 3 are installed is set as the first strip area 21, and the area on one side (or the other side) of the second track 5 where the track fasteners 3 are installed is set as the second strip area 22. At this time, if the number of track fasteners 3 detected in the left area of ​​the first track 4 (i.e., the first strip area 21) is X and the number of track fasteners 3 detected in the left side of the second track 5 (i.e., the second strip area 22) is Y, calculate the sum n of X and Y, and then calculate the travel distance of the train vehicle 1 by the formula S=m×(n / 2). It should be known that using (X+Y) / 2 as the value of the track fastener 3 for calculating the travel distance of the train vehicle 1 is beneficial to reducing the impact of curves on the positioning accuracy of the train vehicle and improving the positioning accuracy of the train vehicle 1.

[0075] Furthermore, in one embodiment of the present invention, the method for positioning the train vehicle 1 further includes the step of clearing the counted number of the track fasteners 3 after the train vehicle 1 receives the arrival information;

[0076] After the train vehicle starts again, take photos of the new target area in real time to identify the track fasteners that the train vehicle passes through;

[0077] The number of currently identified track fasteners is counted; and the travel distance between the train vehicle and the last stop is calculated based on a predetermined average track fastener spacing and the counted number of track fasteners.

[0078] It should be known that the positioning method is used to calculate the position of the train vehicle 1 between two adjacent stations, and clear the counted number of track fasteners 3 after the train vehicle 1 arrives at the station. After the train vehicle 1 departs from the new departure station, the number of track fasteners 3 is restarted to avoid the superposition of positioning accuracy errors of the train vehicle between two different routes, thereby reducing the positioning accuracy of the train vehicle in the underground tunnel.

[0079] Embodiment 2

[0080] See also Figure 1 - Figure 4 The present invention discloses a train vehicle positioning device, comprising:

[0081] An image acquisition module 6 disposed at the bottom of the train vehicle 1 is used to take pictures of the target area in real time when the train vehicle is traveling in the target area;

[0082] An image processing module 7, used to identify track fasteners that a train vehicle passes by;

[0083] The image processing module 7 is also used to count the number of currently identified track fasteners, and calculate the travel distance of the train vehicle based on the predetermined average spacing of the track fasteners and the number of counted track fasteners;

[0084] The image processing module 7 is electrically connected to the image acquisition module 6, and the image processing module 7 and the image acquisition module 6 are used to execute the train vehicle positioning method described in any one of the first embodiments.

[0085] Furthermore, in one embodiment of the present invention, the train vehicle positioning device also includes a positioning module for acquiring the train vehicle position, arrival information and departure information, and the positioning module is electrically connected to the image processing module 7 .

[0086] Specifically, the positioning module is used to obtain the current position information, arrival information, exit information, etc. of the train through the train TCMS network. In addition, the arrival information and exit information can also be linked to the door switch of the train vehicle 1. After determining that the train vehicle 1 has arrived at the station, the positioning module sends an arrival signal to the image processing module 7, and the image processing module 7 clears the number of track fasteners 3 counted after receiving the arrival signal. Similarly, after determining that the train vehicle 1 has left the station, the positioning module sends a departure signal to the image processing module 7, and the image processing module 7 starts to count the number of track fasteners 3 in the target area image collected by the image acquisition module 6 after receiving the arrival signal. After the train vehicle 1 departs from the new station, it starts to count the number of track fasteners 3 again to avoid the superposition of positioning accuracy errors between two different routes, thereby reducing the positioning accuracy of the train vehicle in the underground tunnel.

[0087] Further, in one embodiment of the present invention, the image acquisition module 6 includes a linear array camera 62 and a linear light source 61 arranged side by side along the traveling direction of the train, and the linear light source 61 is used to emit a straight laser line 611 toward the target area 2 in a direction perpendicular to the tread of the first track 4 or the second track 5, and the straight laser line 611 formed by the linear light source 61 in the target area 2 is perpendicular to the traveling direction of the train vehicle 1. It should be clear that during the traveling of the train vehicle 1, the straight laser line 611 will sweep across the track fastener 3, and the spatial shape of the light spot (bright area) formed by the straight laser line 611 when sweeping across the track fastener 3 is different from the spatial shape of the light spot (bright area) formed when the straight laser line 611 does not sweep across the track fastener 3. After deep learning, the image processing module 7 recognizes the different light spot shapes of the straight laser line 611 in the target area image, and recognizes the track fastener 3 based on the different light spot shapes. It should be known that the line light source 61 of the present invention does not need to illuminate the entire area, and only needs to scan a straight laser line across the track fastener 3, which can effectively reduce the energy consumption of the line light source 61. In addition, it should be clear that the imaging line of the line array camera 62 coincides with the lateral center of the line light source (forming a straight line of light in the target area 2).

[0088] Furthermore, in one embodiment of the present invention, a rail fastener recognition model is stored in the image processing module 7, and the rail fastener recognition model is a recognition model for recognizing images of rail fasteners 3 formed based on deep learning.

[0089] Furthermore, in one embodiment of the present invention, the line light source 61 is an infrared light source, and the wavelength of the infrared laser emitted by the infrared light source is 808nm to 950nm. The startup process of the infrared light source 61 is short, that is, it can quickly reach a stable radiation state after being powered on. In addition, the infrared light source 61 has the advantages of high radiation efficiency, small size, light weight and long life, which can effectively improve the accuracy of identifying the track fastener 3.

[0090] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A train vehicle positioning method, characterized in that: The following steps are involved: When the train is traveling in the target area, take photos of the target area in real time to identify the track fasteners that the train passes through; Count the number of currently identified rail fasteners; Calculating the travel distance of the train vehicle based on a predetermined average spacing of the rail fasteners and the number of the counted rail fasteners; The step of taking photos of the target area in real time and identifying the track fasteners that the train passes through while the train is traveling in the target area also includes: Count the number of track fasteners in the target area and confirm the distance between the departure station and the stop station of the train vehicle; Based on the number of rail fasteners in the route and the target area, the average spacing of the rail fasteners between the departure station and the stop station is calculated; wherein the target area is a strip area connecting the departure station and the stop station; The target area includes a first strip area on one side of the first track and a track fastener installed thereon, and a second strip area on one side or the other side of the second track and a track fastener installed thereon, wherein the first strip area and the second strip area are both connected to the departure station and the stop station; The step of calculating the travel distance of the train vehicle based on the predetermined average spacing of the track fasteners and the number of the track fasteners counted is specifically as follows: Calculate the driving distance using the following formula: S = m × (n / 2); Wherein, m is the average spacing between rail fasteners, and n is the total number of rail fasteners counted.

2. The train vehicle positioning method according to claim 1, characterized in that: The step of taking photos of the target area in real time and identifying the track fasteners that the train passes through while the train is traveling in the target area also includes: Acquire a plurality of sample rail fastener images for deep learning to form a sample set; Using Deep CNN to perform deep learning on the sample set, a rail fastener recognition model for identifying rail fasteners in the target area image is obtained; The track fasteners for identifying the train vehicle passing through are specifically: The linear light source of the train vehicle shines towards the target area to form a straight line of light; The rail fastener recognition model recognizes the rail fastener according to the three-dimensional shape formed by the straight line of light in the target area.

3. The train vehicle positioning method according to claim 1, characterized in that: Also includes: After receiving the arrival information, the train vehicle clears the counted number of the track fasteners; After the train vehicle starts again, take photos of the new target area in real time to identify the track fasteners that the train vehicle passes through; Count the number of currently identified rail fasteners; Based on the predetermined average spacing of the rail fasteners and the number of the counted rail fasteners, the travel distance between the train vehicle and the last stop is calculated.

4. A train vehicle positioning device, characterized in that: include: An image acquisition module disposed at the bottom of the train vehicle is used to take pictures of the target area in real time while the train vehicle is traveling in the target area; An image processing module for identifying track fasteners that a train vehicle passes by; The image processing module is also used to count the number of currently identified track fasteners, and calculate the travel distance of the train vehicle based on a predetermined average spacing of the track fasteners and the number of counted track fasteners; The image processing module is electrically connected to the image acquisition module, and the image processing module and the image acquisition module are used to execute the train vehicle positioning method according to any one of claims 1-3.

5. The train vehicle positioning device according to claim 4, characterized in that: It also includes a positioning module for obtaining the train vehicle position, arrival information and departure information, and the positioning module is electrically connected to the image processing module.

6. The train vehicle positioning device according to claim 5, characterized in that: The image acquisition module includes a linear array camera and a linear light source arranged side by side along the direction of train travel. The linear light source is used to emit a straight laser line toward a target area in a direction perpendicular to the tread of the first track or the second track, and the straight light formed by the linear light source in the target area is perpendicular to the direction of travel of the train vehicle.

7. The train vehicle positioning device according to claim 4, characterized in that: The image processing module stores a rail fastener recognition model, which is a recognition model formed based on deep learning and used to recognize rail fastener images.

8. The train vehicle positioning device according to claim 6, characterized in that: The line light source is an infrared light source, and the wavelength of the infrared laser emitted by the infrared light source is 808nm-950nm.

Images