An automatic centering method and system for an orbital flaw detection vehicle

Abstract

The present invention discloses an automatic centering method and system for an orbital flaw detection vehicle. The method includes the following steps: S1: Collect image data in the moving direction of the orbital flaw detection vehicle, identify and obtain the data of the traveling section of the flaw detection rail vehicle, and real-time obtain the attitude data of the current flaw detection wheel of the orbital flaw detection vehicle; S2: According to the attitude data, the centering mechanism of the orbital flaw detection vehicle is controlled in real time to perform centering adjustment on the flaw detection wheel. Among them, according to the data of the traveling section, the line change of the flaw detection rail vehicle during traveling is predicted in advance, and the adjustment control instruction of the flaw detection wheel is set in advance based on the line change, so that when the orbital flaw detection vehicle travels to the position of the line change, the centering mechanism is directly controlled to perform centering adjustment on the flaw detection wheel. The present invention not only performs real-time centering adjustment, but also can prepare for centering adjustment in advance according to the line change, greatly improving the flaw detection work efficiency and flaw detection accuracy.

Description

Technical Field

[0001] The present invention belongs to the technical field of track flaw detection, and particularly relates to an automatic centering method and system for a track flaw detection vehicle. Background Art

[0002] Nowadays, most of the domestic track flaw detection systems use a double-track flaw detection vehicle driven manually for flaw detection. Before flaw detection, manual centering operation of the detection wheels is required. After observing the ultrasonic feedback signal situation and repeatedly adjusting to determine the position of the detection wheels, the flaw detection task can be started. And during the operation, due to route changes, the relative position between the detection wheels and the rail will shift, and the position of the detection wheels needs to be adjusted in time. This method takes a long time, has low efficiency, and has a relatively high operation difficulty.

[0003] In the prior art, the patent with the patent number "CN206113874U" and the name "Offset detection device, automatic centering device and water wheel support of water wheel" discloses an offset detection device of a water wheel, including a laser emitter and a detector installed on a water wheel support. The laser emitter emits a laser beam with a certain width, and this width ensures that the laser beam can irradiate one side of the bottom foot of the I-shaped rail and the ground adjacent to this side. The laser on the upper surface of the bottom foot of the I-shaped rail and the laser hitting the ground form a turning point. The detector detects the position of this turning point, that is, the side boundary position of the bottom foot of the I-shaped rail. When the water wheel moves, the position of this turning point will move, and this movement amount is the offset amount between the water wheel and the center line of the rail. The upper computer is connected to the detector. It has the following advantages: It does not require manual adjustment and intervention, is efficient and accurate, and does not affect the detection of the rail.

[0004] The above prior art uses a laser emitter to detect the position of the rail bottom corner to judge the offset amount of the wheel-type probe. This method can only detect the offset distance during actual operation, rather than the most suitable position of the wheel-type probe on the line. The detection effect still depends on the initial debugging and positioning of the wheel-type probe. At the same time, this effect can be achieved by a displacement sensor, without the need for a laser emitter and the algorithm for calculating the turning point position, saving costs and simplifying the equipment. In addition, there are also the following defects:

[0005] 1) During the operation of the equipment, the centering mechanism requires technical personnel to watch the ultrasonic image for adjustment, which takes a long time, has low efficiency, and has relatively high requirements for the ability of the staff;

[0006] 2) The method of using a laser emitter to detect and calculate the offset amount of the wheel-type probe can only adjust the wheel-type probe to the position before offset, but not necessarily the best detection position;

[0007] 3) During the operation of the equipment, due to different line conditions, the position of the centering mechanism needs to be adjusted. However, since the equipment is in a moving state, it is impossible to achieve real-time and accurate adjustment manually. After the adjustment is completed, the equipment has often traveled a certain distance, resulting in a certain data error. Summary of the Invention

[0008] The present invention provides an automatic centering system for an orbital flaw detection vehicle to solve the above technical problems.

[0009] To solve the above problems, the technical solution of the present invention is as follows:

[0010] An automatic centering method for an orbital flaw detection vehicle, comprising the following steps:

[0011] S1: Collect image data in the moving direction of the orbital flaw detection vehicle, identify and obtain the travel section data of the flaw detection rail vehicle, and real-time obtain the attitude data of the current flaw detection wheel of the orbital flaw detection vehicle;

[0012] S2: According to the attitude data, real-time control the centering mechanism of the orbital flaw detection vehicle to adjust the centering of the flaw detection wheel. Among them, according to the travel section data, predict in advance the line change of the flaw detection rail vehicle, and based on the line change, set in advance the adjustment control instruction of the flaw detection wheel, so that when the orbital flaw detection vehicle travels to the position of the line change, directly control the centering mechanism to adjust the centering of the flaw detection wheel.

[0013] In one embodiment, further comprising in identifying and obtaining the travel section data of the flaw detection rail vehicle:

[0014] Extract the track line in the image data through image recognition;

[0015] Calculate the distance D between the curved track and the straight track according to the pixel points of the track line at the preset distance d in the image data, and calculate the curvature radius R of the curved track according to the formula D 2 -2RD + d 2 = 0.

[0016] In one embodiment, further comprising in real-time obtaining the attitude data of the current flaw detection wheel of the orbital flaw detection vehicle:

[0017] Measure the attitude data of the current flaw detection wheel of the orbital flaw detection vehicle in real-time through the cooperation of an inertial navigation module and a gyroscope, wherein the attitude data includes the displacement and tilt angle of the flaw detection wheel.

[0018] In one embodiment, step S2 further comprises:

[0019] S21: Calculate the superelevation value C of the track based on the displacement and tilt angle of the flaw detection wheel:

[0020]

[0021] Wherein, α is the tilt angle of the flaw detection wheel, dR is the displacement between the right rail apex and the vehicle body, and dL is the displacement between the left rail apex and the vehicle body;

[0022] S22: Calculate according to the radius of curvature R and the superelevation value C to obtain the adjustment offset D of the flaw detection wheel:

[0023] D = 10000C / R

[0024] Among them, the change of the line where the flaw detection rail vehicle travels is predicted in advance according to the change of the radius of curvature R, and the radius of curvature in the control of the adjustment offset D is set in advance based on the line change, so that when the rail flaw detection vehicle travels to the position of the line change, the centering mechanism is directly controlled according to the superelevation value to perform centering adjustment on the flaw detection wheel.

[0025] An automatic centering system for a rail flaw detection vehicle, wherein a flaw detection wheel and a centering mechanism for centering adjustment of the flaw detection wheel are carried on the rail flaw detection vehicle, and the system includes: an attitude detection module, an image recognition module, and a processing controller, and the processing controller is respectively in signal connection with the attitude detection module and the image recognition module;

[0026] The attitude detection module is used to obtain the attitude data of the current flaw detection wheel of the rail flaw detection vehicle in real time, and the processing controller is used to control the centering mechanism to perform centering adjustment on the flaw detection wheel in real time according to the attitude data;

[0027] The image recognition module is used to collect image data in the moving direction of the rail flaw detection vehicle, identify and obtain the data of the traveling section of the flaw detection rail vehicle, and the processing controller is also used to predict in advance the line change of the flaw detection rail vehicle according to the traveling section data, and set in advance the adjustment control instruction of the flaw detection wheel based on the line change, so that when the rail flaw detection vehicle travels to the position of the line change, the centering mechanism is directly controlled to perform centering adjustment on the flaw detection wheel.

[0028] In one embodiment, the attitude detection module includes an inertial navigation module and a gyroscope, and the inertial navigation module cooperates with the gyroscope to jointly measure the attitude data of the current flaw detection wheel of the rail flaw detection vehicle in real time, wherein the attitude data includes the displacement and tilt angle of the flaw detection wheel.

[0029] In one embodiment, it further includes: a remote intervention module, which is remotely communicatively connected with the processing controller and is used to remotely control the centering mechanism to perform centering adjustment on the flaw detection wheel.

[0030] In one embodiment, it further includes: a on-site intervention module, which is directly in signal connection with the processing controller and is used to on-site control the centering mechanism to perform centering adjustment on the flaw detection wheel.

[0031] In one embodiment, it further includes a displacement sensor signal - connected to the processing controller. The displacement sensor is arranged in the centering mechanism and is used to feedback the current position data of the centering mechanism in real - time.

[0032] In one embodiment, an orbit recognition model trained based on images is provided in the image acquisition module. It is used to extract the orbit line in the image data through image recognition, calculate the distance D between the curved orbit and the straight orbit according to the pixel points of the orbit line at a preset distance d in the image data, and calculate the curvature radius R of the curved orbit according to the formula D 2 -2RD + d 2 = 0.

[0033] The present invention has the following advantages and positive effects compared with the prior art:

[0034] The present invention truly realizes the automatic adjustment of the centering mechanism through attitude detection and image recognition. Among them, attitude detection can obtain the attitude data of the flaw detection wheel in real - time, and the superelevation value of the track can be detected according to the attitude data. Image recognition can perform line analysis in advance, predict line changes, obtain the data of the traveling section of the flaw detection wheel, and the curvature radius of the track can be detected according to the traveling section data. Based on the superelevation value and the curvature radius, it can be deduced how much distance the current centering position should be offset to achieve a better flaw detection effect. In this way, when the track flaw detection vehicle travels to the corresponding position, it can directly adjust the centering mechanism to the corresponding optimal position according to the curvature radius predicted in advance combined with the superelevation value detected in real - time, rather than starting to analyze and adjust the centering when the track flaw detection vehicle reaches the line change position, avoiding the error of some sections caused by untimely adjustment. The cooperation of attitude detection and image recognition greatly improves the flaw detection work efficiency and flaw detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention.

[0036] Figure 1 It is a schematic flow chart of an automatic centering method for a track flaw detection vehicle of the present invention;

[0037] Figure 2 It is a schematic diagram of the curve radius detection principle of an automatic centering method for a track flaw detection vehicle of the present invention;

[0038] Figure 3 It is a schematic diagram of the superelevation value detection principle of an automatic centering method for a track flaw detection vehicle of the present invention;

[0039] Figure 4Schematic structural diagram of an automatic centering system for an orbital flaw detection vehicle according to the present invention;

[0040] Figure 5 Schematic working diagram of an automatic centering system for an orbital flaw detection vehicle according to the present invention. Specific embodiments

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the specific embodiments of the present invention will be described below with reference to the accompanying drawings. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts, and other embodiments can also be obtained.

[0042] To make the drawings concise, only the parts related to the present invention are schematically shown in each drawing, and they do not represent the actual structure of the product. In addition, to make the drawings concise and easy to understand, in some drawings, components with the same structure or function are only schematically shown for one of them, or only one of them is marked. In this article, "one" not only means "only this one", but also means "more than one" situation.

[0043] The following further describes in detail a method and system for automatic centering for an orbital flaw detection vehicle proposed by the present invention in conjunction with the accompanying drawings and specific embodiments.

[0044] Refer to Figure 1 , this application provides a method for automatic centering for an orbital flaw detection vehicle, including the following steps:

[0045] S1: Collect image data in the moving direction of the orbital flaw detection vehicle, identify and obtain the data of the traveling section of the flaw detection vehicle, and real-time obtain the attitude data of the current flaw detection wheel of the orbital flaw detection vehicle;

[0046] S2: According to the attitude data, real-time control the centering mechanism of the orbital flaw detection vehicle to adjust the centering of the flaw detection wheel. Among them, according to the data of the traveling section, predict in advance the line change of the flaw detection vehicle, and based on the line change, set in advance the adjustment control instruction of the flaw detection wheel, so that when the orbital flaw detection vehicle travels to the position of the line change, directly control the centering mechanism to adjust the centering of the flaw detection wheel.

[0047] The present embodiment will now be described in detail, but not limited thereto.

[0048] This embodiment is applicable to the automatic centering of railway line flaw detection equipment to achieve the position alignment between the railway track and the flaw detection equipment, thereby ensuring the best flaw detection effect. Among them, the flaw detection vehicle in this embodiment is equipped with flaw detection wheels, which is a wheel-type probe structure. Since the center of the laid railway track is not strictly on the same straight line and there are errors, and there will also be phenomena such as position deviation during use. Although such errors or deviations do not affect the normal use of the railway track, they have a greater impact on the railway track flaw detection effect. Therefore, in order to ensure the flaw detection effect of the flaw detection wheels, centering adjustment is required. In order to achieve the automatic centering of the flaw detection wheels in this embodiment, attitude detection and image recognition are specifically used to cooperate for centering control, which not only reduces the work intensity of railway personnel, but also improves the flaw detection work efficiency and flaw detection accuracy.

[0049] The identification and acquisition of the travel section data of the flaw detection rail vehicle in this embodiment further includes: extracting the track line in the image data through image recognition; calculating the distance D between the curved track and the straight track based on the pixel points of the track line at the preset distance d in the image data, and calculating the curvature radius R of the curved track according to the formula D 2 -2RD + d 2 = 0. Specifically, referring to Figure 2 , the preset distance d in this embodiment is to take real-time photos at a distance of 30 meters. The track line can be extracted through image recognition, and the distance D between the curved track and the straight track is calculated based on the pixel points of the track line at 30 meters. According to the Pythagorean theorem and other principles, the formula: D 2 -2RD + 900 = 0 can be obtained, and thus the curvature radius R can be calculated.

[0050] The real-time acquisition of the attitude data of the current flaw detection wheel of the track flaw detection vehicle in this embodiment further includes: obtaining the attitude data of the current flaw detection wheel of the track flaw detection vehicle through the cooperation of the inertial navigation module and the gyroscope. Among them, referring to Figure 3 , the attitude data includes the displacement of the flaw detection wheel and the tilt angle α. The displacement includes the displacements dR and dL of the top points of the left and right rails from the vehicle body respectively. Specifically, the inertial navigation module plays a role in positioning and correcting the attitude. Through the inertial navigation module, the attitude parameter information of the flaw detection wheel can be obtained, that is, the displacement and tilt angle of the equipment can be known in time, so as to infer the state of the flaw detection wheel running on the railway track. The gyroscope is a device that can measure displacement and tilt angle. Through one-dimensional or multi-dimensional gyroscopes, different equipment attitude data can be measured, so as to know the situation of the railway line.

[0051] Step S2 of this embodiment further includes:

[0052] S21: Calculating the superelevation value C of the track based on the displacement and tilt angle of the flaw detection wheel:

[0053]

[0054] Wherein, α is the inclination angle of the flaw detection wheel, dR is the displacement between the top of the right rail and the vehicle body, and dL is the displacement between the top of the left rail and the vehicle body;

[0055] Specifically, referring to Figure 3 , the superelevation refers to the elevation difference between the top surfaces of the left and right rails within the same cross-section of the track, usually expressed as the height difference of the rail surface inclination angle under 1500 mm, and is calculated by measuring the inclination angle of the track plane relative to the horizontal plane. The inclination angle can be measured by a gyroscope.

[0056] S22: Calculate according to the radius of curvature R and the superelevation value C to obtain the adjustment offset D of the flaw detection wheel:

[0057] D = 10000C / R

[0058] Among them, according to the change of the radius of curvature R, the line change of the flaw detection track vehicle is predicted in advance, and the radius of curvature in the adjustment offset D control is set in advance based on the line change, so that when the track flaw detection vehicle travels to the position of the line change, the centering mechanism can directly perform centering adjustment on the flaw detection wheel according to the superelevation value control.

[0059] Specifically, by calculating according to the measured radius of curvature and superelevation value, the required lateral distance for adjusting the centering mechanism can be obtained. Among them, according to railway practical experience, it is considered that when the radius of curvature is less than 1000 m and the superelevation is greater than 30 mm, it will affect the flaw detection signal of the centering mechanism's detection wheel. It is considered that the offset can be obtained from the radius of curvature and superelevation through the calculation formula D = 10000C / R, and generally the offset is at the millimeter level. In this embodiment, the automatic adjustment of the centering mechanism is truly realized through attitude detection and image recognition. Among them, attitude detection can obtain the attitude data of the flaw detection wheel in real time, and the superelevation value of the track can be detected according to the attitude data. Image recognition can perform line analysis and predict line changes in advance, obtain the travel section data of the flaw detection wheel, and detect the radius of curvature of the track according to the travel section data. Based on the superelevation value and the radius of curvature, it can be deduced how much distance the current centering position should be offset to achieve a better flaw detection effect. In this way, when the track flaw detection vehicle travels to the corresponding position, it can directly adjust the centering mechanism to the corresponding optimal position according to the radius of curvature predicted in advance combined with the superelevation value detected in real time, rather than starting to analyze and perform centering adjustment when the track flaw detection vehicle travels to the line change position, avoiding the error of some sections caused by untimely adjustment. The cooperation of attitude detection and image recognition greatly improves the flaw detection work efficiency and flaw detection accuracy.

[0060] Referring to Figure 4 and Figure 5, this application provides an automatic centering system for an orbital flaw detector based on the above embodiments. Among them, a flaw detection wheel and a centering mechanism for centering adjustment of the flaw detection wheel are mounted on the orbital flaw detector. The system is characterized in that it includes: an attitude detection module, an image recognition module, and a processing controller. The processing controller is respectively connected to the attitude detection module and the image recognition module in a signal connection;

[0061] The attitude detection module is used to obtain the attitude data of the current flaw detection wheel of the orbital flaw detector in real time. The processing controller is used to control the centering mechanism to perform centering adjustment on the flaw detection wheel in real time according to the attitude data;

[0062] The image recognition module is used to collect image data in the moving direction of the orbital flaw detector, identify and obtain the traveling section data of the flaw detection rail vehicle. The processing controller is also used to predict in advance the line change of the flaw detection rail vehicle according to the traveling section data, and set in advance the adjustment control instruction of the flaw detection wheel based on the line change, so that when the orbital flaw detector travels to the position of the line change, it directly controls the centering mechanism to perform centering adjustment on the flaw detection wheel.

[0063] The present embodiment will be described in detail below, but not limited thereto.

[0064] The attitude detection module of this embodiment adopts an inertial navigation module and a gyroscope. The inertial navigation module cooperates with the gyroscope to jointly measure the displacement and tilt angle of the current flaw detection wheel. Among them, the inertial navigation module plays the role of positioning and correcting the attitude. Through the inertial navigation module, the attitude parameter information of the flaw detection wheel can be obtained, that is, the displacement and tilt angle of the device can be known in time, so as to infer the state of the flaw detection wheel running on the railway track. The gyroscope is a device that can measure displacement and tilt angle. Through one-dimensional or multi-dimensional gyroscopes, different device attitude data can be measured, so as to know the railway line conditions.

[0065] The image recognition module of this embodiment can be based on a camera product to achieve clear shooting of long-distance scenes in daylight and at night. Among them, an image of the driving line is taken through the camera, and the line curve or straight line condition is fitted through the upper computer image recognition algorithm, and the curvature radius of the curve is calculated, so as to infer the offset distance required by the centering mechanism to achieve the best flaw detection effect. Specifically, an orbit recognition model trained based on images is provided in the image acquisition module to identify and obtain the traveling section data of the flaw detection rail vehicle. The traveling section data includes the straight line feature and the curve feature of the orbit. Among them, the image recognition module performs rail modeling through a large number of image trainings, so as to extract the rail route during driving, and then judge through different feature points of the straight line and the curve. At the same time, the curvature radius of the curve can be calculated according to the curve radian, so as to provide it to the control layer to judge the distance of the centering mechanism that needs to be adjusted. For example, see Figure 2, the image recognition module will collect and calculate about 30 meters in advance, enabling the system to perform data analysis and calculation in advance. When the device moves forward the corresponding distance, it can directly start adjusting the centering mechanism, rather than starting the corresponding adjustment when it reaches the curved section of the road, thus avoiding the errors in some sections caused by untimely adjustment.

[0066] Preferably, this embodiment further includes: a remote intervention module, which is remotely communicatively connected to the processing controller and is used to remotely control the centering mechanism to perform centering adjustment on the flaw detection wheel. Further, the real-time image recognition module can capture the road conditions ahead of the vehicle and remotely transmit them to the visual interface of the staff, so that the staff can judge whether to turn off the automatic centering system and manually control the adjustment of the centering mechanism.

[0067] Preferably, this embodiment further includes: a on-site intervention module, which is directly signal-connected to the processing controller and is used to on-site control the centering mechanism to perform centering adjustment on the flaw detection wheel. Further, the real-time image recognition module can capture the road conditions ahead of the vehicle and on-site display them on the visual interface of the staff, so that the staff can judge whether to turn off the automatic centering system and manually control the adjustment of the centering mechanism.

[0068] See Figure 2 , through the above two intervention modules, the automatic centering system can be intervened, and it can be selected whether to turn off the automatic centering system. When there is no one driving the vehicle, when the automatic centering system is turned off, the distance to be adjusted can be judged through the video images remotely transmitted by the real-time video image acquisition system and controlled; when the staff is driving the vehicle, they can directly judge and control the centering mechanism according to the road section. If the automatic centering system is not turned off, the system will move the centering mechanism according to the control layer command.

[0069] The centering mechanism of this embodiment is also provided with a displacement sensor signal-connected to the processing controller. The displacement sensor is arranged in the centering mechanism and is used to real-time feedback the current position data of the centering mechanism. Specifically, the position of the centering mechanism is moved by the electric push rod. The electric push rod contains a displacement sensor, which can accurately feedback the current position data of the centering mechanism to the processing controller in real time for control. At the same time, this data will also be real-time displayed on the human-machine interaction interface for the staff to view.

[0070] Now, the working process of this embodiment will be further described through a more specific scenario.

[0071] During the operation of the rail flaw detection vehicle, it is possible to choose whether to turn on the automatic centering system. If the automatic centering system is turned off, manual control is mainly used. The image recognition module remotely transmits the information to the computer terminal, enabling remote workers to view the driving route ahead in real time. Based on the road conditions and work experience, they can determine whether it is necessary to adjust the position and distance of the centering mechanism, and then remotely control the electric push rod of the centering mechanism to move.

[0072] If the automatic centering system is selected to be turned on, positioning is carried out based on the inertial navigation module and the gyroscope, and calculations are made with the running speed V of the equipment on the track. The running speed V can collect data on the number of wheel rotations through an encoder, and the real-time speed during the vehicle's operation can be calculated using the formula: number of rotations * tire circumference / time = running speed. See Figure 2 , taking 30 meters as a unit, the image recognition module takes real-time photos at a distance of 30 meters. The track line can be extracted through image recognition. Based on the pixel points of the track line at 30 meters, the distance D between the curved track and the straight track can be calculated. According to principles such as the Pythagorean theorem, the formula: D 2 -2RD + 900 = 0, from which the radius of curvature R can be calculated.

[0073] See Figure 3 , the superelevation refers to the elevation difference between the top surfaces of the left and right rails within the same cross-section of the track. It is usually expressed as the height difference of the rail surface inclination angle within 1500 mm, and is calculated by measuring the inclination angle of the track plane relative to the horizontal plane. The inclination angle can be measured by a gyroscope. The basic calculation formula is:

[0074]

[0075] In the formula, α is the inclination angle output by the inertial component, with the unit of radian, dR is the displacement between the right rail vertex and the vehicle body, with the unit of mm, and dL is the displacement between the left rail vertex and the vehicle body, with the unit of mm;

[0076] Based on the measured radius of curvature and superelevation value, the required lateral distance for adjusting the centering mechanism can be calculated. According to railway practical experience, when the radius of curvature is less than 1000 m and the superelevation is greater than 30 mm, it will affect the flaw detection signal of the detection wheel of the centering mechanism. It is considered that the offset can be obtained from the radius of curvature and superelevation through the calculation formula D = 10000C / R. Generally, the offset is at the millimeter level. Therefore, the electric push rod is equipped with a displacement sensor, and mm-level remote point control can be achieved through the control program to improve the accuracy of adjusting the centering mechanism, thereby enhancing the flaw detection ability.

[0077] In this embodiment, the automatic adjustment of the centering mechanism is truly realized through the attitude detection module and the image recognition module. Among them, the attitude detection module can detect the offset angle and tilt angle of the flaw detection wheel in real time. According to the offset angle and tilt angle, the superelevation condition and curvature radius of the track can be calculated, so as to deduce how much distance the current centering position should be offset to achieve a better flaw detection effect. The image recognition module can conduct line analysis and predict line changes in advance, and set the adjustment control instructions for the centering structure in advance based on the line changes. In this way, when the rail flaw detector travels to the corresponding position, it can directly adjust the centering mechanism to the corresponding optimal position in time, rather than starting to analyze and adjust the centering when the rail flaw detector reaches the line change position, avoiding the errors in some sections caused by untimely adjustment. The cooperation between the attitude detection module and the image recognition module greatly improves the flaw detection work efficiency and flaw detection accuracy.

[0078] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, provided that these changes fall within the scope of the claims of the present invention and their equivalent technologies, they still fall within the protection scope of the present invention.

Claims

1. An automatic centering method for an orbital flaw detector vehicle, characterized in that, it includes the following steps: S1: Collect image data in the moving direction of the orbital flaw detector vehicle, identify and obtain the travel section data of the orbital flaw detector vehicle, and real-time obtain the attitude data of the current flaw detection wheel of the orbital flaw detector vehicle; S2: According to the attitude data, the centering mechanism of the orbital flaw detector vehicle is controlled in real time to perform centering adjustment on the flaw detection wheel. Among them, according to the travel section data, the line change of the orbital flaw detector vehicle is predicted in advance, and based on the line change, the adjustment control instruction of the flaw detection wheel is set in advance, so that when the orbital flaw detector vehicle travels to the position of the line change, the centering mechanism is directly controlled to perform centering adjustment on the flaw detection wheel; The identification and acquisition of the travel section data of the orbital flaw detector vehicle further includes: Extracting the track line in the image data through image recognition; Calculating the distance D between the curved track and the straight track according to the pixel points of the track line at a preset distance d in the image data, and calculating the curvature radius R of the curved track according to the formula D² - 2RD + d² = 0; Calculating the superelevation value C of the track based on the displacement and tilt angle of the flaw detection wheel; Calculate according to the curvature radius R and the superelevation value C to obtain the adjustment offset D of the flaw detection wheel 1 .

2. The automatic centering method for an orbital flaw detector vehicle according to claim 1, characterized in that, The real-time acquisition of the attitude data of the current flaw detection wheel of the orbital flaw detector vehicle further includes: The attitude data of the current flaw detection wheel of the orbital flaw detector vehicle is measured in real time through the cooperation of an inertial navigation module and a gyroscope. Among them, the attitude data includes the displacement and tilt angle of the flaw detection wheel.

3. The automatic centering method for an orbital flaw detector vehicle according to claim 2, characterized in that, The step S2 further includes: S21: Calculating the superelevation value C of the track based on the displacement and tilt angle of the flaw detection wheel: In the formula, α is the tilt angle of the flaw detection wheel, dR is the displacement between the right rail apex and the vehicle body, and dL is the displacement between the left rail apex and the vehicle body; S22: Calculate based on the radius of curvature R and the superelevation value C to obtain the adjustment offset D of the flaw detection wheel 1 : D 1 = 10000 C / R Among them, the line change of the traveling of the rail flaw detection vehicle is predicted in advance according to the change of the curvature radius R, and the adjustment offset D is set in advance based on the line change. 1 The curvature radius in the control is such that when the rail flaw detection vehicle travels to the position of the line change, the centering mechanism directly controls the centering adjustment of the flaw detection wheel according to the superelevation value.

4. An automatic centering system adopting the automatic centering method for an orbital flaw detector vehicle according to any one of claims 1-3, wherein, A flaw detection wheel and a centering mechanism for centering adjustment of the flaw detection wheel are mounted on the orbital flaw detector vehicle. It is characterized in that the system includes: an attitude detection module, an image recognition module, and a processing controller, and the processing controller is respectively connected to the attitude detection module and the image recognition module in a signal connection; The attitude detection module is used to real-time obtain the attitude data of the current flaw detection wheel of the orbital flaw detector vehicle, and the processing controller is used to control the centering mechanism to perform centering adjustment on the flaw detection wheel according to the attitude data in real time; The image recognition module is used to collect image data in the moving direction of the rail flaw detector, and identify and obtain the travel section data of the rail flaw detector. The processing controller is further used to predict in advance the line change of the rail flaw detector according to the travel section data, and set in advance the adjustment control instruction of the flaw detection wheel based on the line change, so that when the rail flaw detector travels to the position of the line change, the centering mechanism is directly controlled to perform centering adjustment on the flaw detection wheel; The identification and acquisition of the travel section data of the rail flaw detector further includes: Extracting the rail line from the image data through image recognition; Calculating the distance D between the curved track and the straight track according to the pixel points of the rail line at a preset distance d in the image data, and calculating the radius of curvature R of the curved track according to the formula D² - 2RD + d² = 0; Calculating the superelevation value C of the track based on the displacement and tilt angle of the flaw detection wheel; Calculate according to the curvature radius R and the superelevation value C to obtain the adjustment offset D of the flaw detection wheel 1 .

5. The automatic centering system according to claim 4, wherein, The attitude detection module includes an inertial navigation module and a gyroscope. The inertial navigation module cooperates with the gyroscope to jointly measure in real time the attitude data of the current flaw detection wheel of the rail flaw detector. Among them, the attitude data includes the displacement and tilt angle of the flaw detection wheel.

6. The automatic centering system according to claim 5, wherein, further includes: A remote intervention module, which is remotely communicatively connected to the processing controller and is used to remotely control the centering mechanism to perform centering adjustment on the flaw detection wheel.

7. The automatic centering system according to claim 6, wherein, further includes: A on-site intervention module, which is directly signal-connected to the processing controller and is used to on-site control the centering mechanism to perform centering adjustment on the flaw detection wheel.

8. The automatic centering system according to claim 7, wherein, further includes a displacement sensor signal-connected to the processing controller. The displacement sensor is arranged in the centering mechanism and is used to feedback the current position data of the centering mechanism in real time.

Images