A method and device for adaptive inspection of key components under a subway or train

By identifying the vehicle model and adjusting the lifting platform, combining laser ranging array and cubic spline interpolation technology, the low efficiency and inaccurate position caused by fixed positions in the subway or train bottom patrol are solved, and adaptive inspection is realized. Clear car image information.

CN114419019BActive Publication Date: 2024-11-08CHENGDU YUNDA TECH CO LTD +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202210093722.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-26
Publication Date
2024-11-08
Estimated Expiration
2042-01-26

AI Technical Summary

Technical Problem

The prior art is difficult to realize adaptive inspection of key components at the bottom of subways or trains, resulting in low inspection efficiency and inaccurate positioning, and fixed line scanning camera position, resulting in partial out of focus on the shooting screen.

Method used

By identifying the vehicle model and adjusting the lifting platform, the image acquisition mechanism maintains the optimal distance from the vehicle bottom, and using laser ranging arrays and cubic spline interpolation technology to ensure the optimal distance between the image acquisition mechanism and the vehicle bottom, and perform longitudinal distortion correction. Get clear undercarriage images.

Benefits of technology

Adaptive inspection of the undercarriage structures of different trains is realized, and the image information of the undercarriage is accurately and clearly captured, improving the inspection efficiency and accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114419019B_ABST
    Figure CN114419019B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of intelligent robot technology, and discloses an adaptive inspection method for key components of the undercarriage of a subway or train, comprising the following steps: Step 1, identifying the vehicle type of the subway or train to be inspected, and determining whether there is a data record for the vehicle type of the subway or train to be inspected, and if so, proceeding to Step 3, otherwise, proceeding to Step 2; Step 2, obtaining the data entry of the vehicle type of the subway or train to be inspected and the optimal distance for image acquisition during the inspection process; Step 3, inspecting the subway or train to be inspected by an inspection device, and during the inspection process, adjusting the lifting platform of the inspection device to keep the image acquisition mechanism at the optimal distance from the undercarriage of the subway or train to be inspected. The present invention also discloses an adaptive inspection device for key components of the undercarriage of a subway or train. The adaptive inspection method and device disclosed in the present invention can adapt to different undercarriage structures of subways or trains to achieve accurate and clear capture of undercarriage image information.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent robots, and in particular to a method and a device for adaptively inspecting key components of a subway or train bottom. Background Art

[0002] In order to ensure the safe operation of subway trains, it is very important to regularly inspect the key components of subway trains. Conventional inspection schemes use manual inspections to obtain relevant information through visual inspections by workers. This method is inefficient and the comprehensiveness of the inspection is difficult to ensure. A new inspection method is to use robots to replace manual inspections. By installing various sensors on the robots, such as line scan cameras and 3D cameras, the robots can autonomously obtain and analyze the information of key components under the vehicle. The method of using inspection robots to inspect the underside of trains has the advantages of high efficiency and accurate positioning. A key point of the inspection robot research is how to obtain the information of key components under the train more accurately, efficiently and clearly. The camera systems usually carried are fixed-focus cameras. However, the chassis heights of different trains are different, and the required line scan camera positions are different. The installation heights of key components under the vehicle are different. The fixed line scan camera position will cause the shooting picture to be partially out of focus.

[0003] Therefore, it is urgent to develop an inspection device that can adapt to different train undercarriage structures to accurately and clearly capture the undercarriage image information. Summary of the invention

[0004] The present invention provides a method and device for adaptively inspecting key components of a subway or train bottom, which can adaptively inspect and focus on the key components of the bottom.

[0005] The present invention is achieved through the following technical solutions:

[0006] An adaptive inspection method for key components under a subway or train comprises the following steps:

[0007] Step 1: Identify the type of subway or train being inspected, and determine whether there is a data record of the type of subway or train being inspected, wherein the data record includes the optimal distance for image acquisition during the inspection process. If yes, proceed to step 3; otherwise, proceed to step 2.

[0008] Step 2: Obtain data entries of the type of subway or train being inspected and the optimal distance for image acquisition during the inspection process;

[0009] Step 3: inspecting the subway or train by using the inspection device, and during the inspection process, adjusting the lifting platform of the inspection device so that the image acquisition mechanism maintains an optimal distance from the bottom of the subway or train being inspected, and at the same time, acquiring an image of the bottom of the subway or train being inspected to obtain an original image;

[0010] Step 4: Process the original image to obtain a clear image of the underbody of the inspected subway or train.

[0011] As an optimization, the specific implementation method of step 1 is:

[0012] Step 1.1, scanning the barcode set on the bottom of the subway or train to be inspected by the image acquisition mechanism on the inspection device, and transmitting the acquired information of the barcode to the main control box;

[0013] Step 1.2, the main control box determines the currently scanned vehicle model based on the information of the bar identification code, and queries the database whether there is a data record for the vehicle model. If there is a data record for the vehicle model, jump directly to step 3, otherwise, enter step 2.

[0014] As an optimization, the specific steps in step 2 are:

[0015] Step 2.1, add the data entry of the vehicle model without data record in the database;

[0016] Step 2.2: Add the chassis height and corresponding mileage information of the vehicle model without data record to the database to obtain the optimal distance for image acquisition during the inspection process.

[0017] As an optimization, in step 2.2, the specific method of obtaining the chassis height of the vehicle type with no data record is: adjust the lifting platform on the inspection device to the lowest posture, drive the inspection device to move forward along the track direction from the front to the rear of the inspected subway or train, obtain the distance information between the lifting platform and the bottom of the inspected subway or train by the laser ranging array on the inspection device, and calculate the lifting height of the lifting platform on the inspection device according to the distance information to ensure that the image acquisition mechanism on the inspection device can maintain the optimal distance for image acquisition with the bottom of the inspected subway or train during the inspection process, obtain the corresponding mileage information by the odometer of the inspection device, and finally record the distance information, lifting height and mileage information through the main control box.

[0018] As an optimization, when the inspection device moves to the rear of the subway or train being inspected, the lifting height and mileage information are matched and corrected.

[0019] As an optimization, the specific steps for matching and correcting the lift height and mileage information are as follows:

[0020] a. Let x n is the mileage information sampled at the nth moment, y(x n ) is the distance from the laser ranging array to the bottom of the subway or train being inspected at the nth moment;

[0021] b. Obtain the point where the distance from the bottom of the vehicle measured by the laser ranging array changes, that is, n satisfies |y(x n )-y(x n-1 )|>y MAX point, where y MAX The preset vehicle bottom distance jump threshold is used, and according to the jump point, the boundary position of the compartment of the inspected subway or train is determined, and the data of the connection part between the two compartments is deleted;

[0022] c. The mileage information data corresponding to the measured data with the connection between the two carriages deleted is decomposed and numbered according to the carriages, and the mileage information data obtained from each carriage is interpolated with cubic spline to solve the problem of the lifting height and mileage information being out of sync.

[0023] As an optimization, in step c, the specific method of performing cubic spline interpolation on the mileage information data obtained from each carriage is:

[0024] Let n = k, 2k, 3k ..... mk, m is the number of cars of the inspected subway or train, k represents the number of sampling points in each car, and the distance between the received laser ranging array and the bottom of the inspected subway or train is S (x k )=y k , divide the mileage information data obtained from each carriage into the interval [(x0,x k ),(x k ,x 2k ),(x 2k ,x 3k ),.....,(x (m-1)k ,x mk )], construct a cubic spline interpolation curve S in each interval, and the interpolation curve S satisfies the following conditions: Let each segmented small interval [x (m-1)k ,x mk ]S(x)=S j (x), where S j (x) is in the corresponding small interval [x (m-1)k ,x mk ], S(x) is a cubic function consisting of several S j (x) and satisfy the interpolation condition S(x i )=y i ,y i Indicates that at mileage x i The actual vehicle bottom distance obtained by the laser ranging array is i, which represents the number of the sampling point; let the cubic equation system S(x), Continuous and for the cubic system S(x), Solve to get S(x), substitute n∈((m-1)k,mk) into S(xn ), that is, interpolate the odometer points within the sampling interval of the laser ranging array.

[0025] As an optimization, in step 3, the specific steps to keep the image acquisition mechanism at the optimal distance from the bottom of the subway or train to be inspected by adjusting the lifting platform of the inspection device during the inspection process are as follows:

[0026] Predict the height between the image acquisition mechanism and the chassis in front of the subway or train to be inspected, and raise or lower the height of the lifting platform in advance. Specifically: Assume that the imaging focal length range of the image acquisition mechanism is [A1, A2], and obtain the first derivative of the interpolation curve S(x)=S k (x) where S k (x) represents the distance from the laser ranging array to the bottom of the vehicle, the first derivative of S k (x) is the speed, k represents the number of sampling points for each carriage, x represents the mileage information, traverse x n , x n represents the mileage information sampled at the nth moment, find the set of points of x that satisfy , denoted as X0, where is the fastest lifting and lowering speed that the lifting platform can provide. For all x i ∈X0, select several j points belonging to the neighborhood centered on i with a radius of δ. Under the condition that A1 < y0 - y1 < A2, adjust y0 and δ, and re - perform cubic spline interpolation on the points within the neighborhood centered on i with a radius of δ, so that y0 refers to the distance from the laser ranging array to the vehicle chassis, then the distance from the line - scan camera to the vehicle chassis is y j0 - y1, and y1 represents the distance between the line - scan camera and the ranging module in the vertical direction.

[0027] As an optimization, in step 4, the specific steps to process the original image to obtain a clear image of the bottom of the subway or train to be inspected are as follows: Distortion correction based on the height information of the lifting platform. Specifically:

[0028] Step 4.1: Assume that the average height of the lifting platform during the chassis scanning of the subway or train to be inspected is h0, and the platform height of the lifting platform corresponding to the line - scan data of the nth row of the image acquisition mechanism is h. Set the corresponding longitudinal distortion correction coefficient K = αh / h0.

[0029] The present invention also discloses an adaptive inspection device for key components under the subway or train, which executes the aforementioned adaptive inspection method for key components under the subway or train, including a walking robot, a lifting platform and a laser ranging array are arranged on the top of the walking robot, an image acquisition mechanism is arranged on the upper surface of the lifting platform, and also includes a main control box, and the walking robot, the lifting platform, the laser ranging array and the image acquisition mechanism are electrically connected to the main control box.

[0030] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0031] The adaptive inspection method and device disclosed in the present invention can adapt to different subway or train undercarriage structures to achieve accurate and clear capture of undercarriage image information. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative work. In the drawings:

[0033] Figure 1 This is a flow chart of a method for adaptive inspection of key components under a subway or train according to the present invention;

[0034] Figure 2 The present invention is a schematic structural diagram of an adaptive inspection device for key components on the underbody of a subway or train.

[0035] Marks and corresponding parts names in the attached drawings:

[0036] 1- walking robot, 2- lifting platform, 3- laser ranging array, 4- line scan camera, 5- main control box. DETAILED DESCRIPTION

[0037] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0038] Example 1

[0039] A self-adaptive inspection device for key components under the undercarriage of a subway or a train, used for implementing the self-adaptive inspection method for key components under the undercarriage of a subway or a train described later, comprising a walking robot 1, a lifting platform 2 and a laser ranging array 3 are arranged on the top of the walking robot 1, an image acquisition mechanism 4 is arranged on the upper surface of the lifting platform 2, and also comprising a main control box 5, wherein the walking robot 1, the lifting platform 2, the laser ranging array 3 and the image acquisition mechanism 4 are electrically connected to the main control box 5.

[0040] Specifically, the image acquisition mechanism 4 uses a line scan camera. The chassis of the lifting platform 2 is vertically installed on the upper surface of the shell of the walking robot 1, and the line scan camera is fixed to the top of the lifting platform 2 through a structural member. The observation direction of the lens of the line scan camera is vertically upward, perpendicular to the bottom plane of the train. The laser ranging array 3 is installed on the top of the walking robot 1, and the ranging direction is vertically upward. In this embodiment, the measurement method of the laser ranging array 3 is multi-sensor combination single-row ranging. The walking robot 1, the lifting platform 2, the laser ranging array 3, and the line scan camera 4 are electrically connected to the main control box 5 respectively.

[0041] In this embodiment, the line scan camera uses a color global shutter CMOS camera to improve the camera shooting effect.

[0042] In this embodiment, the walking robot selects a trolley with an independently suspended four-wheel chassis, keeps the wheels in reliable contact with the ground, and relies on the encoder built into the trolley's walking motor to read the distance the trolley has traveled.

[0043] In this embodiment, four closed-loop linear motors with absolute position encoders (Jayander Pneumatic RQZ32-150-1002-0000-2T+100W servo motor) are used as the driving power mechanism of the lifting platform to ensure high stability and fast response speed when the lifting platform is raised and lowered.

[0044] Example 2

[0045] The present invention also discloses an adaptive inspection method for key components of a subway or train underbody, which includes three processes of identification, detection and inspection of the inspected subway or train (hereinafter referred to as "the inspected subway or train" as "vehicle"):

[0046] The specific steps include:

[0047] Vehicle identification stage (step 1): identify the vehicle model and determine whether there is a data record for the vehicle model, the data record including the optimal distance for image acquisition during the inspection process. If yes, proceed to step 3, otherwise, proceed to step 2.

[0048] Step 1.1, scanning the barcode set on the bottom of the vehicle through the image acquisition mechanism on the inspection device, and transmitting the acquired information of the barcode to the main control box;

[0049] Step 1.2, the main control box determines the currently scanned vehicle model based on the information of the bar identification code, and queries the database whether there is a data record for the vehicle model. If there is a data record for the vehicle model, jump directly to step 3, otherwise, enter step 2.

[0050] Simply put, the walking robot autonomously walks to the designated location on the front of the vehicle, lowers the height of the lifting platform to the lowest point, and uses a line scan camera to scan the barcode identification code attached to the bottom of the train being tested. The main control box determines the current scanned vehicle model and queries the database to see if there is a data record for the model. If the model is not registered, the model is added to the database and the vehicle inspection phase begins.

[0051] Vehicle detection phase (step 2): obtain the data entry of the vehicle model and the optimal distance for image acquisition during the inspection process; the detection phase includes two parts: registering the vehicle and obtaining chassis height data. Registering a vehicle means adding the data entry of the corresponding vehicle model to the database.

[0052] The specific steps of step 2 are:

[0053] Step 2.1, add the data entry of the vehicle model without data record in the database;

[0054] Step 2.2: Add the chassis height and corresponding mileage information of the vehicle model without data record to the database to obtain the optimal distance for image acquisition during the inspection process.

[0055] Specifically, the specific method for obtaining the chassis height of the vehicle without data record is: adjust the lifting platform on the inspection device to the lowest posture, drive the inspection device to move forward from the front to the rear of the vehicle along the track direction, and obtain the distance information between the lifting platform and the chassis of the vehicle by the laser ranging array on the inspection device. The specific evaluation process of the distance information is: perform mean filtering on the data obtained by each sensor in the laser ranging array at the same time, and calculate the average value of the values ​​obtained by all sensors at the same time. The average value is the distance from the laser ranging array to the bottom of the vehicle at a certain moment. The distance information from the lifting platform (or line scanning camera) to the bottom of the vehicle can be obtained by conversion. Then, the lifting height of the lifting platform on the inspection device is preliminarily calculated based on the obtained distance information to ensure that the image acquisition mechanism on the lifting platform can maintain the best distance for image acquisition with the bottom of the subway or train being inspected during the inspection process, and the corresponding mileage information is obtained by the odometer of the inspection device. Finally, the distance information, lifting height, and mileage information are recorded by the main control box.

[0056] Among them, when the inspection device moves to the rear of the subway or train being inspected, since the speed at which the laser sensor in the laser ranging array obtains distance information is lower than the update speed of mileage information, it is necessary to match and correct the lifting height and mileage information.

[0057] The specific matching steps are:

[0058] a. Let x n is the mileage information sampled at the nth moment, y(x n ) is the distance from the laser ranging array to the bottom of the subway or train being inspected at the nth moment;

[0059] b. Obtain the point where the distance from the bottom of the vehicle measured by the laser ranging array changes, that is, n satisfies |y(x n )-y(x n-1 )|>y MAX point, where y MAX It is a preset bottom vehicle distance jump threshold, and according to the above jump point, the boundary position of the inspected subway or train is determined, and the data of the connecting part of the two carriages is deleted to avoid the interference of the gap between the carriages on the chassis ranging; here, the boundary position of the carriage is the corresponding jump point, therefore, by deleting the jump point, the data of the distance from the bottom vehicle to the connecting part of the two carriages measured by the laser ranging array can be deleted.

[0060] c. The mileage information data corresponding to the measured data with the connection between the two carriages deleted is decomposed and numbered according to the carriages, and the mileage information data obtained from each carriage is interpolated with cubic spline to solve the problem of the lifting height and mileage information being out of sync.

[0061] Specifically, in step c, the specific method of performing cubic spline interpolation on the mileage information data obtained from each carriage is as follows:

[0062] Let n = k, 2k, 3k ..... mk, m is the number of cars of the inspected subway or train, k represents the number of sampling points in each car, and the distance between the received laser ranging array and the bottom of the inspected subway or train is S (x k )=y k , divide the mileage information data obtained from each carriage into the interval [(x0,x k ),(x k ,x 2k ),(x 2k ,x 3k ),.....,(x (m-1)k ,x mk)], there are M intervals in total, with M+1 points, and a cubic spline interpolation curve S is constructed in each interval. The interpolation curve S satisfies the following conditions: Let each segmented small interval [x (m-1)k ,x mk ]S(x)=S j (x), where S j (x) is in the corresponding small interval [x (m-1)k ,x mk ], S(x) is a cubic function consisting of several S j (x) and satisfy the interpolation condition S(x i )=y i ,y i Indicates that at mileage x i The actual vehicle bottom distance obtained by the laser ranging array is i, which represents the number of the sampling point; let the cubic equation system S(x), Continuous and for the cubic system S(x), Solve to get S(x), substitute n∈((m-1)k,mk) into S(x n ), that is, to interpolate the odometer points within the sampling interval of the laser ranging array, thus solving the problem of asynchronous sampling between the laser sensor and the odometer.

[0063] Vehicle inspection stage (step 3): The inspected subway or train is inspected by an inspection device, and during the inspection process, the lifting platform of the inspection device is adjusted to keep the image acquisition mechanism at an optimal distance from the bottom of the inspected subway or train, and at the same time, the bottom of the inspected subway or train is imaged to obtain an original image.

[0064] That is, the walking robot moves along the track direction, and according to the registered vehicle information and the relative height between the key components and the vehicle, drives the lifting platform to keep the preset distance (optimal distance) between the line scan camera and the bottom of the vehicle at all times, and the line scan camera collects the original image information of the key components under the vehicle corresponding to the mileage.

[0065] Since the actual lifting speed of the lifting platform is limited, when the required lifting height changes too quickly, the actual lifting height will not be able to keep up with the required lifting height, resulting in partial out-of-focus of the top or bottom image. The focal length of a line scan camera is usually a range, and clear imaging can be ensured by ensuring that the distance between the line scan camera and the chassis is within a certain range. Therefore, the problem of out-of-focus caused by the limited lifting speed of the lifting platform can be solved by optimizing the movement trajectory of the lifting platform, predicting the required height ahead, and raising or lowering the lifting height in advance.

[0066] Specifically, assuming that the imaging focal length interval of the image acquisition mechanism is [A1, A2], the interpolation curve S(x)=Sk The first derivative of (x) where S k (x) represents the distance from the laser ranging array to the bottom of the vehicle, S k The first derivative of (x) is the speed, k represents the number of sampling points for each carriage, x represents the mileage information, and traverse x n , x n represents the mileage information sampled at the nth moment, find the set of points of x that satisfy , and denote it as X0. The purpose of the above scheme is to find the mileage corresponding to the points where the lifting platform speed changes too fast within the entire length range of the subway or train. Where is the fastest lifting speed that the lifting platform can provide. For all x i ∈X0, select several points belonging to the neighborhood centered at i with a radius of δ. Under the condition that A1 < y0 - y1 < A2, adjust y0 and δ, and re - perform cubic spline interpolation on the points within the neighborhood centered at i with a radius of δ, so that y0 refers to the distance from the laser ranging array to the vehicle chassis, then the distance from the line - scan camera to the vehicle chassis is y0 - y1, and y1 represents the distance between the line - scan camera and the ranging module in the vertical direction. Adjusting in this way, the trajectory of the lifting platform obtained is continuous and smooth, and the shooting stability is good.

[0067] Step 4: Process the original image to obtain a clear image of the bottom of the subway or train to be inspected.

[0068] In this embodiment, the specific steps in Step 4 are: distortion correction based on the height information of the lifting platform, specifically:

[0069] Assume that the average height of the lifting platform during the chassis scanning of the subway or train to be inspected is h0, and the platform height of the lifting platform corresponding to the line - scan data of the nth - row image acquisition mechanism is h. Set the corresponding longitudinal distortion correction coefficient as K = αh / h0. Where α is related to the lens parameters of the line - scan camera and is measured through experiments.

[0070] The above - mentioned specific implementation manners further elaborate on the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above - mentioned is only the specific implementation manners of the present invention and is not used to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An adaptive inspection method for key components under a subway or train, characterized in that: The steps include: Step 1: Identify the type of subway or train being inspected, and determine whether there is a data record of the type of subway or train being inspected, wherein the data record includes the optimal distance for image acquisition during the inspection process. If yes, proceed to step 3; otherwise, proceed to step 2. Step 2: Obtain data entries of the type of subway or train being inspected and the optimal distance for image acquisition during the inspection process; Step 3: inspecting the subway or train by using the inspection device, and during the inspection process, adjusting the lifting platform of the inspection device so that the image acquisition mechanism maintains an optimal distance from the bottom of the subway or train being inspected, and at the same time, acquiring an image of the bottom of the subway or train being inspected to obtain an original image; In step 3, during the inspection process, the lifting platform of the inspection device is adjusted so that the image acquisition mechanism maintains an optimal distance from the bottom of the inspected subway or train. The specific steps are: Predict the height between the image acquisition mechanism and the chassis in front of the subway or train to be inspected, and raise or lower the height of the lifting platform in advance. Specifically: Assume that the imaging focal length range of the image acquisition mechanism is [A1, A2], and obtain the first derivative of the interpolation curve S(x) = S k (x) where S k (x) represents the distance from the laser ranging array to the vehicle bottom, the first derivative of S k (x) is the speed, k represents the number of sampling points per carriage, x represents the mileage information, traverse x n , x n represents the mileage information sampled at the nth moment, find the point set of x that satisfies , denoted as X0, where is the fastest lifting and lowering speed that the lifting platform can provide. For all x i ∈X0, select several j points belonging to the neighborhood centered at i with a radius of δ. i represents the serial number of the sampling point. When A1 < y0 - y1 < A2, adjust y0 and δ, and re - perform cubic spline interpolation on the points within the neighborhood centered at i with a radius of δ, so that y0 refers to the distance from the laser ranging array to the vehicle chassis, then the distance from the line - scan camera to the vehicle chassis is y0 - y1, and y1 represents the distance between the line - scan camera and the ranging module in the vertical direction; Step 4: Process the original image to obtain a clear image of the underbody of the inspected subway or train.

2. The method for adaptive inspection of key components under the subway or train according to claim 1, characterized in that: The specific implementation method of step 1 is: Step 1.1, scanning the barcode set on the bottom of the subway or train to be inspected by the image acquisition mechanism on the inspection device, and transmitting the acquired information of the barcode to the main control box; Step 1.2, the main control box determines the currently scanned vehicle model based on the information of the bar identification code, and queries the database whether there is a data record for the vehicle model. If there is a data record for the vehicle model, jump directly to step 3, otherwise, enter step 2.

3. The method for adaptive inspection of key components under the subway or train according to claim 2, characterized in that: The specific steps in step 2 are: Step 2.1, add data entries of vehicle models without data records in the database; Step 2.2: Add the chassis height and corresponding mileage information of the vehicle model without data record to the database to obtain the optimal distance for image acquisition during the inspection process.

4. The method for adaptive inspection of key components of a subway or train underbody according to claim 3, characterized in that: In step 2.2, the specific method for obtaining the chassis height of the vehicle type with no data record is: adjust the lifting platform on the inspection device to the lowest posture, drive the inspection device to move forward along the track direction from the front to the rear of the inspected subway or train, and use the laser ranging array on the inspection device to obtain the distance information between the lifting platform and the bottom of the inspected subway or train, and calculate the lifting height of the lifting platform on the inspection device through the distance information to ensure that the image acquisition mechanism on the inspection device can maintain the optimal distance for image acquisition with the bottom of the inspected subway or train during the inspection process, and use the odometer of the inspection device to obtain the corresponding mileage information, and finally record the distance information, lifting height and mileage information through the main control box.

5. The method for adaptive inspection of key components under the subway or train according to claim 4, characterized in that: When the inspection device moves to the rear of the subway or train being inspected, the lifting height and mileage information are matched and corrected.

6. The method for adaptive inspection of key components of a subway or train underbody according to claim 5, characterized in that: The specific steps of matching and correcting the lift height and mileage information are as follows: a. Let x n is the mileage information sampled at the nth moment, y(x n ) is the distance from the laser ranging array to the bottom of the subway or train being inspected at the nth moment; b. Obtain the point where the distance from the bottom of the vehicle measured by the laser ranging array changes, that is, n satisfies |y(x n )-y(x n-1 )|>y MAX point, where y MAX The preset vehicle bottom distance jump threshold is used, and according to the jump point, the boundary position of the compartment of the inspected subway or train is determined, and the data of the connection part between the two compartments is deleted; c. The mileage information data corresponding to the measured data with the connection between the two carriages deleted is decomposed and numbered according to the carriages, and the mileage information data obtained from each carriage is interpolated with cubic spline to solve the problem of the lifting height and mileage information being out of sync.

7. The method for adaptive inspection of key components of a subway or train underbody according to claim 6, characterized in that: In step c, the specific method of performing cubic spline interpolation on the mileage information data obtained from each carriage is as follows: Let n = k, 2k, 3k ..... mk, m is the number of cars of the inspected subway or train, k represents the number of sampling points in each car, and the distance between the received laser ranging array and the bottom of the inspected subway or train is S (x k )=y k , divide the mileage information data obtained from each carriage into the interval [(x0,x k ),(x k ,x 2k ),(x 2k ,x 3k ),.....,(x (m-1) k,x mk )], construct a cubic spline interpolation curve S in each interval, and the interpolation curve S satisfies the following conditions: Let each segmented small interval [x (m-1)k ,x mk ]S(x)=S j (x), where S j (x) is in the corresponding small interval [x (m-1)k ,x mk ], S(x) is a cubic function consisting of several S j (x) and satisfy the interpolation condition S(x i )=y i ,y i Indicates that at mileage x i The actual vehicle bottom distance obtained by the laser ranging array, i represents the serial number of the sampling point; Let the cubic equation system S(x), Continuous and for the cubic system S(x), Solve to get S(x), substitute n∈((m-1)k,mk) into S(x n ), that is, to interpolate the odometer points within the sampling interval of the laser ranging array.

8. The method for adaptive inspection of key components under a subway or train according to claim 1, characterized in that: In step 4, the original image is processed to obtain a clear image of the underside of the inspected subway or train. The specific steps are: distortion correction based on the height information of the lifting platform, specifically: Step 4.1, assuming that the average height of the lifting platform when scanning the chassis of the inspected subway or train is h0, the platform height of the lifting platform corresponding to the line scan data of the nth row of image acquisition mechanism is h, and the corresponding longitudinal distortion correction coefficient is set to K = αh / h0, where α is the lens zoom ratio of the line scan camera, which can be obtained by referring to the line scan camera parameter manual.

9. An adaptive inspection device for key components under a subway or train, which implements an adaptive inspection method for key components under a subway or train as claimed in any one of claims 1 to 8, characterized in that: The invention comprises a walking robot, a lifting platform and a laser ranging array are arranged on the top of the walking robot, an image acquisition mechanism is arranged on the upper surface of the lifting platform, and a main control box is also included, and the walking robot, the lifting platform, the laser ranging array and the image acquisition mechanism are electrically connected to the main control box.

Citation Information

Patent Citations

  • Inspection robot system and rail transit vehicle bottom detection method

    CN113450475A

  • Machine room patrol system and working method therefor

    WO2020107957A1