CN112860952A - Method and system for positioning and repairing parts of inspection robot - Google Patents
Method and system for positioning and repairing parts of inspection robot Download PDFInfo
- Publication number
- CN112860952A CN112860952A CN202011499197.XA CN202011499197A CN112860952A CN 112860952 A CN112860952 A CN 112860952A CN 202011499197 A CN202011499197 A CN 202011499197A CN 112860952 A CN112860952 A CN 112860952A
- Authority
- CN
- China
- Prior art keywords
- train
- axle
- position data
- positioning
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000007689 inspection Methods 0.000 title claims abstract description 42
- 238000012423 maintenance Methods 0.000 claims abstract description 17
- 230000008569 process Effects 0.000 claims abstract description 10
- 238000004422 calculation algorithm Methods 0.000 claims description 31
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000010365 information processing Effects 0.000 claims description 6
- 238000001514 detection method Methods 0.000 abstract description 7
- 238000007405 data analysis Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 4
- 238000007621 cluster analysis Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000013102 re-test Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5846—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using extracted text
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/907—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/909—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/63—Scene text, e.g. street names
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Tourism & Hospitality (AREA)
- Human Resources & Organizations (AREA)
- Library & Information Science (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Multimedia (AREA)
- Economics (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Software Systems (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of subway vehicle maintenance, and the embodiment particularly discloses a method and a system for positioning and maintaining parts by an inspection robot, wherein the method comprises the steps of acquiring a train number and a distance deviation value of a train stopping position relative to a preset stopping position; the method comprises the steps of judging whether a train type is an existing train type in a database or not according to the train number, calling historical axle position data of the train type from the database if the train type is the existing train type in the database, generating axle position data of the train according to the historical axle position data and a distance deviation value of a train stopping position relative to a preset stopping position, positioning to maintenance parts according to the axle position data of the train, constructing a train axle distribution relation by using a big data analysis method, calculating the axle position data according to the distance deviation value of the train stopping position and further positioning to the maintenance parts, enabling an inspection robot not to position each bogie in the operation process, and achieving the purpose of improving the overall detection efficiency.
Description
Technical Field
The invention relates to the technical field of subway vehicle maintenance, in particular to a method and a system for positioning and maintaining parts by an inspection robot.
Background
Subway vehicles are important components in urban rail transit, run on railway tracks and are responsible for carrying guests. The subway vehicle has a complex structure, and in order to ensure the daily safe operation of the subway vehicle, the subway vehicle needs to return to a special overhaul warehouse for overhaul after carrying tasks are executed every day. The overhaul content mainly comprises the loosening, the breakage, the loss, the deformation and the like of the parts of the subway vehicle. In order to improve the efficiency of maintenance work, it is an increasingly obvious trend to replace complete manual work with intelligent equipment. In the middle of the operation link of automatic maintenance, accurate positioning detection part is basic and key technique, and how to let the high-efficient accurate positioning detection part of inspection robot be the problem that needs to solve at present urgently.
Disclosure of Invention
In view of the above, the present application provides a method and a system for positioning and repairing a component by an inspection robot, which can solve or at least partially solve the existing problems.
In order to solve the technical problems, the technical scheme provided by the invention is a method for positioning and repairing components of an inspection robot, which comprises the following steps:
acquiring the train number and the distance deviation value of the train stop position relative to the preset stop position;
judging whether the train type is the existing train type in the database or not according to the train number, if so, calling historical axle position data of the train type from the database, generating each axle position data of the train according to the historical axle position data and the distance deviation value of the train stopping position relative to the preset stopping position, positioning each maintenance component according to each axle position data of the train, and if not, acquiring each axle position data of the train and storing the axle position data in the database.
Preferably, the method for acquiring the train number and the distance deviation value of the train stop position relative to the preset stop position includes:
the method comprises the steps of shooting a train number according to a preset path in a pre-stored electronic map and a preset shooting position to obtain a train number image, and extracting the train number from the train number image;
and calculating the distance deviation value between the train number character and the preset train number character position according to the position of the train number character in the train number image, and further calculating the distance deviation value between the train stop position and the preset stop position.
Preferably, the method for collecting and storing the position data of each axle of the train into the database includes:
taking a train bogie as a unit, and positioning a position of a first axle of a first bogie relative to a preset stop position as a train stop position;
sequentially reaching the axle position of each bogie to be positioned, completing a plurality of times of positioning of each axle, and obtaining a plurality of axle position data corresponding to each axle;
and storing the axle data with the most frequent occurrence in a plurality of axle position data corresponding to each axle into a database as the actual position data of the axle.
Preferably, the method for locating the inspection part by the inspection robot further comprises the following steps: generating historical axle position data of the train type; the method for generating the historical axle position data of the train type comprises the following steps:
storing a large amount of acquired position data of each axle of a train type into an array with the size of n;
performing clustering analysis by using a K-means algorithm, and taking each clustered clustering center as target data representing all position data in the array;
and calculating all clustering centers by using a mean algorithm, namely finding the actual position data of the axle.
Preferably, the method for performing cluster analysis by using a K-means algorithm and using each clustered cluster center as target data representing all position data in the array includes:
randomly selecting k initial clustering centers according to the classification number k, wherein the k clustering centers are from position data in the original array;
traversing each position data in the array, and allocating each position data to a clustering center closest to the position data to complete k initial clusters;
re-moving the clustering centers to the own center positions of the k initial clusters in the previous step, wherein the value of the position is changed with the value of the initial clustering center, and the new clustering center is used as the input of the next calculation;
redistributing all position data in the array by using the new clustering center, and continuously iterating and circulating until the minimum value of the target function appears;
wherein the objective function uses a square error criterion, i.e.
Where E represents the sum of squared differences of all the cluster objects, S is the cluster object, and Mi is the average of each cluster object of class Ci, that is:
the absolute value of Ci represents the number of the clustering objects of Ci, in each iteration process, each point needs to calculate the distance between the point and the clustering center once, and takes the shortest distance as the class to which the point belongs, the clustering algorithm ends and obtains the minimum value by E, namely the last clustering center stops after no movement, and the number of the initial clustering centers is controlled to be half of the size of the array.
Preferably, the method for calculating all cluster centers by using a mean algorithm includes: and solving the mean value of the clustering center which is finished by the last iteration as the axle position data, and sequentially finishing the position data calculation of all axles of the whole train.
The invention also provides a system for positioning and repairing the parts of the inspection robot, which comprises:
the positioning information acquisition module is used for acquiring the train number and the distance deviation value of the train stop position relative to the preset stop position;
and the positioning information processing module is used for judging whether the train type is the existing train type in the database or not according to the train number, if so, calling historical axle position data of the train type from the database, generating each axle position data of the train according to the historical axle position data and the distance deviation value of the train stopping position relative to the preset stopping position, positioning each maintenance component according to each axle position data of the train, and if not, acquiring each axle position data of the train and storing the axle position data into the database.
Preferably, the positioning information obtaining module includes:
the train number recognition unit is used for shooting the train number according to the preset shooting position when the train moves to the preset path in the pre-stored electronic map to obtain a train number image and extracting the train number from the train number image;
and the train position acquisition unit is used for calculating the distance deviation value between the train number character and the preset train number character position according to the position of the train number character in the train number image, and further calculating the distance deviation value between the train stop position and the preset stop position.
Preferably, the method for acquiring the position data of each axle of the train and storing the position data into the database by the positioning information processing module comprises the following steps:
taking a train bogie as a unit, and positioning a position of a first axle of a first bogie relative to a preset stop position as a train stop position;
sequentially reaching the axle position of each bogie to be positioned, completing a plurality of times of positioning of each axle, and obtaining a plurality of axle position data corresponding to each axle;
and storing the axle data with the most frequent occurrence in a plurality of axle position data corresponding to each axle into a database as the actual position data of the axle.
Preferably, the system for positioning and repairing the parts by the inspection robot further comprises a historical data generation module, and the method for generating the historical axle position data of the train type by the historical data generation module comprises the following steps:
storing a large amount of acquired position data of each axle of a train type into an array with the size of n;
performing clustering analysis by using a K-means algorithm, and taking each clustered clustering center as target data representing all position data in the array;
and calculating all clustering centers by using a mean algorithm, namely finding the actual position data of the axle.
Compared with the prior art, the beneficial effects of the method are detailed as follows: the method comprises the steps of obtaining the train number and the distance deviation value of the train stop position relative to a preset stop position; judging whether the train type is the existing train type in the database according to the train number, if so, calling historical train axis position data of the train type from the database, generating position data of each axle of the train according to the historical position data of the axle and the distance deviation value of the train stopping position relative to the preset stopping position, positioning each maintenance component according to each axle position data of the train, if not, collecting each axle position data of the train and storing the axle position data into a database, constructing a train axle distribution relation by utilizing a big data analysis method, the position data of each axle of the train is calculated according to the distance deviation value of the train stopping position relative to the preset stopping position, and then the data are positioned to each maintenance component for maintenance, so that the inspection robot does not need to position each bogie in the operation process, and the aim of improving the overall detection efficiency is fulfilled.
Drawings
In order to illustrate the embodiments of the present invention more clearly, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained by those skilled in the art without inventive effort.
Fig. 1 is a schematic flow chart of a method for positioning and repairing a component by an inspection robot according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a method for acquiring a train number and a distance deviation value of a train stop position relative to a preset stop position according to an embodiment of the present invention;
FIG. 3 is an internal network diagram of license plate character recognition according to an embodiment of the present invention;
fig. 4 is a schematic flow chart of a method for collecting position data of each axle of the train and storing the position data into a database according to an embodiment of the present invention;
fig. 5 is a flowchart illustrating a method for generating historical axle position data of a train model according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of a system for positioning and repairing components of an inspection robot according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without any creative work belong to the protection scope of the present invention.
In order to make the technical solutions of the present invention better understood by those skilled in the art, the present invention will be further described in detail with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1, an embodiment of the present invention provides a method for positioning and repairing a component by an inspection robot, including:
s11: acquiring the train number and the distance deviation value of the train stop position relative to the preset stop position;
s12: judging whether the train type is the existing train type in the database or not according to the train number, if so, calling historical axle position data of the train type from the database, generating each axle position data of the train according to the historical axle position data and the distance deviation value of the train stopping position relative to the preset stopping position, positioning each maintenance component according to each axle position data of the train, and if not, acquiring each axle position data of the train and storing the axle position data in the database.
Specifically, robot major structure includes is patrolled and examined to intelligence: laser navigation, a positioning component, a vision acquisition unit and the like. The intelligent inspection robot freely walks in the subway vehicle maintenance trench to realize image acquisition and analysis of key parts of the train, so as to judge abnormal conditions.
As shown in fig. 2, the method for acquiring the train number and the distance deviation value of the train stop position from the preset stop position in S11 includes:
s111: the method comprises the steps of moving to a preset shooting position according to a preset path in a prestored electronic map to shoot a train number to obtain a train number image, and extracting the train number from the train number image;
s112: and calculating the distance deviation value between the train number character and the preset train number character position according to the position of the train number character in the train number image, and further calculating the distance deviation value between the train stop position and the preset stop position.
Specifically, after the train stops and enters the garage, the train number is identified first. Before vehicle number identification, the inspection robot accurately positions the inspection robot, and a laser SLAM technology is adopted to scan an inspection environment to form a map so as to accurately obtain the relative position relation between the inspection robot and a detected vehicle. The inspection robot moves to a preset vehicle number position according to a preset electronic map to complete the identification of the vehicle type (image vehicle number identification) and the deviation of the vehicle number relative to the preset vehicle number character position. Namely, the accurate positioning of the stop position of the subway vehicle is completed once in the vehicle number identification process at this stage.
Specifically, as shown in fig. 3, the method for extracting the train number from the train number image by the inspection robot adopts a deep learning image processing technology, and the specific process includes: 1. the high-definition digital camera is used for completing the collection and storage of the train number image of the target train; 2. carrying out a series of preprocessing work (including image graying, image contrast enhancement and denoising, image edge detection, binarization processing and the like) on the stored train number image; 3. the function and the characteristics of the preprocessed train number image are pertinently suitable for recognizing the license plate characters by adopting a YOLOv3 target detection algorithm to process the preprocessed train number image. The internal network diagram of the license plate character recognition is shown in FIG. 3; 4. labeling and training the train number image by using labelImag software, and generating a configuration and weight file; 5. and calling the algorithm to realize train number identification.
Specifically, the method for calculating the distance deviation value between the train number character and the preset train number character position according to the position of the train number character in the train number image and further calculating the distance deviation value between the train stop position and the preset stop position is as follows: and calculating the pixel position deviation. And finding out the position information of the actual train according to the difference value between the pixel initial point pixel coordinate value of the train number character in the train number image and the pixel position in the preset train number image and by combining the relation between the preset train number image and the real stop position. . Note that, as shown in fig. 4: the method for collecting and storing the position data of each axle of the train into the database in the S12 comprises the following steps:
s121: taking a train bogie as a unit, and positioning a position of a first axle of a first bogie relative to a preset stop position as a train stop position;
s122: sequentially reaching the axle position of each bogie to be positioned, completing a plurality of times of positioning of each axle, and obtaining a plurality of axle position data corresponding to each axle;
s123: and storing the axle data with the most frequent occurrence in a plurality of axle position data corresponding to each axle into a database as the actual position data of the axle.
Specifically, the vehicle positioning determination positions the parking position of the first axle of the first bogie relative to the front end of the train (regarded as 0 point) in units of the car bogies, which represents the current position at which the car is parked. The inspection robot sequentially reaches the position of a first axle needing to be positioned of each bogie, a plurality of times of positioning of a single axle is completed, and a value with the most frequency of occurrence of the times of data is used as an actual space relative position value of the axle position (taking n positioned axles as an example). Because a large number of data sets are efficiently collected in a short time, errors can be reduced to a certain extent, and the positioning precision of the axle is improved.
The technology adopted for positioning the axle is to scan the cross section of the axle by a laser range finder, and a data set consisting of a plurality of times of lowest points of the shape (the lowest point of a semi-circular arc shape, the diameter of which is similar to the diameter of a real axle) of the scanned data is used as a selection range of the position of the axle. In this range (see table 1 below for a large number of approximate data sets as an example), the output position information (1-n value) of the axle with higher similarity is found in the multiple sets of data by analyzing the rule through an algorithm.
TABLE 1 axle positioning data example
It should be noted that, as shown in fig. 5, the method for positioning the inspection part by the inspection robot further includes: s10: and generating historical axle position data of the train type. Specifically, the method for generating the historical axle position data of the train type in S10 includes:
s101: storing a large amount of acquired position data of each axle of a train type into an array with the size of n;
s102: performing clustering analysis by using a K-means algorithm, and taking each clustered clustering center as target data representing all position data in the array;
s103: and calculating all clustering centers by using a mean algorithm, namely finding the actual position data of the axle.
It should be noted that, the method for performing cluster analysis by using the K-means algorithm in S102 and using each clustered cluster center as target data representing all position data in the array includes:
randomly selecting k initial clustering centers according to the classification number k, wherein the k clustering centers are from position data in the original array;
traversing each position data in the array, and allocating each position data to a clustering center closest to the position data to complete k initial clusters;
re-moving the clustering centers to the own center positions of the k initial clusters in the previous step, wherein the value of the position is changed with the value of the initial clustering center, and the new clustering center is used as the input of the next calculation;
redistributing all position data in the array by using the new clustering center, and continuously iterating and circulating until the minimum value of the target function appears;
wherein the objective function uses a square error criterion, i.e.
Where E represents the sum of squared differences of all the cluster objects, S is the cluster object, and Mi is the average of each cluster object of class Ci, that is:
the absolute value of Ci represents the number of the clustering objects of Ci, in each iteration process, each point needs to calculate the distance between the point and the clustering center once, and takes the shortest distance as the class to which the point belongs, the clustering algorithm ends and obtains the minimum value by E, namely the last clustering center stops after no movement, and the number of the initial clustering centers is controlled to be half of the size of the array.
It should be noted that, in step S103, the method for calculating all cluster centers by using the mean algorithm includes: and solving the mean value of the clustering center which is finished by the last iteration as the axle position data, and sequentially finishing the position data calculation of all axles of the whole train.
Specifically, in order to generate historical axle position data of a train type, the adopted method comprises the following steps: (1) storing a large amount of data scanned by each axle of a train into an array nums with the size of n, (2) carrying out clustering analysis by using a K-means algorithm, (3) taking each clustered clustering center as a target value which can mainly represent all values in the array, and (4) calculating all clustering centers by using the mean algorithm, namely finding the actual position value of the axle.
Specifically, (2) the method for performing cluster analysis by using the K-means algorithm includes: 1) randomly selecting k initial clustering centers according to the classification number k, wherein the k clustering centers are from values in the original array; 2) traversing each numerical value in the array, and allocating each numerical value to a clustering center closest to the numerical value to complete k initial clusters; 3) moving the clustering centers again to the own center positions of the k initial clusters in the step 3, wherein the value of the position is changed from the value of the initial clustering center, and the new clustering center is used as the input of the next calculation; 4) And reallocating all values in the array by using the new clustering center, and continuously iterating and circulating until the minimum value of the target function appears (the clustering center does not move).
Wherein, the objective function usually adopts the square error criterion, namely:
wherein E represents the sum of squared differences of all the cluster objects, S is the cluster object, and Mi is the average of each cluster object of class Ci, that is:
wherein the absolute value of Ci represents the number of Ci clustering objects, each point in each iteration process needs to calculate the distance between the point and the clustering center, and the shortest distance is taken outAnd E is used for obtaining the minimum value at the end of the clustering algorithm from the class to which the point belongs, namely the last clustering center stops after no movement occurs, and the number of the initial clustering centers is controlled to be half of the size of the array.
Specifically, (4) the method for calculating all the cluster centers by using the mean algorithm includes: and finally, taking the average value of the values of the sequential clustering centers to complete the positioning of the actual position of the axle, and sequentially and simultaneously completing the position calculation of all axles of the whole train.
After the historical axle position data of the train type is generated, after the train stops and enters a warehouse, the train number is identified, and when the identified train number and the identified train type are consistent with the previously detected data, the inspection robot automatically matches the position data of each axle of the train type in the database to the relative position information of all bogie axles of the current train, and the information is considered to be reliable, namely, the measurement and the positioning of the axles are not required again.
The invention has the following principle: the method has the advantages that the big data analysis technology is utilized, the axle distribution relation of a certain subway vehicle is constructed through a large amount of historical data, then the deviation information relative to the preset train number character position, which is given in the train number identification stage, is utilized, the deviation position of the train parking position relative to the preset parking position is automatically calculated, and further, the inspection robot does not need to position each bogie in the operation process, so that the aim of improving the overall detection efficiency is fulfilled.
The advantages of the invention are as follows: 1. the axle positioning mode is improved, in the past, after each train is put in storage, the inspection robot needs to perform axle positioning on each bogie once to effectively ensure that pictures of key components can be effectively shot in the subsequent steps, and after the axle positioning mode is improved, the detection positioning precision is more accurate by taking the average value of large data as a measurement template through multiple measurements of a single axle; 2. under original axletree locate mode, supposing to need retest once after every m train overall positioning to investigate the positioning accuracy problem, after improving, only need carry out m-1 times whole car location equivalently, and can prolong the interval time of retesting, improve axletree locate efficiency greatly, improve 90% in the aspect of axletree locate efficiency for traditional mode.
As shown in fig. 6, the present invention also provides a system for positioning and repairing components of an inspection robot, comprising:
the positioning information acquisition module 21 is used for acquiring the train number and the distance deviation value of the train stop position relative to the preset stop position;
and the positioning information processing module 22 is configured to determine whether the train type is an existing train type in the database according to the train number, if so, call historical axle position data of the train type from the database, generate each axle position data of the train according to the historical axle position data and a distance deviation value of the train stopping position relative to a preset stopping position, position each maintenance component according to each axle position data of the train, and if not, acquire each axle position data of the train and store the axle position data in the database.
It should be noted that the positioning information obtaining module 21 includes:
the train number recognition unit is used for shooting the train number according to the preset shooting position when the train moves to the preset path in the pre-stored electronic map to obtain a train number image and extracting the train number from the train number image;
and the train position acquisition unit is used for calculating the distance deviation value between the train number character and the preset train number character position according to the position of the train number character in the train number image, and further calculating the distance deviation value between the train stop position and the preset stop position.
It should be noted that the method for the positioning information processing module 22 to collect the position data of each axle of the train and store the data into the database includes:
taking a train bogie as a unit, and positioning a position of a first axle of a first bogie relative to a preset stop position as a train stop position;
sequentially reaching the axle position of each bogie to be positioned, completing a plurality of times of positioning of each axle, and obtaining a plurality of axle position data corresponding to each axle;
and storing the axle data with the most frequent occurrence in a plurality of axle position data corresponding to each axle into a database as the actual position data of the axle.
It should be noted that, the system for locating and repairing the component by the inspection robot further includes: and a historical data generation module. The method for generating the historical axle position data of the train type by the historical data generation module comprises the following steps:
storing a large amount of acquired position data of each axle of a train type into an array with the size of n;
performing clustering analysis by using a K-means algorithm, and taking each clustered clustering center as target data representing all position data in the array;
and calculating all clustering centers by using a mean algorithm, namely finding the actual position data of the axle.
For the description of the features in the embodiment corresponding to fig. 6, reference may be made to the related description of the embodiments corresponding to fig. 1 to fig. 5, which is not repeated here.
The method and the system for positioning and repairing the components of the inspection robot provided by the embodiment of the invention are described in detail above. The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
Claims (10)
1. A method for positioning and repairing parts of an inspection robot is characterized by comprising the following steps:
acquiring the train number and the distance deviation value of the train stop position relative to the preset stop position;
judging whether the train type is the existing train type in the database or not according to the train number, if so, calling historical axle position data of the train type from the database, generating each axle position data of the train according to the historical axle position data and the distance deviation value of the train stopping position relative to the preset stopping position, positioning each maintenance component according to each axle position data of the train, and if not, acquiring each axle position data of the train and storing the axle position data in the database.
2. The inspection robot positioning repair part method according to claim 1, wherein the method of obtaining the train number and the distance deviation value of the train stopping position relative to the preset stopping position includes:
the method comprises the steps of shooting a train number according to a preset path in a pre-stored electronic map and a preset shooting position to obtain a train number image, and extracting the train number from the train number image;
and calculating the distance deviation value between the train number character and the preset train number character position according to the position of the train number character in the train number image, and further calculating the distance deviation value between the train stop position and the preset stop position.
3. The inspection robot locating service component method according to claim 1, wherein the method of collecting and storing the position data of each axle of the train into the database includes:
taking a train bogie as a unit, and positioning a position of a first axle of a first bogie relative to a preset stop position as a train stop position;
sequentially reaching the axle position of each bogie to be positioned, completing a plurality of times of positioning of each axle, and obtaining a plurality of axle position data corresponding to each axle;
and storing the axle data with the most frequent occurrence in a plurality of axle position data corresponding to each axle into a database as the actual position data of the axle.
4. The inspection robot method for locating service components according to claim 1, further including: generating historical axle position data of the train type; the method for generating the historical axle position data of the train type comprises the following steps:
storing a large amount of acquired position data of each axle of a train type into an array with the size of n;
performing clustering analysis by using a K-means algorithm, and taking each clustered clustering center as target data representing all position data in the array;
and calculating all clustering centers by using a mean algorithm, namely finding the actual position data of the axle.
5. The inspection robot positioning inspection component method according to claim 4, wherein the clustering analysis using the K-means algorithm, and the method of using each clustered cluster center as target data representing all position data in the array includes:
randomly selecting k initial clustering centers according to the classification number k, wherein the k clustering centers are from position data in the original array;
traversing each position data in the array, and allocating each position data to a clustering center closest to the position data to complete k initial clusters;
re-moving the clustering centers to the own center positions of the k initial clusters in the previous step, wherein the value of the position is changed with the value of the initial clustering center, and the new clustering center is used as the input of the next calculation;
redistributing all position data in the array by using the new clustering center, and continuously iterating and circulating until the minimum value of the target function appears;
wherein the objective function uses a square error criterion, i.e.
Where E represents the sum of squared differences of all the cluster objects, S is the cluster object, and Mi is the average of each cluster object of class Ci, that is:
the absolute value of Ci represents the number of the clustering objects of Ci, in each iteration process, each point needs to calculate the distance between the point and the clustering center once, and takes the shortest distance as the class to which the point belongs, the clustering algorithm ends and obtains the minimum value by E, namely the last clustering center stops after no movement, and the number of the initial clustering centers is controlled to be half of the size of the array.
6. The inspection robot positioning service part method according to claim 4, wherein the method for calculating all cluster centers using a mean algorithm includes: and solving the mean value of the clustering center which is finished by the last iteration as the axle position data, and sequentially finishing the position data calculation of all axles of the whole train.
7. The utility model provides a system for patrolling and examining robot location maintenance part, its characterized in that includes:
the positioning information acquisition module is used for acquiring the train number and the distance deviation value of the train stop position relative to the preset stop position;
and the positioning information processing module is used for judging whether the train type is the existing train type in the database or not according to the train number, if so, calling historical axle position data of the train type from the database, generating each axle position data of the train according to the historical axle position data and the distance deviation value of the train stopping position relative to the preset stopping position, positioning each maintenance component according to each axle position data of the train, and if not, acquiring each axle position data of the train and storing the axle position data into the database.
8. The system for routing inspection robot positioning repair parts according to claim 7, wherein the positioning information acquisition module includes:
the train number recognition unit is used for shooting the train number according to the preset shooting position when the train moves to the preset path in the pre-stored electronic map to obtain a train number image and extracting the train number from the train number image;
and the train position acquisition unit is used for calculating the distance deviation value between the train number character and the preset train number character position according to the position of the train number character in the train number image, and further calculating the distance deviation value between the train stop position and the preset stop position.
9. The inspection robot positioning repair parts system according to claim 7, wherein the method for the positioning information processing module to collect the position data of each axle of the train and store the data in the database includes:
taking a train bogie as a unit, and positioning a position of a first axle of a first bogie relative to a preset stop position as a train stop position;
sequentially reaching the axle position of each bogie to be positioned, completing a plurality of times of positioning of each axle, and obtaining a plurality of axle position data corresponding to each axle;
and storing the axle data with the most frequent occurrence in a plurality of axle position data corresponding to each axle into a database as the actual position data of the axle.
10. The inspection robot positioning service part system according to claim 7, further comprising a historical data generation module, wherein the method for generating historical axle position data of train models by the historical data generation module comprises:
storing a large amount of acquired position data of each axle of a train type into an array with the size of n;
performing clustering analysis by using a K-means algorithm, and taking each clustered clustering center as target data representing all position data in the array;
and calculating all clustering centers by using a mean algorithm, namely finding the actual position data of the axle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011499197.XA CN112860952B (en) | 2020-12-17 | 2020-12-17 | Method and system for positioning and overhauling parts by inspection robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011499197.XA CN112860952B (en) | 2020-12-17 | 2020-12-17 | Method and system for positioning and overhauling parts by inspection robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112860952A true CN112860952A (en) | 2021-05-28 |
| CN112860952B CN112860952B (en) | 2023-06-20 |
Family
ID=75997514
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011499197.XA Active CN112860952B (en) | 2020-12-17 | 2020-12-17 | Method and system for positioning and overhauling parts by inspection robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112860952B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113448333A (en) * | 2021-06-25 | 2021-09-28 | 北京铁道工程机电技术研究所股份有限公司 | Bottom routing inspection positioning method and device based on sensor combination and electronic equipment |
| CN115439451A (en) * | 2022-09-09 | 2022-12-06 | 哈尔滨市科佳通用机电股份有限公司 | Denoising detection method for spring supporting plate of railway wagon bogie |
| CN115900553A (en) * | 2023-01-09 | 2023-04-04 | 成都盛锴科技有限公司 | Compound positioning method and system for train inspection robot |
| CN116039699A (en) * | 2023-03-28 | 2023-05-02 | 湖北工业大学 | System for replacing bogie of wide-narrow rail train |
| CN116090802A (en) * | 2023-04-12 | 2023-05-09 | 成都盛锴科技有限公司 | Train inspection task intelligent distribution and scheduling system oriented to vehicle bottom part identification |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101020463A (en) * | 2007-03-22 | 2007-08-22 | 北京交通大学 | Precise fault locating method and device for railway track checking car |
| CN102107777A (en) * | 2009-12-24 | 2011-06-29 | 同方威视技术股份有限公司 | Vehicle conveying device and vehicle radiation detecting system provided with same |
| CN106468914A (en) * | 2015-08-21 | 2017-03-01 | 苏州华兴致远电子科技有限公司 | Train overhaul method and system |
| CN206594518U (en) * | 2015-08-21 | 2017-10-27 | 苏州华兴致远电子科技有限公司 | Train overhaul system |
| US20180035606A1 (en) * | 2016-08-05 | 2018-02-08 | Romello Burdoucci | Smart Interactive and Autonomous Robotic Property Maintenance Apparatus, System, and Method |
| CN109176513A (en) * | 2018-09-04 | 2019-01-11 | 北京华开领航科技有限责任公司 | A kind of method for inspecting and cruising inspection system of intelligent inspection robot |
| CN109738213A (en) * | 2019-02-03 | 2019-05-10 | 北京新联铁集团股份有限公司 | Rail transit rolling stock inspection pose detection system and its method |
| CN110259226A (en) * | 2019-06-28 | 2019-09-20 | 陈载发 | Multistory parking frame installation vehicle hoisting mechanism, multistory parking frame installation and Transport Vehicle method |
| CN210879665U (en) * | 2019-11-07 | 2020-06-30 | 成都铁安科技有限责任公司 | Comprehensive positioning system of subway vehicle train inspection robot |
| CN111554005A (en) * | 2020-04-27 | 2020-08-18 | 浙江库科自动化科技有限公司 | Intelligent inspection method for railway freight train |
| CN111845864A (en) * | 2020-06-12 | 2020-10-30 | 佛山中国空间技术研究院创新中心 | Rail train identification and positioning method, system, device and storage medium |
-
2020
- 2020-12-17 CN CN202011499197.XA patent/CN112860952B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101020463A (en) * | 2007-03-22 | 2007-08-22 | 北京交通大学 | Precise fault locating method and device for railway track checking car |
| CN102107777A (en) * | 2009-12-24 | 2011-06-29 | 同方威视技术股份有限公司 | Vehicle conveying device and vehicle radiation detecting system provided with same |
| CN106468914A (en) * | 2015-08-21 | 2017-03-01 | 苏州华兴致远电子科技有限公司 | Train overhaul method and system |
| CN206594518U (en) * | 2015-08-21 | 2017-10-27 | 苏州华兴致远电子科技有限公司 | Train overhaul system |
| US20180035606A1 (en) * | 2016-08-05 | 2018-02-08 | Romello Burdoucci | Smart Interactive and Autonomous Robotic Property Maintenance Apparatus, System, and Method |
| CN109176513A (en) * | 2018-09-04 | 2019-01-11 | 北京华开领航科技有限责任公司 | A kind of method for inspecting and cruising inspection system of intelligent inspection robot |
| CN109738213A (en) * | 2019-02-03 | 2019-05-10 | 北京新联铁集团股份有限公司 | Rail transit rolling stock inspection pose detection system and its method |
| CN110259226A (en) * | 2019-06-28 | 2019-09-20 | 陈载发 | Multistory parking frame installation vehicle hoisting mechanism, multistory parking frame installation and Transport Vehicle method |
| CN210879665U (en) * | 2019-11-07 | 2020-06-30 | 成都铁安科技有限责任公司 | Comprehensive positioning system of subway vehicle train inspection robot |
| CN111554005A (en) * | 2020-04-27 | 2020-08-18 | 浙江库科自动化科技有限公司 | Intelligent inspection method for railway freight train |
| CN111845864A (en) * | 2020-06-12 | 2020-10-30 | 佛山中国空间技术研究院创新中心 | Rail train identification and positioning method, system, device and storage medium |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113448333A (en) * | 2021-06-25 | 2021-09-28 | 北京铁道工程机电技术研究所股份有限公司 | Bottom routing inspection positioning method and device based on sensor combination and electronic equipment |
| CN113448333B (en) * | 2021-06-25 | 2024-02-06 | 北京铁道工程机电技术研究所股份有限公司 | Bottom inspection positioning method and device based on sensor combination and electronic equipment |
| CN115439451A (en) * | 2022-09-09 | 2022-12-06 | 哈尔滨市科佳通用机电股份有限公司 | Denoising detection method for spring supporting plate of railway wagon bogie |
| CN115900553A (en) * | 2023-01-09 | 2023-04-04 | 成都盛锴科技有限公司 | Compound positioning method and system for train inspection robot |
| CN116039699A (en) * | 2023-03-28 | 2023-05-02 | 湖北工业大学 | System for replacing bogie of wide-narrow rail train |
| CN116090802A (en) * | 2023-04-12 | 2023-05-09 | 成都盛锴科技有限公司 | Train inspection task intelligent distribution and scheduling system oriented to vehicle bottom part identification |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112860952B (en) | 2023-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112860952A (en) | Method and system for positioning and repairing parts of inspection robot | |
| CN110378869B (en) | Steel rail fastener abnormity detection method with automatic sample marking function | |
| CN109658387B (en) | Method for detecting defects of pantograph carbon slide plate of electric train | |
| CN111489339B (en) | Method for detecting backup and main defects of high-speed railway locator bolts | |
| CN112330591A (en) | Steel rail surface defect detection method and device capable of achieving sample-less learning | |
| CN108846331B (en) | Video identification method for judging whether screw fastener of motor train unit chassis falls off or not | |
| CN112949715A (en) | SVM (support vector machine) -based rail transit fault diagnosis method | |
| CN106251361A (en) | A kind of rail surface defects image self-adapting division method | |
| CN110047070B (en) | Method and system for identifying rail wear degree | |
| CN111311567A (en) | Method for identifying fastener and steel rail diseases of track line image | |
| CN113487533B (en) | Part assembly quality digital detection system and method based on machine learning | |
| CN115482195B (en) | Train part deformation detection method based on three-dimensional point cloud | |
| CN114049356B (en) | Method, device and system for detecting structure apparent crack | |
| CN111079748A (en) | Method for detecting oil throwing fault of rolling bearing of railway wagon | |
| CN112115770A (en) | Method and system for identifying autonomous inspection defects of unmanned aerial vehicle of overhead line | |
| CN103679214A (en) | Vehicle detection method based on online area estimation and multi-feature decision fusion | |
| CN115546223A (en) | Method and system for detecting loss of fastening bolt of equipment under train | |
| CN114693617A (en) | Method, system, equipment and storage medium for detecting defects of railway steel rails | |
| CN113295142A (en) | Terrain scanning analysis method and device based on FARO scanner and point cloud | |
| CN117058348A (en) | Three-dimensional point cloud electric tower segmentation method based on geometric feature and registration joint optimization | |
| CN116681653A (en) | Three-dimensional point cloud extraction method and extraction system | |
| CN116681912A (en) | Rail gauge detection method and device for railway turnout | |
| CN116385292A (en) | IMU-assisted LiDAR dynamic point cloud eliminating method | |
| CN112036403B (en) | Intelligent detection method for missing of bolt pin of power transmission tower based on attention mechanism | |
| Evstafev et al. | Surface Defect Detection and Recognition Based on CNN |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |