CN113581244B - Intelligent skate track recognition system and method based on information fusion - Google Patents
Intelligent skate track recognition system and method based on information fusion Download PDFInfo
- Publication number
- CN113581244B CN113581244B CN202110821256.9A CN202110821256A CN113581244B CN 113581244 B CN113581244 B CN 113581244B CN 202110821256 A CN202110821256 A CN 202110821256A CN 113581244 B CN113581244 B CN 113581244B
- Authority
- CN
- China
- Prior art keywords
- positioning
- track
- stock
- identification
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 105
- 230000004927 fusion Effects 0.000 title claims abstract description 40
- 238000001914 filtration Methods 0.000 claims abstract description 37
- 238000012544 monitoring process Methods 0.000 claims abstract description 33
- 238000007619 statistical method Methods 0.000 claims abstract description 21
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 186
- 229910052742 iron Inorganic materials 0.000 claims description 93
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000013507 mapping Methods 0.000 claims description 9
- 239000008186 active pharmaceutical agent Substances 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 6
- 239000002184 metal Substances 0.000 claims description 4
- 229910052751 metal Inorganic materials 0.000 claims description 4
- 239000000523 sample Substances 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 9
- 230000002411 adverse Effects 0.000 abstract description 2
- 230000008094 contradictory effect Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61K—AUXILIARY EQUIPMENT SPECIALLY ADAPTED FOR RAILWAYS, NOT OTHERWISE PROVIDED FOR
- B61K9/00—Railway vehicle profile gauges; Detecting or indicating overheating of components; Apparatus on locomotives or cars to indicate bad track sections; General design of track recording vehicles
- B61K9/08—Measuring installations for surveying permanent way
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention discloses an intelligent skate track recognition system based on information fusion, which comprises more than 4 positioning satellites, a reference station, a 4G private base station, a plurality of pairs of intelligent shoes, a positioning terminal and a monitoring upper computer, wherein the positioning satellites are arranged on the reference station; the positioning terminal fuses the recognition results of the two stock track recognition methods, namely the precision weight statistical method and the variance-varying Gaussian filtering method, comprehensively judges and obtains the final stock track recognition result so as to realize accurate recognition of the stock track. When the stock track identification is performed by using the precision weight statistical method, the precision weight of the positioning data is reasonably introduced, the effect of the high-precision data on the identification effect is improved, the adverse effect of the low-precision data on the identification effect is effectively reduced, and the identification result is more accurate and effective. The stock way identification method based on information fusion is provided, and the problem that the results of the two identification methods are contradictory is effectively solved.
Description
Technical Field
The invention relates to the technical field of railways, in particular to an intelligent shoe track recognition system and an intelligent shoe track recognition method, which are suitable for realizing accurate recognition of a railway track where intelligent shoes are positioned in the process of anti-slip operation of trains in a railway station.
Background
As railway lines increase and coverage areas become wider, intelligent shoes are increasingly used, and intelligent shoes have become important devices for protecting personnel and vehicles.
At present, most traditional iron shoes do not have the stock track recognition function, and even if part of intelligent iron shoes can perform stock track recognition, the recognition accuracy is not high, and the main reasons and the existing problems are as follows:
Because GPS positioning of the intelligent iron shoes is greatly influenced by various factors such as environment, but the railway station is severe in environment, the electromagnetic environment is complex, a large amount of electromagnetic interference can reduce the positioning precision of the positioning terminal, and because a human body, a train, a building and the like generate shielding effect on satellite positioning signals, the positioning result of the positioning terminal is deviated.
The intelligent shoes are used for track identification, and the position is determined mainly through high-precision positioning information provided by the intelligent shoes, so that the positioning deviation problem of GPS positioning can influence a positioning terminal to obtain accurate positioning data, certain deviation exists between the positioning data obtained by the intelligent shoes and a true value, even the intelligent shoes receive incorrect positioning data, the intelligent shoes are used for track identification errors, and disorder on intelligent shoe management is caused.
Disclosure of Invention
The invention aims to solve the problems, and provides an intelligent system and method for identifying the railway track of the intelligent shoe, which are suitable for accurately identifying the railway track where the intelligent shoe is positioned in the process of anti-slip operation of a train in a railway station yard.
The technical scheme of the invention is as follows:
An intelligent skate strand identification system based on information fusion comprises more than 4 positioning satellites, a reference station, a 4G private base station, a plurality of pairs of intelligent shoes, a positioning terminal and a monitoring upper computer;
The satellite refers to a satellite which establishes communication with the reference station and the positioning terminal in the Beidou positioning system, establishes communication with the reference station and the positioning terminal, and sends positioning signals to the reference station and the positioning terminal;
the reference station receives a positioning signal of a positioning satellite, so as to calculate self positioning information, calculate with self mapping position coordinates to obtain differential data, and send the differential data to the positioning terminal through the 4G private base station for positioning correction;
The positioning terminal is placed beside the intelligent iron shoe and is used in a binding mode with the intelligent iron shoe, the positioning terminal receives positioning signals of positioning satellites to determine self positioning information, and high-precision positioning information is provided for the intelligent iron shoe based on a differential data correction method; the positioning terminal fuses the recognition results of the two stock track recognition methods, namely the precision weight statistical method and the variance-variable Gaussian filtering method, comprehensively judges and obtains the final stock track recognition result so as to realize accurate recognition of the stock track; the positioning terminal sends the positioning result to the monitoring upper computer through the 4G private base station;
The 4G private base station receives the reference station differential data and forwards the reference station differential data to the positioning terminal, and simultaneously receives the positioning data corrected by the positioning terminal; on the other hand, the 4G private base station interacts with the monitoring upper computer data through the wired Ethernet interface, and the received positioning data after the positioning terminal is corrected is sent to the upper computer.
Preferably, the differential data correction method is as follows: and correcting the current self-position observed value determined by the positioning terminal based on the differential data obtained from the reference station, namely adding the current self-position observed value and the current self-position observed value to obtain high-precision positioning data corrected by the positioning terminal.
The invention also discloses an intelligent iron shoe strand identification method based on information fusion, which is based on any system and comprises the following steps: the method comprises four steps of system initialization, terminal and iron shoe identification matching, station track positioning and identification by a terminal machine and upper computer update state monitoring, and comprises the following specific processes:
step 1: system initialization
The system initialization comprises the initialization of a Beidou positioning module in a positioning terminal and the initialization of stock track information of a train yard, wherein the initialization comprises the geographical information of a stock track center line and the number k of stock track of the yard, which are obtained according to geographical mapping information of a yard track, and the number i=5 of positioning data sets are acquired before and after a reference moment in the first recognition;
step 2: terminal and skate identification matching
The railway station staff performs anti-slip operation, performs intelligent skate placement operation, performs RFID radio frequency identification with the intelligent skate by using a positioning terminal, and performs matching with the skate;
step 3: locating terminal locating and identifying stock way
After the positioning terminal is successfully matched with the intelligent iron shoe, taking the RFID radio frequency identification matching time as a reference time, respectively taking i seconds of high-precision positioning data before and after to form a data sequence with the length of 2 x i+1, and sequentially carrying out precision weight statistical method strand identification and variance-variable Gaussian filtering method strand identification by the identification system by utilizing the group of positioning data, wherein the strand identification result is determined and output when the two identification results are consistent; if the identification results are inconsistent, the number of the positioning data acquisition groups i=i+5, and the identification results of the two methods are inconsistent after the step 3 is circulated for 3 times, fusion judgment is carried out on the two identification results by using a DS evidence reasoning algorithm, and the information fusion judgment result is used as a final identification result;
step 4: monitoring update status of host computer
After the stock way of the iron shoe is successfully identified, the monitoring upper computer detects whether the stock way has a train or not, if the monitoring upper computer displays that the stock way has a train, the iron shoe is normally on line, and the monitoring upper computer updates the state of the train and the iron shoe; if the stock road is displayed without the train, whether the iron shoes are placed in place or not is confirmed again, and whether the train number information is required to be recorded by an upper computer or whether the iron shoes are replaced or not is judged.
Specifically, the specific steps of matching and identifying the positioning terminal and the iron shoe in the step 2 are as follows:
2-1), firstly, taking out the intelligent iron shoes and the positioning terminal from the anti-slip device box by on-site operators, transferring the intelligent iron shoes to an anti-slip operation place, and placing the intelligent iron shoes on a preset rail;
2-2), matching and binding the positioning terminal with the intelligent iron shoe through RFID;
2-3), judging whether the positioning terminal is successfully matched with the iron shoe or not: if the matching is unsuccessful, executing the step 2-2); if the matching is successful, the step 3 is entered.
Specifically, the specific steps of locating and identifying the stock track by the locating terminal in the step 3 are as follows:
3-1), after the positioning terminal is successfully matched with the iron shoe, placing the positioning terminal beside the iron shoe, taking the RFID (radio frequency identification) time of the terminal machine and the iron shoe as a reference time, respectively taking i groups of high-precision positioning data from front to back to form a positioning data sequence with the length of 2 x i+1, and executing the step 3-2);
3-2), carrying out stock track identification by using an accuracy weight statistical method: respectively carrying out track recognition on the 2x i+1 group positioning data, distributing the precision weights of the track recognition data to the recognized tracks, finally counting the precision weights of all tracks, taking the track with the highest precision weight statistic value as a track recognition result, and executing the step 3-3) after the track recognition result is obtained;
3-3), carrying out stock track identification by using a variational Gaussian filtering method: carrying out variational Gaussian filtering calculation on the 2x i+1 group of positioning data to obtain variational Gaussian filtered coordinates (x ', y'), calculating the vertical distance between the variational Gaussian filtered coordinates and the center line of each track, wherein the track with the nearest distance is the track recognition result, and executing the step 3-4 after the track recognition result is obtained;
3-4), judging whether the stock way identification results of the two methods are consistent or not: if the results are consistent, uploading the stock channel identification results, and entering step 4; if the results are inconsistent, executing the step 3-5);
3-5), i=i+5, enlarging the time range of positioning data acquisition, and judging whether i is less than or equal to 15; if i is less than or equal to 15, executing the step 3-1); if i >15, executing the step 3-6);
3-6), adopting a recognition result fusion judgment method based on different recognition methods, respectively calculating the credibility of the two methods on different recognition results, utilizing a DS evidence reasoning algorithm to fusion-calculate and judge the stock track recognition result, taking the fusion judgment result as a final result, uploading, and entering the step 4.
Specifically, the specific steps for identifying the tracks by using the precision weight statistical method in the step 3-2) are as follows:
3-2-1), taking card swiping time as a reference, respectively taking i groups of positioning data before and after to obtain a positioning data sequence with the length of 2 x i+1, and accurately distributing weights to the 2 x i+1 groups of positioning data, wherein the weight distribution is specifically shown in the following table:
Wherein delta j is the positioning data precision of the jth positioning data, omega j is the precision weight of the jth positioning data; let j=1, g 1=G2=···=Gk=0,G1、G2···Gk be the accuracy weight statistics of k tracks, respectively, and execute step 3-2-2);
3-2-2), taking the j-th group of positioning data (x j,yj) in the positioning data sequence, calculating the vertical distance L 1···Lk between the positioning point (x j,yj) and the center line of each track, and executing the step 3-2-3);
3-2-3), determining a minimum value L m, m epsilon [1, k ] in L 1···Lk, namely that the distance between the locating point and the stock track m is the minimum, and the accuracy weight of the locating data is omega j, counting the accuracy weight on the corresponding stock track weight, enabling G m=Gm+ωj, j=j+1, and executing the step 3-2-4);
3-2-4), judging whether j is less than or equal to 2 x i+1: if j is less than or equal to 2 x i+1, executing the step 3-2-2); if j >2 x i+1, executing step 3-2-5);
3-2-5), determining that the maximum value in G 1、G2···Gk is G N, N epsilon [1, k ], and ending when the stock track identification result of the precision weight statistics method is the stock track N.
Specifically, the specific steps of the stock track identification using the variational Gaussian filtering method in the step 3-3) are as follows:
3-3-1), taking the card swiping time as a reference, respectively taking i groups of positioning data before and after to obtain a positioning data sequence P 1(x1,y1)、P2(x2,y2)、···、P2i+1 (x, y) with the length of (2 x i+1), giving a Gaussian weight value to each group of data in the sequence, and changing coordinate data after variance Gaussian filtering into (x ', y'), wherein the calculation mode of x 'and y' is as follows:
Wherein x j、yj and sigma correspond to x coordinate, y coordinate and Gaussian standard deviation of the j-th set of positioning data respectively. For positioning data sequences with different lengths, selecting different sigma values, changing weight distribution, when i is gradually increased and the length of the positioning data sequence is enlarged, reducing the sigma value to improve the weight of positioning data near the reference moment, reducing the weight of positioning data at the edge moment, and when i=5, taking sigma=10; when i=10, σ=4.5 is taken; when i=15, σ=4 is taken;
carrying out variational Gaussian filtering on all positioning data in the sequence to obtain coordinates (x 'and y'), and executing the step 3-3-2);
3-3-2), calculating the vertical distance L 1···Lk between the coordinate point (x ', y') and the center line of each track, and then executing the step 3-3-3);
3-3-3), determining the minimum value in L 1···Lk as L M, M epsilon [1, k ], and the track recognition result as track M,
And (5) ending.
Specifically, the algorithm for calculating the vertical distance between the positioning coordinate point and the center line of each track is specifically as follows:
(1) According to the rail geographical mapping data of the train station, determining the geographical position of the track center line, and determining the coordinates of two end points of the track center line as A (X 1,Y1),B(X2,Y2);
(2) The positioning point coordinates P (X, Y), the vertical distance between the positioning point P and the track center line is as follows:
specifically, the method for determining the fusion of the two stock track recognition results in step 3-6) is to perform the stock track recognition respectively by using a precision weight statistics method and a variance-varying Gaussian filtering method, and perform fusion calculation to determine the stock track recognition result under the condition that the stock track recognition results are inconsistent, wherein the specific fusion determination process is as follows:
recording the recognition result of the precision weight statistical method as p and the recognition result of the variance-changing Gaussian filtering method as q; p, q epsilon [1, k ] and p not equal q, namely the identification results obtained by the two identification methods are different;
The credibility of the accuracy weight statistical method for the strand identification result p and q is respectively as follows:
wherein, G p,Gq is the sum of the precision weight statistics values of the p tracks and the q tracks respectively;
The credibility of the variational Gaussian filtering method for the strand identification result p and q is respectively as follows:
based on the recognition results of the two stock way recognition methods, the reliability m (p) of the matched stock way p and the reliability m (q) of the matched stock way q are calculated in a fusion mode, wherein:
Comparing m (p) with m (q), and if m (p) is greater than m (q), determining that the track p is matched by fusion; if m (p) < m (q), the fusion judgment is that the stock way q is matched; if m (p) =m (q), the likelihood of the hint etc. matches the track p, the track q.
Specifically, the specific steps for monitoring the update status of the upper computer in step 4 are as follows:
4-1), monitoring whether the upper computer has a car or not according to the intelligent skate track recognition result: if the upper computer displays that the stock road has no vehicles, executing the step 4-2); if the upper computer displays that the track has a car, executing the step 4-4);
4-2), detecting whether the skate is put in place: the placement of the skate in place includes two conditions: 1. determining that the shoe is on track by a metal detector on the shoe; 2. detecting the distance between the skate and the wheel by using an acoustic ranging probe on the skate, and determining that the skate is placed in an anti-slip range; when the iron shoes are on the track and in the anti-slip range and the iron shoes are put in place, executing the step 4-3); if any condition is not met, the iron shoes are not placed in place, and the iron shoes need to be replaced, and the step 2 is executed;
4-3), putting the iron shoe in place, and executing the step 4-4 after the upper computer does not timely record the train number information and the train number information is complemented;
4-4), the upper computer is monitored to update the state of the train and the intelligent iron shoes on the track, and the process is finished.
The beneficial effects of the invention are that
Compared with the prior art, the invention has the beneficial effects that:
1. the intelligent shoe track recognition system and method based on information fusion are suitable for intelligent track recognition by applying intelligent shoes in a railway station; according to the intelligent shoe track recognition method, the intelligent shoe positioning system and the track recognition method are optimized, so that the problems of large positioning deviation and inaccurate track recognition of the intelligent shoe in a complex train station environment are solved, and the accuracy of intelligent shoe track recognition is effectively improved.
2. The positioning terminal is used in binding with the intelligent iron shoe, and corrects the positioning data of the positioning terminal by acquiring the differential correction data of the reference station, so that high-precision positioning data is provided for the intelligent iron shoe.
3. According to the track identification method, the matching time of the positioning terminal and the RFID of the intelligent iron shoe is taken as the reference time, i groups of positioning data are collected front and back respectively to form a positioning data sequence with the length of (2 x i+1), the track identification is carried out by using the groups of data, and the data reliability is high.
4. According to the stock way identification method, the stock way identification results obtained through a reasonable fusion precision weight statistical method and a variance-varying Gaussian filtering method are used as final identification results when the two method identification results are consistent; if the results are inconsistent, the acquisition range of the positioning data is enlarged, the stock way identification is circularly carried out for a plurality of times, and the probability of the stock way identification error is reduced.
5. When the stock track identification is performed by using the precision weight statistical method, the precision weight of the positioning data is reasonably introduced, the effect of the high-precision data on the identification effect is improved, the adverse effect of the low-precision data on the identification effect is effectively reduced, and the identification result is more accurate and effective.
6. When the stock track identification is performed by using a variance-varying Gaussian filtering method, based on a selected positioning data sequence with the length of (2 x i+1), not only the data sequence is processed according to Gaussian weights, but also the weight distribution can be dynamically and automatically changed for positioning data sequences with different (2 x i+1) lengths, when i is gradually increased, the length of the positioning data sequence is increased, the sigma value is reduced to improve the weight of positioning data near the reference moment, the weight of positioning data at the edge moment is reduced, and the error caused by low credibility data at the edge moment can be effectively reduced; when i is gradually reduced and the length of the positioning data sequence is reduced, the useful information at the edge moment can be effectively utilized by improving the sigma value, and finally the purpose of improving the accuracy of stock track identification by a variance-varying Gaussian filtering method is achieved.
7. If the acquisition range of the positioning data exceeds a set maximum value, namely, the identification results obtained by the accuracy weight statistical method and the variance Gaussian filtering method are still inconsistent when i is greater than 15, carrying out fusion judgment by adopting a DS evidence reasoning algorithm, respectively calculating the credibility of the two identification results, taking the result with higher credibility as the final identification result, effectively solving the problem that the results of the two identification methods are contradictory, fusing multiple aspects of information in the process of carrying out credibility calculation, and leading the final identification result to be more credible.
8. After the monitoring upper computer obtains the stock way recognition result, detect this stock way train state, through metal detection device, sound wave range unit on the intelligent skate, confirm whether intelligent skate is put in place, further judge whether need carry out operations such as supplementary recording train number or repositioning the skate, can prevent effectively that the train accident from causing because of the skate is placed erroneously, ensure train operation safety.
Drawings
FIG. 1 is a diagram showing the overall construction of an intelligent railway track recognition system according to the present invention.
FIG. 2 is an overall workflow diagram of the intelligent skate track identification method of the present invention.
FIG. 3 is a workflow diagram of the accuracy weight statistics stock track identification in the present invention.
Fig. 4 is a flowchart of the operation of the variational gaussian filter stock-in-process of the present invention.
Fig. 5 is a schematic diagram of the present invention for calculating the distance between a positioning coordinate point and the track centerline.
Detailed Description
As shown in FIG. 1, the intelligent skate track identification system comprises more than 4 positioning satellites, a reference station, a 4G private base station, a plurality of pairs of intelligent shoes, a positioning terminal and a monitoring upper computer.
The satellite refers to a satellite which establishes communication with the reference station and the positioning terminal in the Beidou positioning system, and the satellite establishes communication with the reference station and the positioning terminal and sends positioning signals to the reference station and the positioning terminal.
The reference station receives the positioning signals of 4 satellites and more, can calculate and obtain self positioning information, calculates and obtains differential data with self mapping position coordinates, and sends the differential data to the positioning terminal through the 4G private base station for positioning correction.
The positioning terminal is placed beside the intelligent iron shoe and is bound with the intelligent iron shoe for use, the positioning terminal receives positioning signals of 4 or more satellites to determine self positioning information, differential data obtained from the reference station corrects the self position observation value, namely, the two are added to obtain high-precision positioning data corrected by the positioning terminal. The positioning terminal takes the high-precision positioning data of the positioning terminal as the high-precision positioning data of the intelligent iron shoe, provides high-precision positioning information for the intelligent iron shoe, realizes the functions of stock way judgment, theft alarm and the like of the intelligent iron shoe, sends the positioning result to the monitoring upper computer through the 4G private base station, finally realizes the high-precision positioning of the intelligent iron shoe, and improves the monitoring function of the intelligent anti-slip system.
The intelligent skate comprises an intelligent box and a skate body, wherein the intelligent box is added on the basis of the traditional mechanical skate to realize intelligent anti-slip.
The 4G private base station receives the reference station differential data and forwards the reference station differential data to the positioning terminal, and simultaneously receives the positioning data after the positioning terminal corrects the positioning data, (used for expanding the signal coverage); on the other hand, the 4G private base station interacts with the monitoring upper computer data through the wired Ethernet interface, and the received positioning data after the positioning terminal is corrected is sent to the upper computer.
The upper computer is monitored for 24 hours without interruption to monitor the state of the placed iron shoes, and the stock way where the iron shoes are is displayed.
When the intelligent skate is used by binding the positioning terminal and the intelligent skate, the intelligent skate and the positioning terminal are taken out of the anti-slip device box and then transferred to an anti-slip operation place when on-site operators perform anti-slip operation; then carrying out anti-slip operation such as placing the iron shoes, after finishing the anti-slip operation, carrying out matching binding on the positioning terminal and the intelligent iron shoes through RFID radio frequency identification, hanging and attaching the positioning terminal on a train carriage, then pressing a button on the equipment, and uploading high-precision positioning data to a server by the positioning terminal to finish the track positioning function of the iron shoes; after the anti-slip operation is finished, an operator withdraws the iron shoes from the wheels, removes the positioning terminal from the train carriage, bounces the keys, and finally returns the iron shoes to the anti-slip device box, and the operation is finished.
As shown in fig. 2, an intelligent method for identifying a stock track of an iron shoe is divided into: the method comprises four steps of system initialization, terminal and iron shoe identification matching, station track positioning and identification by a terminal machine and upper computer update state monitoring, and comprises the following specific processes:
step 1: system initialization
The system initialization comprises the initialization of a Beidou positioning module in a positioning terminal and the initialization of stock track information of a train station, wherein the initialization comprises the geographical information of a stock track center line and the number k (the maximum k value is 8) of the stock track center line obtained according to the geographical mapping information of the station track, and the number i=5 of positioning data sets is acquired before and after a reference moment in the first recognition;
step 2: terminal and skate identification matching
The railway station staff performs anti-slip operation, performs intelligent skate placement operation, performs RFID radio frequency identification with the intelligent skate by using a positioning terminal, and performs matching with the skate;
The specific steps of matching and identifying the positioning terminal and the iron shoe in the step 2 are as follows:
2-1), firstly, taking out the intelligent iron shoes and the positioning terminal from the anti-slip device box by on-site operators, transferring the intelligent iron shoes to an anti-slip operation place, and placing the intelligent iron shoes on a preset rail;
2-2), matching and binding the positioning terminal with the intelligent iron shoe through RFID;
2-3), judging whether the positioning terminal is successfully matched with the iron shoe or not: if the matching is unsuccessful, executing the step 2-2); if the matching is successful, the step 3 is entered;
step 3: locating terminal locating and identifying stock way
After the positioning terminal is successfully matched with the intelligent iron shoe, taking the RFID radio frequency identification matching time as a reference time, respectively taking i seconds of high-precision positioning data before and after to form a data sequence with the length of (2 x i+1), and sequentially carrying out precision weight statistical method track identification and variance-variable Gaussian filtering method track identification by the identification system by utilizing the group of positioning data, wherein the track identification result is determined and output when the two identification results are consistent; if the identification results are inconsistent, the number of the positioning data acquisition groups i=i+5, and the identification results of the two methods are inconsistent after the step 3 is circulated for 3 times, fusion judgment is carried out on the two identification results by using a DS evidence reasoning algorithm, and the information fusion judgment result is used as a final identification result;
The specific steps of the positioning terminal in the step 3 for positioning and identifying the stock tracks are as follows:
3-1), after the positioning terminal is successfully matched with the iron shoe, placing the positioning terminal beside the iron shoe, taking the RFID (radio frequency identification) time of the terminal machine and the iron shoe as a reference time, respectively taking i groups of high-precision positioning data from front to back to form a positioning data sequence with the length of (2 x i+1), and executing the step 3-2);
3-2), carrying out stock track identification by using an accuracy weight statistical method: respectively carrying out track identification on the (2 x i+1) group positioning data, distributing the accuracy weights of the track identification data to the identified tracks, finally counting the total weights of the tracks, taking the track with the highest total weight as a track identification result, and executing the step 3-3) after the track identification result is obtained;
3-3), carrying out stock track identification by using a variational Gaussian filtering method: carrying out variational Gaussian filtering calculation on the positioning data of the (2 x i+1) group to obtain variational Gaussian filtered coordinates (x ', y'), calculating the vertical distance between the variational Gaussian filtered coordinates and the center line of each track, wherein the track with the nearest distance is the track recognition result, and executing the step 3-4 after the track recognition result is obtained;
3-4), judging whether the stock way identification results of the two methods are consistent or not: if the results are consistent, uploading the stock channel identification results, and entering step 4; if the results are inconsistent, executing the step 3-5);
3-5), i=i+5, enlarging the time range of positioning data acquisition, and judging whether i is less than or equal to 15; if i is less than or equal to 15, executing the step 3-1); if i >15, step 3-6) is performed.
3-6), Adopting a recognition result fusion judgment method based on different recognition methods, respectively calculating the credibility of the two methods on different recognition results, utilizing a DS evidence reasoning algorithm to fusion-calculate and judge the stock track recognition result, taking the fusion judgment result as a final result and uploading, and entering the step 4;
step 4: monitoring update status of host computer
After the stock way of the iron shoe is successfully identified, the monitoring upper computer detects whether the stock way has a train or not, if the monitoring upper computer displays that the stock way has a train, the iron shoe is normally on line, and the monitoring upper computer updates the state of the train and the iron shoe; if the stock is displayed to be without a car, whether the iron shoes are placed in place or not is confirmed again, and whether the upper computer is required to carry out the supplementary recording of the car number information or replace the iron shoes is judged.
The specific steps for monitoring the update state of the upper computer in the step 4 are as follows:
4-1), monitoring whether the upper computer has a car or not according to the intelligent skate track recognition result: if the upper computer displays that the stock road has no vehicles, executing the step 4-2); if the upper computer displays that the track has a car, executing the step 4-4);
4-2), detecting whether the skate is put in place: the placement of the skate in place includes two conditions: 1. determining that the shoe is on track by a metal detector on the shoe; 2. the distance between the skate and the wheels is detected by an acoustic ranging probe on the skate, and the skate is determined to be placed in the anti-slip range. When the iron shoes are on the track and in the anti-slip range and the iron shoes are put in place, executing the step 4-3); if any condition is not met, the iron shoes are not placed in place, and the iron shoes need to be replaced, and the step 2-1) is executed;
4-3), putting the iron shoe in place, and executing the step 4-4 after the upper computer does not timely record the train number information and the train number information is complemented;
4-4), the upper computer is monitored to update the state of the train and the intelligent iron shoes on the track, and the process is finished.
As shown in fig. 3, in step 3-2), the method for identifying the tracks by using the precision weight statistics method is to allocate precision weights to the (2 x i+1) group positioning data, and respectively identify the tracks, count the precision weight statistics value of each track, and use the track with the highest precision weight statistics value as the track identification result, and the specific steps for identifying the tracks by using the precision weight statistics method are as follows:
3-2-1), taking card swiping time as a reference, respectively taking i groups of positioning data from front and back to obtain a positioning data sequence with the length of (2 x i+1), and distributing weights to the (2 x i+1) groups of positioning data according to the precision, wherein the weight formula is as follows:
Where δ j is the positioning data precision of the jth positioning data, ω j is the precision weight of the jth positioning data. Let j=1, g 1=G2=···=Gk=0,G1、G2···Gk be the identification weight statistics of k tracks, respectively, and execute step 3-2-2);
3-2-2), taking the j-th set of positioning data (x j,yj) in the positioning data sequence, and calculating the vertical distance L 1···Lk between the positioning point (x j,yj) and the center line of each track. Executing the step 3-2-3);
3-2-3), determining a minimum value L m, m epsilon [1, k ] in L 1···Lk, namely that the positioning point is nearest to the track m, and the precision weight of the positioning data is delta j, counting the precision weight on the corresponding track weight, enabling G m=Gm+δj, j=j+1, and executing the step 3-2-4);
3-2-4), judging whether j is less than or equal to 2 x i+1: if j is less than or equal to 2 x i+1, executing the step 3-2-2); if j >2 x i+1, executing step 3-2-5);
3-2-5), determining that the maximum value in G 1、G2···Gk is G N, N epsilon [1, k ], and ending when the stock track identification result of the precision weight statistics method is the stock track N.
As shown in fig. 4, the method for identifying the tracks by using the variational gaussian filtering method in step 3-3) is to perform variational gaussian filtering calculation on (2 x i+1) group positioning data to obtain coordinates (x ', y') after variational gaussian filtering, calculate the vertical distance between the variational gaussian filtering coordinates and the center line of each track, and the track closest to the coordinates is the track identification result, and the specific steps for identifying the tracks by using the variational gaussian filtering method are as follows:
3-3-1), taking the card swiping time as a reference, respectively taking i groups of positioning data before and after to obtain a positioning data sequence P 1(x1,y1)、P2(x2,y2)、···、P2i+1 (x, y) with the length of (2 x i+1), giving a Gaussian weight value to each group of data in the sequence, and changing coordinate data after variance Gaussian filtering into (x ', y'), wherein the calculation mode of x 'and y' is as follows:
Wherein x j、yj and sigma correspond to x coordinate, y coordinate and Gaussian standard deviation of the j-th set of positioning data respectively. For positioning data sequences with different lengths, selecting different sigma values, changing weight distribution, when i is gradually increased and the length of the positioning data sequence is enlarged, reducing the sigma value to improve the weight of positioning data near the reference moment, reducing the weight of positioning data at the edge moment, and when i=5, taking sigma=10; when i=10, σ=4.5 is taken; when i=15, σ=4 is taken.
Carrying out variational Gaussian filtering on all positioning data in the sequence to obtain coordinates (x 'and y'), and executing the step 3-3-2);
3-3-2), calculating the vertical distance L 1···Lk between the coordinate point (x ', y') and the center line of each track, and then executing the step 3-3-3);
3-3-3), determining the minimum value in L 1···Lk as L M, M epsilon [1, k ], and the track recognition result as track M,
And (5) ending.
Step 3-6) a recognition result fusion judging method based on two stock track recognition methods, wherein the specific fusion judging process is as follows:
and (5) recording an identification result of the precision weight statistical method as p, and recording an identification result of the variance-changing Gaussian filtering method as q. p, q epsilon [1, k ] and p not equal q, namely the identification results obtained by the two identification methods are different.
The credibility of the accuracy weight statistical method for the strand identification result p and q is respectively as follows:
wherein, And G P,Gq is the sum of the precision weight statistics values of the p tracks and the q tracks respectively.
The credibility of the variational Gaussian filtering method for the strand identification result p and q is respectively as follows:
based on the recognition results of the two stock way recognition methods, the reliability m (p) of the matched stock way p and the reliability m (q) of the matched stock way q are calculated in a fusion mode, wherein:
Comparing m (p) with m (q), and if m (p) is greater than m (q), determining that the track p is matched by fusion; if m (p) < m (q), the fusion judgment is that the stock way q is matched; if m (p) =m (q), the likelihood of the hint etc. matches the track p, the track q.
As shown in fig. 5, the specific algorithm for calculating the vertical distance between the positioning coordinate point and the center line of each track is as follows:
(1) According to the rail geographical mapping data of the train station, determining the geographical position of the track center line, and determining the coordinates of two end points of the track center line as A (X 1,Y1),B(X2,Y2);
(2) The positioning point coordinates P (X, Y), the vertical distance between the positioning point P and the track center line is as follows:
The specific embodiments described herein are offered by way of example only to illustrate the spirit of the invention. Those skilled in the art may make various modifications or additions to the described embodiments or substitutions thereof without departing from the spirit of the invention or exceeding the scope of the invention as defined in the accompanying claims.
Claims (5)
1. An intelligent skate track recognition method based on information fusion is based on an intelligent skate track recognition system based on information fusion, and the system comprises more than 4 positioning satellites, a reference station, a 4G private base station, a plurality of pairs of intelligent shoes, a positioning terminal and a monitoring upper computer;
The satellite refers to a satellite which establishes communication with the reference station and the positioning terminal in the Beidou positioning system, establishes communication with the reference station and the positioning terminal, and sends positioning signals to the reference station and the positioning terminal;
the reference station receives a positioning signal of a positioning satellite, so as to calculate self positioning information, calculate with self mapping position coordinates to obtain differential data, and send the differential data to the positioning terminal through the 4G private base station for positioning correction;
The positioning terminal is placed beside the intelligent iron shoe and is used in a binding mode with the intelligent iron shoe, the positioning terminal receives positioning signals of positioning satellites to determine self positioning information, and high-precision positioning information is provided for the intelligent iron shoe based on a differential data correction method; the positioning terminal fuses the recognition results of the two stock track recognition methods, namely the precision weight statistical method and the variance-variable Gaussian filtering method, comprehensively judges and obtains the final stock track recognition result so as to realize accurate recognition of the stock track; the positioning terminal sends the positioning result to the monitoring upper computer through the 4G private base station;
The 4G private base station receives the reference station differential data and forwards the reference station differential data to the positioning terminal, and simultaneously receives the positioning data corrected by the positioning terminal; on the other hand, the 4G private base station interacts with the monitoring upper computer data through a wired Ethernet interface, and the received positioning data after the positioning terminal is corrected is sent to the upper computer;
The intelligent skate track identification method is characterized by comprising the following steps of: the method comprises four steps of system initialization, terminal and iron shoe identification matching, station track positioning and identification by a terminal machine and upper computer update state monitoring, and comprises the following specific processes:
step 1: system initialization
The system initialization comprises the initialization of a Beidou positioning module in a positioning terminal and the initialization of stock track information of a train yard, wherein the initialization comprises the geographical information of a stock track center line and the number k of stock track of the yard, which are obtained according to geographical mapping information of a yard track, and the number i=5 of positioning data sets are acquired before and after a reference moment in the first recognition;
step 2: terminal and skate identification matching
The railway station staff performs anti-slip operation, performs intelligent skate placement operation, performs RFID radio frequency identification with the intelligent skate by using a positioning terminal, and performs matching with the skate;
step 3: locating terminal locating and identifying stock way
After the positioning terminal is successfully matched with the intelligent iron shoe, taking the RFID radio frequency identification matching time as a reference time, respectively taking i seconds of high-precision positioning data before and after to form a data sequence with the length of 2 x i+1, and sequentially carrying out precision weight statistical method strand identification and variance-variable Gaussian filtering method strand identification by the identification system by utilizing the group of positioning data, wherein the strand identification result is determined and output when the two identification results are consistent; if the identification results are inconsistent, the number of the positioning data acquisition groups i=i+5, and the identification results of the two methods are inconsistent after the step 3 is circulated for 3 times, fusion judgment is carried out on the two identification results by using a DS evidence reasoning algorithm, and the information fusion judgment result is used as a final identification result;
The specific steps of the positioning terminal in the step 3 for positioning and identifying the stock tracks are as follows:
3-1), after the positioning terminal is successfully matched with the iron shoe, placing the positioning terminal beside the iron shoe, taking the RFID (radio frequency identification) time of the terminal machine and the iron shoe as a reference time, respectively taking i groups of high-precision positioning data from front to back to form a positioning data sequence with the length of 2 x i+1, and executing the step 3-2);
3-2), carrying out stock track identification by using an accuracy weight statistical method: respectively carrying out track recognition on the 2x i+1 group positioning data, distributing the precision weights of the track recognition data to the recognized tracks, finally counting the precision weights of all tracks, taking the track with the highest precision weight statistic value as a track recognition result, and executing the step 3-3) after the track recognition result is obtained;
The specific steps for identifying the stock way by using the precision weight statistical method are as follows:
3-2-1), taking card swiping time as a reference, respectively taking i groups of positioning data before and after to obtain a positioning data sequence with the length of 2 x i+1, and accurately distributing weights to the 2 x i+1 groups of positioning data, wherein the weight distribution is specifically as follows:
δ j e (0,0.05 ], ω j =10;
δ j e (0.05,0.2 ], ω j =8;
δ j e (0.2, 0.5], ω j =4;
δ j e (0.5, 1], ω j =2;
δ j e (1, 1.5], ω j =1;
delta j epsilon (1.5, +.infinity), omega j =0;
wherein delta j is the positioning data precision of the jth positioning data, and the unit is meter; omega j is the precision weight of the jth positioning data; let j=1, g 1=G2=···=Gk=0,G1、G2···Gk be the accuracy weight statistics of k tracks, respectively, and execute step 3-2-2);
3-2-2), taking the j-th group of positioning data (x j,yj) in the positioning data sequence, calculating the vertical distance L 1···Lk between the positioning point (x j,yj) and the center line of each track, and executing the step 3-2-3);
3-2-3), determining a minimum value L m, m epsilon [1, k ] in L 1···Lk, namely that the distance between the locating point and the stock track m is the minimum, and the accuracy weight of the locating data is omega j, counting the accuracy weight on the corresponding stock track weight, enabling G m=Gm+ωj, j=j+1, and executing the step 3-2-4);
3-2-4), judging whether j is less than or equal to 2 x i+1: if j is less than or equal to 2 x i+1, executing the step 3-2-2); if j >2 x i+1, executing step 3-2-5);
3-2-5), determining that the maximum value in G 1、G2···Gk is G N, N is E [1, k ], and ending when the stock track identification result of the precision weight statistics method is the stock track N;
3-3), carrying out stock track identification by using a variational Gaussian filtering method: carrying out variational Gaussian filtering calculation on the 2x i+1 group of positioning data to obtain variational Gaussian filtered coordinates (x ', y'), calculating the vertical distance between the variational Gaussian filtered coordinates and the center line of each track, wherein the track with the nearest distance is the track recognition result, and executing the step 3-4 after the track recognition result is obtained;
The specific steps of the stock way identification by using the variational Gaussian filtering method are as follows:
3-3-1), taking the card swiping time as a reference, respectively taking i groups of positioning data before and after to obtain a positioning data sequence P 1(x1,y1)、P2(x2,y2)、···、P2i+1 (x, y) with the length of (2 x i+1), giving a Gaussian weight value to each group of data in the sequence, and changing coordinate data after variance Gaussian filtering into (x ', y'), wherein the calculation mode of x 'and y' is as follows:
Wherein x j、yj and sigma correspond to the x coordinate, y coordinate and Gaussian distribution standard deviation of the j-th group of positioning data respectively; for positioning data sequences with different lengths, selecting different sigma values, changing weight distribution, when i is gradually increased and the length of the positioning data sequence is enlarged, reducing the sigma value to improve the weight of positioning data near the reference moment, reducing the weight of positioning data at the edge moment, and when i=5, taking sigma=10; when i=10, σ=4.5 is taken; when i=15, σ=4 is taken;
carrying out variational Gaussian filtering on all positioning data in the sequence to obtain coordinates (x 'and y'), and executing the step 3-3-2);
3-3-2), calculating the vertical distance L 1···Lk between the coordinate point (x ', y') and the center line of each track, and then executing the step 3-3-3);
3-3-3), determining the minimum value in L 1···Lk as L M, M epsilon [1, k ], and the track recognition result as track M,
Ending;
3-4), judging whether the stock way identification results of the two methods are consistent or not: if the results are consistent, uploading the stock channel identification results, and entering step 4; if the results are inconsistent, executing the step 3-5);
3-5), i=i+5, enlarging the time range of positioning data acquisition, and judging whether i is less than or equal to 15; if i is less than or equal to 15, executing the step 3-1); if i >15, executing the step 3-6);
3-6), adopting a recognition result fusion judgment method based on different recognition methods, respectively calculating the credibility of the two methods on different recognition results, utilizing a DS evidence reasoning algorithm to fusion-calculate and judge the stock track recognition result, taking the fusion judgment result as a final result and uploading, and entering the step 4;
step 4: monitoring update status of host computer
After the stock way of the iron shoe is successfully identified, the monitoring upper computer detects whether the stock way has a train or not, if the monitoring upper computer displays that the stock way has a train, the iron shoe is normally on line, and the monitoring upper computer updates the state of the train and the iron shoe; if the stock road is displayed without the train, whether the iron shoes are placed in place or not is confirmed again, and whether the train number information is required to be recorded by an upper computer or whether the iron shoes are replaced or not is judged.
2. The method according to claim 1, wherein the specific steps of matching and identifying the positioning terminal and the iron shoe in the step 2 are as follows:
2-1), firstly, taking out the intelligent iron shoes and the positioning terminal from the anti-slip device box by on-site operators, transferring the intelligent iron shoes to an anti-slip operation place, and placing the intelligent iron shoes on a preset rail;
2-2), matching and binding the positioning terminal with the intelligent iron shoe through RFID;
2-3), judging whether the positioning terminal is successfully matched with the iron shoe or not: if the matching is unsuccessful, executing the step 2-2); if the matching is successful, the step 3 is entered.
3. The method of claim 1, wherein the algorithm for calculating the vertical distance between the locating point and the center line of each track is as follows:
(1) According to the rail geographical mapping data of the train station, determining the geographical position of the track center line, and determining the coordinates of two end points of the track center line as A (X 1,Y1),B(X2,Y2);
(2) The positioning point coordinates P (X, Y), the vertical distance between the positioning point P and the track center line is as follows:
4. The method of claim 1, wherein the method of step 3-6) is characterized in that the method of determining the fusion of the recognition results based on different track recognition methods performs the following fusion calculation to determine the track recognition result when the accuracy weight statistics method and the variance gaussian filtering method are respectively used for track recognition, and the track recognition result is inconsistent, and the specific fusion determination process is as follows:
recording the recognition result of the precision weight statistical method as p and the recognition result of the variance-changing Gaussian filtering method as q; p, q epsilon [1, k ] and p not equal q, namely the identification results obtained by the two identification methods are different;
The credibility of the accuracy weight statistical method for the strand identification result p and q is respectively as follows:
wherein, G p,Gq is the sum of the precision weight statistics values of the p tracks and the q tracks respectively;
The credibility of the variational Gaussian filtering method for the strand identification result p and q is respectively as follows:
based on the recognition results of the two stock way recognition methods, the reliability m (p) of the matched stock way p and the reliability m (q) of the matched stock way q are calculated in a fusion mode, wherein:
Comparing m (p) with m (q), and if m (p) is greater than m (q), determining that the track p is matched by fusion; if m (p) < m (q), the fusion judgment is that the stock way q is matched; if m (p) =m (q), the likelihood of the hint etc. matches the track p, the track q.
5. The method of claim 1, wherein the specific steps of monitoring the update status of the host computer in step 4 are as follows:
4-1), monitoring whether the upper computer has a car or not according to the intelligent skate track recognition result: if the upper computer displays that the stock road has no vehicles, executing the step 4-2); if the upper computer displays that the track has a car, executing the step 4-4);
4-2), detecting whether the skate is put in place: the placement of the skate in place includes two conditions: 1. determining that the shoe is on track by a metal detector on the shoe; 2. detecting the distance between the skate and the wheel by using an acoustic ranging probe on the skate, and determining that the skate is placed in an anti-slip range; when the iron shoes are on the track and in the anti-slip range and the iron shoes are put in place, executing the step 4-3); if any condition is not met, the iron shoes are not placed in place, and the iron shoes need to be replaced, and the step 2 is executed;
4-3), putting the iron shoe in place, and executing the step 4-4 after the upper computer does not timely record the train number information and the train number information is complemented;
4-4), the upper computer is monitored to update the state of the train and the intelligent iron shoes on the track, and the process is finished.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110821256.9A CN113581244B (en) | 2021-07-20 | 2021-07-20 | Intelligent skate track recognition system and method based on information fusion |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110821256.9A CN113581244B (en) | 2021-07-20 | 2021-07-20 | Intelligent skate track recognition system and method based on information fusion |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113581244A CN113581244A (en) | 2021-11-02 |
CN113581244B true CN113581244B (en) | 2024-04-26 |
Family
ID=78248540
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110821256.9A Active CN113581244B (en) | 2021-07-20 | 2021-07-20 | Intelligent skate track recognition system and method based on information fusion |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113581244B (en) |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104765814A (en) * | 2015-04-03 | 2015-07-08 | 中铁第四勘察设计院集团有限公司 | Multi-time-scale coexisting type management method for railway yard plan station tracks |
CN105398474A (en) * | 2015-12-16 | 2016-03-16 | 郑州北斗七星通讯科技有限公司 | Beidou technology based intelligent real-time monitoring visualization management system for railway anti-slip iron brake shoe |
CN205220706U (en) * | 2015-12-16 | 2016-05-11 | 郑州北斗七星通讯科技有限公司 | Visual management system of swift current skate real time monitoring is prevented to big dipper intelligence railway |
CN108001477A (en) * | 2018-01-03 | 2018-05-08 | 伊军庆 | A kind of skate remote monitoring and management system and its method |
CN109552365A (en) * | 2019-01-15 | 2019-04-02 | 兰州运通铁路科技有限责任公司 | A kind of anti-slip remote monitoring system of rolling stock |
CN110406565A (en) * | 2019-08-01 | 2019-11-05 | 南京富岛信息工程有限公司 | The positioning and rectifying of Intelligent iron shoe and stolen determination method and device |
CN111003026A (en) * | 2019-12-30 | 2020-04-14 | 中国铁道科学研究院集团有限公司通信信号研究所 | In-station vehicle protection method based on intelligent iron shoes and STP |
CN111055877A (en) * | 2019-12-24 | 2020-04-24 | 南京富岛信息工程有限公司 | Wide-temperature-range intelligent iron shoe and anti-slip state judgment method thereof |
CN210573918U (en) * | 2019-10-30 | 2020-05-19 | 中国神华能源股份有限公司神朔铁路分公司 | Railway anti-running management system |
WO2021063136A1 (en) * | 2019-09-30 | 2021-04-08 | 江苏大学 | Data-driven high-precision integrated navigation data fusion method |
CN112793616A (en) * | 2021-02-08 | 2021-05-14 | 南京富岛信息工程有限公司 | Novel intelligent skate and in-place distance acquisition method thereof |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20200405223A1 (en) * | 2015-07-17 | 2020-12-31 | Chao-Lun Mai | Method, apparatus, and system for automatic and adaptive wireless monitoring and tracking |
US10229363B2 (en) * | 2015-10-19 | 2019-03-12 | Ford Global Technologies, Llc | Probabilistic inference using weighted-integrals-and-sums-by-hashing for object tracking |
-
2021
- 2021-07-20 CN CN202110821256.9A patent/CN113581244B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104765814A (en) * | 2015-04-03 | 2015-07-08 | 中铁第四勘察设计院集团有限公司 | Multi-time-scale coexisting type management method for railway yard plan station tracks |
CN105398474A (en) * | 2015-12-16 | 2016-03-16 | 郑州北斗七星通讯科技有限公司 | Beidou technology based intelligent real-time monitoring visualization management system for railway anti-slip iron brake shoe |
CN205220706U (en) * | 2015-12-16 | 2016-05-11 | 郑州北斗七星通讯科技有限公司 | Visual management system of swift current skate real time monitoring is prevented to big dipper intelligence railway |
CN108001477A (en) * | 2018-01-03 | 2018-05-08 | 伊军庆 | A kind of skate remote monitoring and management system and its method |
CN109552365A (en) * | 2019-01-15 | 2019-04-02 | 兰州运通铁路科技有限责任公司 | A kind of anti-slip remote monitoring system of rolling stock |
CN110406565A (en) * | 2019-08-01 | 2019-11-05 | 南京富岛信息工程有限公司 | The positioning and rectifying of Intelligent iron shoe and stolen determination method and device |
WO2021063136A1 (en) * | 2019-09-30 | 2021-04-08 | 江苏大学 | Data-driven high-precision integrated navigation data fusion method |
CN210573918U (en) * | 2019-10-30 | 2020-05-19 | 中国神华能源股份有限公司神朔铁路分公司 | Railway anti-running management system |
CN111055877A (en) * | 2019-12-24 | 2020-04-24 | 南京富岛信息工程有限公司 | Wide-temperature-range intelligent iron shoe and anti-slip state judgment method thereof |
CN111003026A (en) * | 2019-12-30 | 2020-04-14 | 中国铁道科学研究院集团有限公司通信信号研究所 | In-station vehicle protection method based on intelligent iron shoes and STP |
CN112793616A (en) * | 2021-02-08 | 2021-05-14 | 南京富岛信息工程有限公司 | Novel intelligent skate and in-place distance acquisition method thereof |
Non-Patent Citations (3)
Title |
---|
叶彦斐.智能铁鞋定位纠偏及被盗判定的设计与实现.工业控制计算机.2020,第54-55页卷第54-55页. * |
智能铁鞋定位纠偏及被盗判定的设计与实现;叶彦斐;黄家辉;胡文杰;徐涛;陈天石;;工业控制计算机;20200125(01);第57-58页 * |
车辆防溜铁鞋智能管理***的研究和设计;刘鑫鹏;中国优秀硕士学位论文全文数据库;20190115;第11-12、39-47 * |
Also Published As
Publication number | Publication date |
---|---|
CN113581244A (en) | 2021-11-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109215347B (en) | Traffic data quality control method based on crowdsourcing trajectory data | |
CN111862659B (en) | GPS track data matching and complementing method | |
CN105741546B (en) | The intelligent vehicle Target Tracking System and method that roadside device is merged with vehicle sensor | |
CN107976697B (en) | Train safety positioning method and system based on Beidou/GPS combination | |
US9533626B2 (en) | Method and system for determining the availability of a lane for a guided vehicle | |
CN108454652A (en) | A kind of method, apparatus and system of safe and reliable real time speed measuring and consecutive tracking | |
CN110406565B (en) | Method and device for positioning, correcting and judging theft of intelligent iron shoe | |
CN105241465B (en) | A kind of method of road renewal | |
JP6856979B2 (en) | Radio interference detection system and radio interference detection method along the route | |
CN103010265B (en) | Be applicable to the static train locating method of CBTC system | |
CN114446048B (en) | Rail transit traveler full travel chain analysis method based on mobile phone signaling data | |
CN110493714B (en) | Bluetooth auxiliary positioning vehicle returning method and system | |
CN101866017B (en) | Magnetic positioning method of traffic vehicle based on displacement cyclic unique code | |
CN110736999A (en) | Railway turnout detection method based on laser radar | |
CN103050014A (en) | Traffic speed detection system and detection method | |
CN113581244B (en) | Intelligent skate track recognition system and method based on information fusion | |
CN114872763A (en) | Method for determining position of train relative to transponder based on satellite positioning | |
US8260480B2 (en) | Automatic creation, maintenance and monitoring of a guideway database | |
CN106981236A (en) | A kind of curved traveling exam detection means and method | |
CN113079516B (en) | Method, device and equipment for determining base station and computer storage medium | |
CN113324560A (en) | Method, system and computer readable medium for obtaining vehicle mileage | |
CN104986188A (en) | Rail train positioning system and method | |
US20100191675A1 (en) | Wireless positional based route tolling | |
CN113353124B (en) | Intelligent iron shoe track identification system and method | |
CN112445879A (en) | Track data processing method and device, medium, terminal and server |
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 |