JP3421139B2 - Railroad point detection method and device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a railway spot detecting method and apparatus for detecting a traveling spot of a running railway vehicle.

[0002]

2. Description of the Related Art In order to pass a curved portion of a route at high speed and comfortably, a pendulum vehicle with a control has been put into practical use which performs a tilt control for tilting the vehicle leftward or rightward by a predetermined angle according to its curvature. This tilt control needs to be started prior to the traveling of the curved portion, and it is necessary to hold accurate curve information representing the bending shape of the traveling line section and the traveling position of the vehicle. It is necessary to detect (so-called point detection).

The curve information can be known from a design drawing etc. at the time of laying the track, but it is necessary to devise the point detection. At first glance, it may be possible to detect points by comparing the above curve information with the running distance of the vehicle, but in practice it is possible to measure the running distance accurately because the wheels slide or slip on the rails. If this is not possible, a deviation will occur between the two, and if a point is detected from the traveled distance, a considerable error will be included.

Therefore, in the past, AT was used as a point detection method.
A method of detecting the position of this ground element using S ground element and matching it with the curve information of the traveling line section has been performed.

[0005]

However, according to the above-mentioned prior art, it is necessary to obtain position information indicating the position of the ATS ground element in the line section, which requires a lot of expense and cost. Need time. Further, the ATS ground element is not for point detection, and it may be necessary to relocate because of its original purpose of use, and it was necessary to collect the above-mentioned position information each time it was relocated.

The present invention has been made in view of the above problems, and a method for detecting a point of a vehicle running without using an ATS ground element and a position detection of a vehicle running without using an ATS ground element. The purpose is to provide a device for performing.

[0007]

A method for detecting a point for a railway vehicle according to claim 1 of the present invention, which has been made to achieve the above object, is a method of rotating a wheel of a vehicle prior to normal traveling of the vehicle. Under the speed condition where the amount and the running distance are substantially the same, the vehicle is run in the running line section during normal running, the behavior of the vehicle in the left-right direction at this time is measured, and this measured value is measured for each predetermined rotation amount of the wheel. The reference data representing the behavior of the vehicle with respect to the travel distance in the travel line zone is created by storing the data in the travel zone, and during normal traveling, the behavior of the vehicle in the left-right direction is measured in the travel zone and externally measured. In accordance with the point search command, the search data is created by storing a predetermined number of wheels for each predetermined rotation amount, and the cross-correlation function between the search data and the reference data is calculated to obtain the search data. Corresponding vehicle behavior pattern There obtains an area that best matches the behavior pattern of the vehicle represented by the reference data, which was based on the subject matter to determine the current location of the vehicle.

As shown in FIG. 1, a railroad vehicle spot detecting device according to a second aspect of the present invention measures rotation amount measuring means for measuring a rotation amount of a wheel of a vehicle and lateral behavior of the vehicle. By storing the behavior measurement means and the value measured by the behavior measurement means in the first storage means for each predetermined rotation amount measured by the rotation amount measurement means in accordance with an external reference data creation command, Reference data creating means for creating reference data representing behavior of the vehicle with respect to the travel distance within the travel line section, and when receiving a point search command from the outside, the value measured by the behavior measuring means is used as the rotation amount measuring means. When a predetermined number of measured rotation amounts are stored in the second storage means by a predetermined number, the search data creation means creates search data, and the search data creation means creates the search data. By calculating a cross-correlation function between the reference data and the search data, a zone in which the behavior pattern of the vehicle corresponding to the search data best matches the behavior pattern of the vehicle represented by the reference data is obtained, and based on this, And a point specifying means for specifying the present location of the vehicle.

According to a third aspect of the present invention, there is provided a railroad vehicle spot detecting device, wherein the traveling line section is divided into a plurality of small sections in advance, and a point search point is provided in each of the small sections. When the measurement result by the rotation amount measuring means reaches each rotation amount of the wheels corresponding to the respective point search points during normal traveling, the point search command is changed to the search data creating means. And a rotation amount correcting means for correcting the measured value of the rotation amount of the wheel by the rotation amount measuring means on the basis of the specifying result by the point specifying means. It is characterized in that a cross-correlation function between each search data created in each small section and the reference data in each small section is calculated.

[0010]

The point detecting method for railway vehicles according to the first aspect of the present invention is performed as follows. First, prior to normal running, the vehicle is run in the running line section during normal running, the lateral behavior of the vehicle at this time is measured, and the measured value is stored for each predetermined rotation amount of the wheel and data is stored. Create a column. This run is performed under speed conditions that prevent the wheels from idling or gliding, for example, under conditions such as low speed and no sudden acceleration or deceleration. It represents the behavior of the vehicle with respect to the distance. This data string is called reference data.

Even during normal traveling, the behavior of the vehicle in the left-right direction is measured within the traveling line section. When the point search command is received, a predetermined number of the measured values are stored for each predetermined rotation amount of the wheel to create search data. Since this running is performed at a higher speed than the running when the reference data is created, the wheels generally run on the rails or slip. That is, the behavior pattern indicated by the search data generally does not match the behavior pattern indicated by the reference data at the same rotation amount.

When the cross-correlation function between the search data thus created and the reference data is calculated, the behavior pattern of the vehicle represented by the search data is calculated in an area where the behavior pattern of the vehicle represented by the reference data is the best match. The function value becomes maximum. Based on this, the present location of the vehicle is specified. As a supplementary example of this specific method, the deviation between the rotation amount of the wheel at the starting point of the above area and the rotation amount of the wheel when the point search command is received is obtained, and the degree of sliding / idling of the wheel is calculated from this. Estimate and correct the current amount of wheel rotation.

In other words, the search data created during normal travel is searched from the reference data created during travel that preceded this, so the travel point of the vehicle can be determined without using the ATS ground element. It can be detected from inside. This search is performed using the cross-correlation function, and when the reference data is created, the wheels are run under the speed conditions that do not cause idling or sliding on the rails, and the wheel rotation amount and the vehicle travel distance match. Therefore, even if the wheels slip or spin during normal driving, the running position of the vehicle at the time when the search data is created can be detected. The traveling position of can also be obtained.

The point detection by the railway vehicle point detection device according to the second aspect is performed as follows. That is,
When the reference data creation means receives a reference data creation command during the travel that precedes the normal travel, the behavior measured by the behavior measurement means is changed every time the rotation quantity measured by the rotation quantity measurement means increases by a certain amount. It is stored in the first storage means and is used as reference data. This traveling is performed under a speed condition such that the wheel and the rail do not slip or slide.

In normal driving, within the above running line section,
When the search data creation means receives the search data creation command, a predetermined number of behaviors measured by the behavior measurement means are stored in the second storage means each time the rotation amount measured by the rotation amount measurement means increases by a certain amount. Store and use this as search data. Since the normal running is performed at a higher speed than the running when the reference data is created, it is expected that the wheels are sliding or idling on the rails. Therefore, the behavior pattern indicated by the search data generally does not match the behavior pattern indicated by the reference data for the same amount of rotation of the wheel.

Subsequently, the point specifying means calculates the cross-correlation function between the reference data and the search data, and from the calculation result,
An area in which the behavior pattern of the vehicle corresponding to the search data best matches the behavior pattern of the vehicle represented by the reference data is obtained, and the current present location of the vehicle is specified based on this area.

In other words, the search data created during normal travel is searched from the reference data created during travel that preceded this, so the target point can be searched for without using the ATS ground element. It can be detected from inside. Moreover, since this search is performed by calculating the cross-correlation function between the reference data and the search data, it is possible to specify the running position of the vehicle even if the wheels slip or slip on the rail during normal running. it can.

The point detection by the point detection device for a railway vehicle according to claim 3 is performed as follows. That is,
Similarly to the invention described in claim 2, the vehicle travels within the traveling line zone to create the reference data prior to the normal traveling. This traveling line section is divided into several small sections, and the search command means later sets a point search point which is an index for determining the timing of issuing the search data creation command to the search data creation means.

During normal traveling, when the rotation amount of the wheels reaches the rotation amount of the wheels corresponding to the point search point, the search command means issues a search data creation command to the search data creation means, and the search data is created. Will start to be created. Creation of the search data is completed when a predetermined number of behaviors are stored in the second storage means.

Subsequently, the point specifying means calculates a cross-correlation function between the search data and the reference data in this small section,
From this calculation result, an area in which the behavior pattern of the vehicle corresponding to the search data most matches the behavior pattern of the vehicle represented by the reference data is obtained. From this result, the rotation amount correction means obtains an error between the rotation amount of the wheel and the actual traveling distance, and corrects the rotation amount of the wheel based on the error.

That is, according to the railway spot detecting device of the third aspect, in addition to the effect of the invention of the second aspect, the running line section is divided into small sections, and the search range is within each small section. By this, the amount of calculation of the cross-correlation function can be reduced and detection can be performed in a short time. Further, by dividing the reference data so as not to include a part showing a confusing behavior pattern in one small section, it is possible to prevent an erroneous search and improve the detection accuracy. Further, since the rotation amount correction means corrects the measured value of the rotation amount of the wheel based on the detection result every time the point detection is performed, no error is accumulated and the detection accuracy can be maintained.

[0022]

Embodiments of the present invention will be described below with reference to the drawings. First, FIG. 2 shows the overall configuration of a tilt control device 1 as an embodiment to which the present invention is applied.

As shown in FIG. 2, the tilt control device 1 includes a lateral acceleration sensor 11 for detecting the lateral acceleration of the vehicle,
A signal processing circuit 13 for waveform-shaping the acceleration signal output from the lateral acceleration sensor 11, and further A / D converting and outputting the waveform;
A storage device 15 for storing a signal output from the signal processing circuit 13 and a point search point described later, a speed generator 17 for outputting a pulse signal according to a rotation amount of wheels, and a speed generator 1
7. A mileage calculation circuit 19 capable of counting 7 pulse signals and receiving the change of the pulse count value from the outside, and an output of the signal processing circuit 13 every time the pulse count value of the mileage calculation circuit 19 increases by a fixed number. A reference data creation unit 21 as a reference data creation unit of the present invention that stores the acceleration thus obtained in the storage device 15, and a signal processing circuit each time the pulse count value of the travel distance calculation circuit 19 increases by a certain number when a search command described later is received. A search data creation unit 23 that stores a predetermined number of accelerations output by 13 in the storage device 15.
Then, when the pulse count value of the mileage calculation circuit 19 reaches a point search point stored in advance in the storage device 15, a search command unit 25 that outputs a search command to the search data creation unit 23.
A cross-correlation function calculation unit 27 that calculates a cross-correlation function between the reference data and the search data, and a travel distance correction unit 29 that corrects the pulse count value of the travel distance calculation circuit 19 based on the calculation result. A curve information storage device 31 and a curve information storage device 31 in which the correspondence relationship between the curve information including the curvature and the cant of the travel line section and the travel distance is stored in advance.
The optimum lean amount of the vehicle is calculated from the curve information stored in the pulse count value of the running distance calculation circuit 19 and the running speed obtained from the output of the speed generator 17, and a control input corresponding to this lean amount is generated. The tilt controller 33 and a tilt actuator 35 for tilting the vehicle to the left or right according to the control input generated by the tilt controller 33.

It should be noted that, in the above-mentioned configuration, the reference data creation unit 21, the search data creation unit 23, the search command unit 25,
The cross-correlation function calculation unit 27 and the traveling distance correction unit 29 are realized by the electronic control unit 3 including a well-known CPU, RAM, ROM, input / output interface, and a bus connecting them, and the storage device 15 is an external storage device. , The storage medium, and a part of the RAM, and correspond to the first and second storage means of the present invention. Further, the lateral acceleration sensor 11 and the signal processing circuit 13 correspond to the behavior measuring means of the present invention, and the speed generator 17 and the traveling distance calculating circuit correspond to the rotation amount measuring means of the present invention.

A vehicle having the inclination control device 1 travels at a low speed in the line section in which the vehicle normally travels prior to being operated in the line section. The reason why it is performed at a low speed is to prevent the wheel and the rail from slipping or sliding. This makes it possible to reduce the error between the vehicle travel distance and the wheel rotation amount. The flow of processing performed at this time is shown by a solid arrow in FIG.

That is, the running line section is defined as the entire section between the starting station and the ending station of the route on which the vehicle runs, and the lateral acceleration detected when the vehicle is run at low speed is converted into an electric signal. Is stored in the storage device 15 as the first storage means every time the vehicle travels for a certain distance. Thus, the storage device 1
The data created in 5 is called reference data, and the process of creating this reference data is called reference data creation processing.

The reference data creating process is started at the time of departure from the starting station and stopped at the ending station. The flow is shown in the flowchart of FIG.
The reference data creation unit 21 performs the entire process shown in FIG. First, at step 100, the pulse count value of the traveling distance calculation circuit 19, which is a count of the pulse signals output from the speed generator 17, is read. This directly corresponds to the amount of rotation of the wheels, but this run is
Since the wheels are run at a low speed so that they do not slip or spin on the rails, it can be called the mileage. Hereinafter, to simplify the explanation, it is assumed that the speed generator 17 outputs one pulse when the vehicle travels 1 m, and the traveling distance is reset at the starting station.

Next, the process proceeds to step 110 and step 10
Based on the value read at 0, it is checked whether the vehicle has traveled a sampling distance. In this embodiment, the sampling distance is 5 pulses, that is, 5 m. If the mileage has not reached 5 m, the process returns to step 100, and steps 100 to 11 are repeated until 5 m is reached.
The loop of 0 is repeated. If it has reached 5 m, the process proceeds to the following step 120.

At step 120, the lateral acceleration detected by the lateral acceleration sensor 11 is read through the signal processing circuit 13 and stored at step 130 in the storage device 15. Finally, in step 140, it is determined whether or not the terminal has arrived at the terminal station by checking whether or not the traveling distance read in step 100 has reached the total distance of the traveling line section. If not, otherwise return to step 100 and repeat the above steps.

When the reference data creation process is completed in this manner, the storage device 15 stores the lateral acceleration of the vehicle sampled every time the vehicle travels 5 m (5 pulses).
For example, the mileage from the starting station to the terminal station is 100km
Then, the reference data consisting of a total of 20,000 lateral accelerations is created. The reference data created in this manner shows a waveform unique to the radius of curvature of the track or rail, the amount of cant, the vehicle that measures the reference data, and the like. FIG. 4A shows a waveform of this reference data in which the horizontal axis represents the number of pulses and the vertical axis represents the lateral acceleration. The number of pulses on the horizontal axis is limited to 2000.

Next, the vehicle is returned to the starting station, and normal traveling, that is, high-speed traveling on which people and cargo are loaded is performed.
In this traveling, in order to drive the vehicle safely, comfortably, and at high speed on a curved portion of the track, tilt control is performed to incline the vehicle to the left or right. The flow of the process is shown by a dashed arrow in FIG.

Originally, the main processing in the tilt control is processing for determining the amount of tilt of the vehicle and tilting the vehicle accordingly. Here, the point detection and the running point based on this are the basis for the timing of starting the tilt. The explanation is centered on the identification of. The outline of this process is that the entire traveling section from the starting station to the terminal station is divided into several small sections, the traveling point of the vehicle is detected in each small section, and the traveling distance is corrected accordingly. Based on this, the tilt control is performed.

Therefore, it is necessary to provide a point search point in each small section before performing this processing. This is the number of pulses corresponding to the trigger point for starting the creation of the search data created to detect the running point. In the present embodiment, as shown in FIG. 100 km (100,000 pulse number) is A1, A2, A
It is equally divided into 50 subsections such as 3, ..., and the midpoint of each subsection is used as the point search point. That is, if the point search points are N1, N2, N3, ... From the station near the starting station, the number of these pulses is 1,000, 3000, 5000 ,. This value is stored in the storage device 15 in advance. The number of pulses N (2000 here) in this small section is
The waveform of the reference data, that is, the waveform of FIG. 4A is investigated over the entire traveling line section, and it is determined that a confusing waveform that may be erroneously searched does not enter the same small section.

In addition to this point search point, the pulse numbers C1 and C corresponding to the point where the tilt control is required, that is, the point where the preceding control of the tilt before the curved portion of the track is started.
2, ... Are also stored in the storage device 15. This point is called the control start point. The difference between the above point search point and the control start point is that the same number of point search points as the number of small sections are set, but the control start point is determined by the shape of the running line section, and it depends on how to divide the running line section. Do not rely on it, and the above point search point is for specifying the timing to start the acquisition of search data, and the number of pulses cannot correspond to the actual running point due to wheel idling / sliding , The control start point corresponds to the number of pulses corresponding to the actual traveling point.

The flow of this tilt control process is shown in the flowchart of FIG. First, step 200 to step 220
The inclination controller 33 performs the process consisting of Step 20
At 0, the pulse count value of the traveling distance calculation circuit 19 which counts the pulse signal output from the speed generator 17 is read. It is assumed that this value has been reset at the starting station as in the reference data creation process.

Next, in step 210, step 200
The value read by is the number of pulses C1 and C at the control start point
2, C3, ... Is checked, and if not reached, the routine proceeds to step 230, and if it is reached, the tilt control is performed at step 220 and then the routine proceeds to step 230. The inclination control is performed by the curvature information stored in the curve information storage device 31, which indicates the bending state of the track, and the travel distance calculation circuit 19
The tilt controller 33 calculates the optimum amount of tilt of the vehicle from the pulse count value of the vehicle speed and the traveling speed obtained from the output of the speed generator 17, and the control input corresponding to this amount of tilt is applied to the tilt actuator 35. However, we will omit detailed explanation here because we will focus on point detection.

At the following step 230, the search command unit 25 performs the processing as the command means of the present invention. Step two
In step 30, it is checked whether the pulse count value read in step 200 has reached the point search point. That is, it is checked whether or not the number of pulses N1, N2, N3, ... Corresponding to the point search points stored in advance in the storage device 15 has been reached. Here, the number of pulses read in step 200 is 1000, and the point search point N1
It is assumed that the value of matches. Then, the process proceeds to step 240, and the search data creating unit 23 performs the search data creating process as the search data creating means of the present invention. If not reached, the process returns to step 200 and the above steps are repeated.

The search data creation process performed in step 240 is as shown in the flowchart of FIG. As can be seen from this, the search data creation process is exactly the same as the reference data creation process except for a part after the process is started. That is, steps 300 to 330 are exactly the same as steps 100 to 130 of the reference data creation process except that the lateral acceleration is stored in the storage device 15 as “search data” in step 330. Therefore, the explanation is omitted.

In the following step 340, step 3
It is largely different in that the storage of the lateral acceleration performed at 30 is a condition for ending the processing whether or not a predetermined number m of preset values are stored. That is, the search data created in the search data creation process consists of m lateral acceleration signals. Here, m = 120, and the search data created from the point search point N1 is shown in FIG. M in this figure is the number of pulses corresponding to m = 120, and here, since 5 lateral pulses store 120 lateral acceleration signals, M = 600.

The description returns to the tilt control process. When the search data creation process is completed as described above, step 250
The processing as the point specifying means of the present invention, which consists of steps 260 to 260, is performed. In Step 250, Step 24
The cross-correlation function calculation unit 27 calculates the cross-correlation function between the search data created by 0 and the reference data created by the reference data creation process. The cross-correlation function used here is as in the following equation [Equation 1].

[0041]

[Equation 1]

N is the number of lateral acceleration signals included in one small section, m is the number of lateral acceleration signals included in one search data, data string x (i) is reference data, and data string y (i).
Is search data, x (1) represents the first lateral acceleration signal of the reference data in the small section, and y (1) represents the first lateral acceleration signal of the search data. Here, the value of m is 120 as described above, and the value of n is 400 because the lateral acceleration signal is sampled every 5 pulses in a small section consisting of 2000 pulses to create reference data. FIG. 4C shows a graph in which the vertical axis represents the cross-correlation function Rxy (i) and the horizontal axis represents the pulse number.

Generally, the cross-correlation function is used for the purpose of retrieving the area that most matches the signal sequence to be searched (search data in this embodiment) from the signal sequence in the search range (reference data in this embodiment). However, the value of the cross-correlation function Rxy (i) is maximized in that area. In the present embodiment, if the wheels rotate without slipping on the rail during normal traveling, the pulse number Nmax that maximizes the cross-correlation function Rxy (i) and the point search point Ni match. Generally, the wheels do not match because the wheels slide and spin. In FIG. 4C, Nmax = 870, which is 130 pulses less than N1 = 1000. That is, in this normal traveling, the wheel idles for 130 pulses with respect to the rail (however,
Ignore the offset of sliding and slipping).

It should be noted that here, in order to clearly show the difference between the waveforms shown in FIGS. 4 (a) and 4 (b) (in particular, the deviation of the search data from the reference data), a deviation of as much as 130 pulses is presented. There is no such slip / sliding, and the deviation is only 1 to 2%.

Next, in step 260, the pulse amount Nmax and the point search point Ni that maximize the cross-correlation function Rxy (i).
The travel distance correction unit 29 corrects the travel distance output from the travel distance calculation circuit 19 based on the deviation. In the above example, since the number of pulses was counted from 0 to 1000, an extra 130 pulses were counted compared to the actual traveling distance, so the output of the traveling distance calculating circuit 19 read by the traveling distance correcting unit 29. If the pulse count value to be set is 1800, the value obtained by subtracting 130 pulses from this is 167
The pulse count value is corrected to 0.

As described above, in step 260, in addition to the processing as the point specifying means of the present invention for specifying the current position of the vehicle, the processing as the rotation amount correction means of the present invention for correcting the rotation amount of the wheel is also performed. That is, the pulse number 167
The process up to the part for calculating 0 corresponds to the point specifying means, and the process for correcting the pulse count value of the running distance calculating circuit 19 to 1670 corresponds to the process as the rotation amount correcting device.

When the correction of the traveling distance is completed, step 27
When it is determined that the vehicle has finished traveling in all the traveling line zones, the inclination control process is terminated. More specifically, the number of pulses matched with the traveled distance as corrected in step 260 is compared with the number of pulses corresponding to the distances of all traveling line sections (here, 100,000) to determine the end of traveling. To do. If it is determined that the vehicle is still running, step 200
Return to and repeat the above steps.

FIG. 8 schematically shows how this point detection and the correction based on it are performed. In this figure, the horizontal axis represents time, and the vertical axis represents the pulse count value of the travel distance calculation circuit 19. The broken line graph replaces the actual mileage with the pulse count value, the solid line graph the pulse count value when the above point detection processing is performed, and the two-dot chain line graph the assumption when the relevant point detection is not performed. This is the pulse count value. If the wheels do not slip or slide on the rail, these three graphs will overlap to form one curve.

When the vehicle starts to run normally, search data is started to be created at the time N1 when the pulse count value of the running distance calculation circuit 19 reaches 1000, and when this is completed, the cross-correlation function is calculated and the number of pulses is corrected. Do (Point P1
). Thereafter, this is repeated, and the number of pulses is corrected at P2, P3, and P4, and a graph like a solid line is obtained. 2 when this point detection and correction are not performed
A curve closer to the broken line graph, which is the ideal number of pulses, is obtained as compared with the broken line graph.

That is, in the tilt control process, the point detection necessary for specifying the time point at which the tilt control is performed is performed as described above, and the tilt control is performed at the time when the traveling point of the vehicle reaches the control start point. To do. As for point detection, the entire running line section is divided into a plurality of small sections as described above, and the rotation amount of the wheel corresponding to the intermediate point of each small section (in the above example, the number of pulses is 1000, 3000, 5000, ...). This is carried out by storing 120 lateral acceleration signals for every 5 pulses from the time point of reaching, and searching from the reference data in the small section created in the same manner in advance.

Therefore, as in the conventional point detection method, A
Since the TS grounding element is not used, even if the ATS grounding element is relocated, the labor for measuring the position of the ATS grounding element is not required. Further, since this search is performed by calculating the cross-correlation function of the reference data and the search data, the point can be detected even when the wheels spin / slide on the rail during normal running.

Further, in the tilt control process, the search is performed in each of the divided small sections, and the number of pulses is corrected from the error obtained as a result, so that the error due to idling / sliding is accumulated. Therefore, it is possible to detect points with high accuracy. Also, since the search range is reduced by dividing the reference data into small sections, the amount of calculation of the cross-correlation function is reduced, correction can be performed quickly, and the confusing lateral acceleration signal waveform has the same small section. Since it is divided so that it does not enter inside, the possibility of false search can be further reduced.

Although the tilt control device 1 to which the spot detecting device of the present invention is applied has been described as an embodiment of the present invention, the present invention is not limited to the embodiment and is implemented in various modes. You can. For example, in the present embodiment, the lateral acceleration signal is used to represent the behavior of the vehicle in the left-right direction, and the reference data or the search data is created by storing the lateral acceleration signal. In addition to this, the amount of tilt of the carriage in the left-right direction, the amount of self-steering of the carriage, the radius of curvature, etc. may be used, and similar point detection can be performed. Further, the reference data may be created from the average value of the behaviors recorded by running the vehicle a plurality of times instead of defining the reference data by one time of the reference data creation processing as described above.

In the inclination control device 1 of the present embodiment, the entire traveling line section is equally divided, but this division may be made unequal in length. In this case, the length of the small section is long in the section in which the tilt control is not necessary and is short in the vicinity of the control start point, whereby the necessary and sufficient tilt control can be performed. Further, in the above embodiment, the length of the search data is m = 120, which is common to all the small sections, but the number may be different for each small section. By doing so, when the reference data has a waveform that may cause an erroneous search, it is possible to prevent the erroneous search by increasing m. In particular, when each small section is made unequal in length as described above, it is preferable to set m to a different value according to the length of the small section. As a special case, m = 0 in a small section that does not require point detection, or a small section in which straight lines are long and difficult to detect, that is, point detection is not performed in that small section. good. By doing so, extra calculation can be omitted, and if the reference data is not created in this section in accordance with this, the capacity of the storage device 15 can be saved.

Further, as a result of calculating the cross-correlation function, when a clear maximum value cannot be calculated, or when the maximum value can be calculated but is significantly smaller than the normal maximum value, the detection thereof is detected. The result may be discarded and the correction may not be performed. Even in this case, since the detection is performed in the following small section, the point detection is possible.

Further, in the above embodiment, the point search point is set corresponding to the center point of each small section, but it may be shifted to the starting station side or the ending station side. For example, empirically, in a section where slipping tends to occur more often than sliding, setting a point search point shifted from the center of a small section toward the terminal station allows more slipping.

The cross-correlation function is not limited to that shown in the equation [Equation 1]. For example, there is a position where a significantly large lateral acceleration signal is detected in a certain small section, and if the above cross-correlation function Rxy (i) is used, Rxy (i) becomes maximum at that position. Moreover, when it is not possible to increase m because the location is close to the control start point, etc., the reference data and search data in each small section are dimensionless with the average of the absolute value of each signal value. It is better to use the cross-correlation function after conversion. That is, the cross-correlation function is calculated by the following expression [expression 2] instead of expression [expression 1].

[0058]

[Equation 2]

Since the maximum value of R'xy (i) is smaller than the maximum value in the cross-correlation function of search data that is rather large, the maximum value of the cross-correlation function is valid when it is equal to or greater than a certain threshold. The threshold should be adjusted by conducting an experiment. Also, in the formula [Formula 1] and the formula [Formula 2] as the search range of the correlation function, the subscript i was moved from 0 to nm, and the last m of the small sections were excluded, but the terminal station was included. It is possible to calculate up to i = n except for small sections, that is, small sections in which the reference data does not exist after i = n. On the contrary, the starting point of the search range may be set to i = -m except for a small section including the starting station, that is, a small section in which the reference data does not exist before i = 0. Further, each search range corresponding to each point search point does not need to be independent from each other as described above, and a part of adjacent search ranges may overlap.

In the above embodiment, the point search point is set by the number of rotations of the wheels, but it may be set by another one. For example, the illuminance sensor indicates that the vehicle entered the first tunnel in the small section. Creation of search data may be started by detecting the ATS ground element or the like. The ATS ground element is intended to be disused in the present invention, but when used as a trigger for a point search point,
Since it is not necessary to know its position accurately, unless it is relocated outside the small section, there is no need to measure the position every time it is relocated as in the conventional case.

The correction by the traveling distance correction unit 29 is not limited to the method shown in the above embodiment. For example, the deviation occurring in the section from the point search point to the moment when the mileage correction unit 29 makes the correction may be predicted and corrected. As a method of prediction, it is assumed that deviation occurs at the same rate as the section to the point search point, prediction from the magnitude of the maximum value of the cross-correlation function and the waveform of the search data, slip / sliding in the past It may be estimated in consideration of the tendency of.

[0062]

According to the method for detecting a point for a railway vehicle according to claim 1 of the present invention, the behavior of the vehicle in the left-right direction stored during traveling prior to normal traveling is used as reference data. Since the behavior signal measured during normal traveling is used as search data to search for this from the reference data, the location of the vehicle is detected without using the ATS ground element. You can Also, since both the reference data and the search data are stored per unit distance, regardless of the running speed of the train,
Both can be compared. Therefore, when the reference signal is generated, accurate point detection can be performed if the vehicle runs at a speed condition such that the wheel does not slip or slide on the rail, for example, at a low speed. Further, even if the wheels spin / slide on the rail during normal traveling, the retrieval data is retrieved by calculating the cross-correlation function with the reference data, so the traveling position of the vehicle can be detected. it can.

According to the railway vehicle spot detecting device of the second aspect, the behavior measuring means measures the behavior of the train in the left-right direction, and the reference data creating means stores in the traveling line section stored in the first storage means. The search data creation means similarly searches the search data stored in the second storage means from the reference data representing the behavior of the vehicle with respect to the travel distance. Therefore, AT
You can search the running point of the vehicle at the time when you started to create the search data from the running line section that stores the reference data, without using the S ground element. Based on this, you can find the current vehicle position. Can be detected. Further, since the reference data and the search data are stored for each constant rotation amount measured by the rotation amount measuring means, the reference data and the search data can be correlated regardless of the traveling speed of the vehicle. Therefore, when creating the reference signal, the speed conditions that prevent the wheel from idling or sliding on the rail,
For example, if the vehicle runs at a low speed, it is possible to detect a point with high accuracy. Further, the cross-correlation function calculating means performs the search of the search data by calculating the cross-correlation function with the reference data, so that the running position of the vehicle can be detected even if the wheels slide on the track or slip. can do.

According to the railway vehicle spot detection device of the third aspect, the following effect can be obtained in addition to the effect of the invention of the second aspect. That is, by dividing the running line section into small sections and setting the cross-correlation function calculating means as the search range within each small section, the calculation amount of the cross-correlation function can be reduced and the search can be performed in a short time. Further, when the reference data has a waveform having a plurality of similar behavior patterns, the behavior patterns are divided so that the behavior patterns belong to different small sections, thereby reducing the possibility of erroneous search and increasing the detection accuracy. be able to. Further, the search command means generates a command for starting the storage of search data when the actual rotation amount reaches the rotation amount of each wheel corresponding to the point search point in each small section.
There is no need for an operator or the like to issue a command in anticipation that the vehicle has reached the point search point in each small section. Especially, if this point search point is set at the center of each subsection,
It can be allowed for both sliding and slipping wheels,
In addition, in a small section in which a lot of gliding tends to occur, if the point search points are biased rearward with respect to the traveling direction, more gliding can be allowed.

Moreover, since the rotation amount correcting means corrects the measured value of the rotation amount of the wheel by the rotation amount measuring means on the basis of the output of the spot output means, the spot detection can be performed even if the running line segment is divided into small sections. It is possible to maintain the accuracy of point detection without accumulating the error of.

[Brief description of drawings]

FIG. 1 is a block diagram illustrating a configuration of the present invention.

FIG. 2 is a block diagram showing an overall configuration of a tilt control device 1 according to an embodiment.

FIG. 3 is a flowchart showing a reference data creation process executed by a reference data creation unit 21.

FIG. 4 shows the reference data created in the small section A1, the search data created in the small section A1 and the cross-correlation function of the two in correspondence with each pulse count value, that is, the rotation speed of the wheel. It is the explanatory view shown.

FIG. 5 is an explanatory diagram showing a state of division of a traveling line section of a vehicle.

FIG. 6 is a flowchart showing a tilt control process performed during normal traveling of the vehicle.

FIG. 7 is a flowchart showing a search data creation process executed by the search data creation unit 23.

FIG. 8 shows the relationship between the time and the pulse count value when the number of pulses is corrected according to the point detection processing of the present embodiment.
It is explanatory drawing shown in a simulation.

[Explanation of symbols]

1 ... Inclination control device 3 ... Electronic control device 11 ... Lateral acceleration sensor 13 ... Signal processing circuit 15 ... Storage device 17 ...
Speed generator 19 ... Running distance calculation circuit 21 ... Standard data creation unit 23 ... Search data creation unit 25 ... Search command unit 27 ... Cross-correlation function calculation unit 29 ... Travel distance correction unit

─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Kuromozawa Chito, 1-1 Sanbonmatsucho, Atsuta-ku, Nagoya, Aichi Prefecture, Japan Vehicle Manufacturing Co., Ltd. (56) Reference JP-A-6-107172 (JP, A) ( 58) Fields investigated (Int.Cl. ⁷ , DB name) G01C 22/02

Claims (3)

(57) [Claims] 1. Prior to the normal running of the vehicle, the vehicle is
Under the speed condition where the rotation amount of the wheel and the mileage substantially match,
Traveling within the traveling line zone by running in the traveling line zone during normal traveling, measuring the behavior of the vehicle in the left-right direction at this time, and storing this measured value for each predetermined rotation amount of the wheel Create reference data that represents the behavior of the vehicle with respect to distance, measure the behavior of the vehicle in the left-right direction within the above travel line section during normal driving, and according to the point search command from the outside, for each predetermined amount of rotation of the wheel. By storing a predetermined number of search data, and calculating the cross-correlation function of the search data and the reference data, the behavior pattern of the vehicle corresponding to the search data of the vehicle represented by the reference data A method for detecting a point for a railway vehicle, which is characterized in that an area most matching a behavior pattern is obtained, and the present location of the vehicle is specified based on the area. 2. A rotation amount measuring means for measuring a rotation amount of a wheel of a vehicle, a behavior measuring means for measuring a behavior of the vehicle in a left-right direction, and a value measured by the behavior measuring means as a reference from the outside. According to the data creation command, the reference data representing the behavior of the vehicle with respect to the travel distance in the travel line section is created by storing in the first storage means for each predetermined rotation quantity measured by the rotation quantity measuring means. When the reference data creating means and the point search command from the outside are received, a value measured by the behavior measuring means is stored in the second storage means for each predetermined rotation amount measured by the rotation amount measuring means. The search data creating means for creating the search data by storing the minutes and the search data created by the search data creating means calculates a cross-correlation function between the reference data and the search data. By obtaining a zone in which the behavior pattern of the vehicle corresponding to the search data most matches the behavior pattern of the vehicle represented by the reference data, and based on this, a point specifying means for specifying the present location of the vehicle. A point detection device for a railway vehicle, comprising: 3. The railway vehicle spot detection device according to claim 2, wherein the traveling line section is divided into a plurality of small sections in advance, and
Point search points are set in each of the small sections, and when the vehicle travels normally and the rotation amount of the wheel corresponding to each point search point reaches the measurement result by the rotation amount measuring means, the point search point is set. Commanding means for issuing a search command to the search data creating means, and rotation amount correcting means for correcting the measured value of the rotation amount of the wheel by the rotation amount measuring means based on the identification result by the point identifying means are further provided. A railway vehicle spot detecting device, wherein the spot identifying means calculates a cross-correlation function between each search data created in each small segment and reference data in each small segment. .