JP2018118595A - Obstacle detection device in railway crossing - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an obstacle detection device in a railway crossing whose software processing load is mitigated.SOLUTION: In a determination program 20 of a failsafe computer 14 that processes measured data on a railway crossing road 4 acquired by flat surface sweeping measurement parts 11 to 13, after data inputting 21, the measured data are contracted for subsequent processing in data limiting 22 before grouping 23. Eventually, the data processing load is decreased in size and volume and an increase of tracking number is suppressed without sacrificing capability of tracking. Tracking 24 is carried out after the grouping and then in the final determination 25, the existence of obstacles is determined in response to the existence of pursued objects in the tracking. This avoids increasing of burden of determination of existence and worsening of correctness even if tracking process which uses representing points and imposes light processing burden is adopted.SELECTED DRAWING: Figure 1

Description

  TECHNICAL FIELD The present invention relates to a level crossing obstacle detection device that detects a level crossing obstacle (obstacle) such as a person or a vehicle on a railroad crossing by irradiating an aerial propagation wave such as infrared rays in a railroad crossing. In particular, the present invention relates to a plane sweeping radar type crossing obstacle detection device that receives a reflected wave and performs position measurement while performing a plane sweep, and more specifically, determines whether an obstacle exists based on measurement data obtained by such a method. The present invention relates to a level crossing obstacle detection apparatus that is performed by a safe computer.

  A plane sweeping radar type crossing obstacle detection device (see, for example, Patent Document 1) is an obstacle in a part of a crossing road that crosses a railroad, located between the fences on both sides of the railroad, and a surface area above it. As a detection area, based on the measurement data obtained by the plane sweep measurement unit that receives the reflected wave and measures the distance and direction while sweeping and transmitting the air propagation wave toward the obstacle detection area And a determination unit for determining the presence or absence of an obstacle in the obstacle detection area.

  Among them, the plane sweep measurement unit (see, for example, Patent Documents 1 and 2) includes an aerial propagation wave transmission / reception unit that transmits an aerial propagation wave and receives a reflected wave toward an obstacle detection region, and the aerial propagation wave transmission / reception unit. A rotation mechanism that normally rotates within a range of a predetermined angle, a rotation control unit that controls the rotation movement and performs or enables measurement of the direction of transmission / reception of airborne waves, and transmission / reception signals of the airborne wave transmission / reception unit And a signal processing unit for measuring the distance from the transmission position to the reflection position.

  The determination unit includes a computer as means for executing data processing for determining the presence / absence of an obstacle based on measurement data and software for determining the contents of the determination processing. And in software processing, after tracing the series of data and grasping the object shape, it is judged whether it is an obstacle by checking the object shape against the memory shape, etc. Since the computer responsible for execution needs to secure the reliability necessary for detecting a crossing obstacle, a fail-safe computer capable of revealing a hardware failure is employed (see, for example, Patent Documents 3 and 4).

In such a plane sweep radar type crossing obstacle detection device, the aerial wave transmission / reception unit is installed at the same height as the breaking fence during descent. Compared to the three-dimensional level crossing obstacle detection device that looks down, it is easier to reduce costs.
In addition, compared with the beam type that has been used from the past, the number of installed transmitters and receivers can be reduced, so the number of installations can be reduced, and the resolution is good, so large vehicles such as automobiles as well as individual passers and wheelchairs are small There is an advantage that even things can be detected.

  Furthermore, in data processing and discrimination processing, each time a group of plane sweeps is performed, the object shape is grasped by tracing a series of measurement point data (a pair of distance and direction) included in the measurement data. If the measurement data is grouped for each obstacle candidate image by tracking within the image, the grouping process is performed in the image, and if a part of the detected object is in the barrier between the railroad crossings It is determined that there is an obstacle in the level crossing (see, for example, Patent Document 1). The grouping process in the image related to the obstacle candidate image increases the original detection target such as an automobile or a passer-by while avoiding an undesired false detection by measurement data related to low-density disturbance such as snowfall or rain. Since detection is based on dense measurement data, it helps to improve detection accuracy.

Japanese Patent Laid-Open No. 2006-007818 JP 2006-216961 A JP 2006-338094 A Japanese Patent Application No. 2006-163679 (Application)

  In this way, the detection performance of the level crossing obstacle detection device has been enhanced by various improvements from hardware to software, but since further improvement is required for practicality, processing based on measurement data is required. It cannot be limited to the grouping process described above. Specifically, in the case of a railroad crossing, just because an object exists in the monitoring area (obstacle detection area), it does not immediately issue an alarm as an obstacle or stop the train. There is also a demand for a function of accurately detecting a staying material that reaches, and excluding, from a detection target, a high-speed moving body that quickly passes through a railroad crossing to prevent overdetection.

  As one means for meeting the demand, it is conceivable to introduce a tracking process over time between a plurality of images related to the obstacle candidate image. In this tracking process, measurement point data grouped as an obstacle candidate image is grouped into a plane sweep and a grouping on the premise that the grouping process in the image related to the obstacle candidate image is performed every time a group of plane sweeps are performed. In the series of multiple measurement data that is repeatedly processed, information over time, such as how it has changed, especially how it has moved, is extracted, so the above functions can be realized by software . In addition, by combining tracking processing with grouping processing, it can be expected to suppress false detection that could not be achieved only by grouping processing.

On the other hand, in the tracking process, it is indispensable to introduce an element when the tracking disappears and to make a plurality of tracking numbers.
Among them, the tracking disappearance element is a predetermined time during which the tracking information and the like are kept after the measurement point data group of the obstacle candidate to be tracked disappears. This is necessary because, for example, the tracking may be temporarily interrupted, for example, when another object passes in front of the object being tracked or the airborne wave reflected from the object being tracked is weakened. This is because the tracking can be resumed promptly after the temporary factor disappears. An appropriate time as the resumption waiting time is previously determined as a tracking disappearance element, and data is held in a state that can be referred to during the tracking process. And even after the disappearance of the measurement point data group, the tracking data (tracking information) is maintained without being deleted until the tracking disappears.

Among them, the tracking number is the number of measurement point data groups of obstacle candidates that are tracked over time simultaneously in parallel.
The reason why the multiple is important is that if the number of tracking is one, another obstacle candidate object enters the monitoring area immediately after the obstacle being tracked leaves the monitoring area (obstacle detection area). In such a case, the tracking of the approaching object is not started immediately after the obstacle has advanced, but since that time is delayed until the time when the tracking disappears, the detection and determination of the obstacle is delayed, which is inconvenient. It is.

In addition, it is ideal that the measurement point data group of obstacle candidates related to one object is combined into one, but when there are large measurement fluctuation factors such as the surface shape and reflectance of the object, it may be divided into a plurality. Therefore, even in such a case, it is beneficial to use multiple tracking numbers.
Furthermore, for the effects of undesired factors that could not be eliminated by the grouping process, for example, undesired effects due to factors such as snowfall and spatial disturbances such as interference from the oncoming radar, the effect of mitigating / suppressing the effect could be reduced by the tracking number. It is expected that it will be progressed by making more than one.

Then, in addition to the grouping process in the image related to the obstacle candidate image described above, the above-described tracking process over time between the plurality of images related to the obstacle candidate image is also performed by software. The processing hardware requires a large processing capacity.
However, as mentioned above, fail-safe computers are used for the hardware of the railroad crossing obstacle detection device, and fail-safe computers are used in many consumer devices because priority is given to ensuring high reliability. Compared to the general computer that is, cost performance is sacrificed, so it is powerless.
Therefore, reducing the load of software processing is an important issue.

Specifically, in the tracking processing over time between multiple images related to the obstacle candidate image, it is assumed that the number of tracking is not limited to a single number as described above, but if the number of tracking is large, the processing load also increases. Since the tendency to become heavy is strong, suppressing the increase in the number of tracking without sacrificing the tracking performance is a first technical problem.
In tracking processing, position estimation and position prediction using a Kalman filter or the like are frequently used. In such position estimation, the representative point is determined instead of using the data of the measurement point data group of the obstacle candidate as it is. An estimation calculation related to the point coordinates is performed.

For this reason, even a part of the detected object is not very compatible with the above-described crossing obstacle presence / absence determination method in which it is determined that an obstacle exists in the crossing if it enters the barrier between the crossing roads.
That is, in the tracking process after the grouping process subsequent to the measurement data acquisition, the position of the representative point of the measurement point data group of the obstacle candidate is calculated using the representative point. Since the process is repeated by returning to the data of a plurality of points related to the measurement point data group of the obstacle candidates obtained by the earlier grouping process, the processing load is heavy.

However, after the measurement data acquisition, grouping processing and tracking processing are performed, and then the position of the representative point simply enters the monitoring area (obstacle detection area) without returning to the multiple data measurement point data group. In the determination method of determining whether or not there is a processing load, the processing load is reduced, but even if the end of the measurement point data group of the obstacle candidate remains in the monitoring area (obstacle detection area), the representative point is the monitoring area If it is out of the (obstacle detection area), the measurement point data group of the obstacle candidate is not detected as an obstacle, and the accuracy of the determination may be sacrificed.
Therefore, it is a second technical problem to improve so that the presence / absence determination burden is not increased and the accuracy is not impaired even if the tracking processing using the representative points with a light processing burden is performed.

  The crossing obstacle detection device of the present invention (Solution 1) was created in order to solve such a problem, and the plane sweep measurement that obtains the measurement data of the reflection position by plane sweeping the air propagation wave. And a fail-safe computer equipped with determination means for holding detection area defining data for defining an obstacle detection area for a level crossing and determining whether or not an obstacle exists based on the measurement data and the detection area defining data The level crossing obstacle detection apparatus, wherein the determination means narrows down the data related to the position belonging to the obstacle detection area from among the measurement data based on the detection area definition data, as the subsequent processing data A grouping process for specifying a measurement point data group of candidate obstacles using the data for subsequent processing and the measurement point data group A tracking process for tracking the position of the surface point over the time when the tracking disappears is performed, and the presence or absence of an obstacle is determined according to the presence or absence of the tracking target of the tracking process. .

  Further, the level crossing obstacle detection device of the present invention (Solution means 2) is the level crossing obstacle detection device of the above solution means 1, wherein the tracking extinction time element in the tracking process has a speed related to the measurement point data group. It is characterized by increasing or decreasing depending on whether it is slow or fast.

  Further, the level crossing obstacle detection device according to the present invention (Solution means 3) is the level crossing obstacle detection device of the above solution means 1 and 2, and the speed related to the measurement point data group in the tracking process is faster than a predetermined speed. The tracking disappearance element is sometimes set to zero time or ignored.

  Further, the level crossing obstacle detection device of the present invention (solution means 4) is the level crossing obstacle detection device of the above solution means 3, wherein the predetermined speed is set to a value corresponding to the average walking speed of a healthy person. It is characterized by that.

  In such a level crossing obstacle detection device of the present invention (Solution means 1), two-dimensional measurement data relating to a passing body on a level crossing is obtained each time a plane sweep of an aerial propagation wave is performed. Instead of using the measurement data as it is and performing the grouping process or the like, the measurement data is narrowed down to the subsequent process data prior to the grouping process or the like, and then the grouping process or the like is performed using the subsequent process data. The data narrowing down reduces the distribution range of the measurement point data group of candidate obstacles, and accordingly, the data processing amount is reduced / reduced and the number of tracking is reduced or at least prevented from increasing. Since the data related to the position belonging to the object detection area is used for subsequent processing such as grouping without being removed, an increase in the number of tracking is suppressed without sacrificing the tracking performance.

  In addition, since the grouping process is performed using the subsequent processing data corresponding to the obstacle detection area, the measurement point data group of the obstacle that has advanced from the obstacle detection area disappears quickly, but the measurement point data group disappears. Since tracking processing is also performed to keep track of the corresponding tracking information until the predetermined tracking disappearance time passes, tracking of the measurement point data group of obstacle candidates over time is temporarily Therefore, the increase in the number of tracking is suppressed without sacrificing the tracking performance.

  Furthermore, since the presence or absence of an obstacle is determined according to the presence or absence of the tracking target of such tracking processing, it relates to a measurement point data group of obstacle candidates obtained by grouping processing prior to tracking processing. Without performing the heavy processing of returning to the data of multiple points and repeating the presence / absence determination, and without performing the attribute determination processing of the representative point position to the obstacle detection area that tends to sacrifice accuracy, The presence or absence of an obstacle can be determined easily and quickly. For this reason, in this level crossing obstacle detection device, even if tracking processing using representative points with a light processing burden is performed, the burden of existence determination does not increase. Accuracy is not compromised.

  Therefore, according to this invention, both the first and second technical problems described above can be solved. In addition, the following effects can be obtained. That is, since the image data to be processed has been narrowed down to those related to the obstacle detection area prior to the grouping process or the like, both when the obstacle enters the obstacle detection area and when it advances, Since the measurement point data group of the obstacle candidate is limited to the part belonging to the obstacle detection area, not the entire obstacle, the scale is expanded and contracted while sticking to the boundary line of the obstacle detection area. And when looking at the moving speed of each part of the image, the part farthest from the boundary moves at or near the speed of the obstacle, while the part along the boundary keeps almost stopping, so the obstacle candidate The inner point such as the center point in the measurement point data group moves at a slower speed than the obstacle.

  On the other hand, since position estimation and position prediction using the Kalman filter etc. calculate the next position from the previous position and speed, etc., it is generally vulnerable to discontinuous jumping speed changes, so at the start and end of tracking It has the property that it can be accurately tracked as the speed change is slow. Therefore, it is easy to adopt the inner point of the measurement point data group as the representative point of the obstacle point measurement point data group to be tracked by the tracking process using the representative point, especially the center position calculation and the center of gravity position among the inner points. By adopting the center point obtained by calculation or the like, the tracking ability in the tracking process can be improved easily and accurately.

  In the level crossing obstacle detection device according to the present invention (solution 2), when tracking a measurement point data group in the tracking process, if the speed related to the measurement point data group is slow, the tracking disappearance time is short. When the speed related to the measurement point data group is high, the time for tracking disappearance is shortened, so that it takes time to cross the railroad crossing because it moves slowly like an elderly person, for example. Can be tracked firmly to ensure safety, while for those that move quickly, such as an automobile, and do not have the risk of staying within the railroad crossing, tracking can be quickly rounded up to reduce the burden of data processing. .

Furthermore, in the level crossing obstacle detection device of the present invention (solution means 3 and 4), for a slow level crossing vehicle such as an elderly person, for the sake of safety, while performing tracking extension based on the tracking extinction element, an automobile By eliminating the extension of tracking based on the disappearance of tracking, the level crossing traffic safety and the data processing burden can be reduced to a high level.
Moreover, this can be done simply by setting the tracking extinction element to zero time or ignoring the tracking extinction element.

1 shows the structure of a level crossing obstacle detection device according to Example 1 of the present invention, where (a) is a schematic plan view of a level crossing road at an installation destination, and (b) is a schematic block diagram of hardware of a level crossing obstacle detection device. c) is a schematic block diagram of software of a crossing obstacle detection device. (A) is a schematic diagram of a plane sweep of an air propagation wave, (b) is an image diagram of measurement data, (c) is an image diagram of data for subsequent processing, and (d) is an image image diagram of a measurement point data group. is there. (A) is an image diagram of tracking information, (b) is an image image diagram of the measurement point data group, (c) is an image diagram of the tracking information, (d) is an image image diagram of the measurement point data group, and (e) is an image diagram of the tracking information. It is. (A)-(f) is an image figure of a measurement point data group and tracking information, (g) is a graph of a speed change example.

About the crossing obstacle detection apparatus of such this invention, the specific form for implementing this is demonstrated by the following Examples 1-3.
The embodiment 1 shown in FIGS. 1 to 4 embodies the above-described solving means 1 (claim 1 at the beginning of the application). 4 (claims 2 to 4 at the beginning of the application) is embodied.
In these drawings, for the sake of simplification, etc., housings, fasteners such as frames and bolts, drive sources such as electric motors, transmission members such as gears, electric circuits such as motor drivers, electronic circuits such as controllers, etc. Circuits and the like are not shown, and are shown in a block diagram and the like focusing on what is necessary for explaining the invention and related ones.

About the Example 1 of a level crossing obstacle detection device of the present invention, the concrete composition is explained referring to drawings.
1A is a schematic plan view showing an installation state of a level crossing obstacle detection device 10 on a level crossing road 4, FIG. 1B is a schematic block diagram showing a hardware configuration of the level crossing obstacle detection device 10, ) Is a schematic block diagram showing a software configuration of the crossing obstacle detection device 10. FIG. 2A is a schematic diagram of a plane sweep of an air propagation wave, FIG. 2B is an image diagram of measurement data, FIG. 2C is an image diagram of data for subsequent processing, and FIG. ), FIG. 3B and FIG. 3D are image images of the measurement point data group, and FIG. 3A, FIG. 3C and FIG. 3E are image diagrams of the tracking information.

  The level crossing obstacle detection device 10 (see FIGS. 1A and 1B, Patent Document 4) is a space region above the level crossing road 4 and sandwiched between barrier bars. Plane sweep measurement units 11 to 11 of a plane sweep radar system that receives a reflected wave and measures a distance and a direction while sweeping and transmitting an air propagation wave (see a two-dot chain line) such as infrared rays toward the obstacle detection region 7. 13 and a fail-safe computer 14 in which a determination program 20 for determining the presence or absence of an obstacle in the obstacle detection area 7 from the measurement data obtained by the measurement is installed.

  The plane sweep measuring units 11 to 13 transmit the aerial propagation wave transmission / reception unit 12 that transmits the aerial propagation wave and receives the reflected wave toward the obstacle detection region 7, and the transmission direction of the aerial propagation wave transmission / reception unit 12, for example, 130. The rotation control unit 11 that controls the rotational movement of the rotation mechanism that sweeps within an angular range of や or 190 ° and that measures or enables the direction of transmission / reception of the air propagation wave, and the transmission / reception signal of the air propagation wave transmission / reception unit 12 And a signal processing unit 13 for measuring the distance from the transmission position to the reflection position. In this example, two sets (plurality) of plane sweep measurement units 11 to 13 are provided in order to measure the entire area of the obstacle detection area 7 by area sharing.

  Since the fail-safe computer 14 suffices with a known product (see, for example, Patent Documents 3 and 4), it is employed, and the data area includes detection area defining data and tracking disappearance, both of which are preset constants. In addition, the measurement data that is a variable or an array that is changed as the determination program 20 is executed, the data for subsequent processing, the data of the measurement point data group, and the tracking information are held.

  In short, the determination program 20 (see FIG. 1C) determines whether or not an obstacle exists based on the measurement data and the detection area defining data. For this purpose, the data input program 21, the data limiting program 22, and the grouping are performed. A program 23, a tracking program 24, and a final determination program 25 are provided, which are repeatedly executed in that order, for example, at a predetermined cycle or whenever a predetermined event occurs.

More specifically, the detection area defining data defines the obstacle detection area 7 related to the railroad crossing 4, and is, for example, a two-dimensional coordinate of each corner position of the obstacle detection area 7 arranged in a circular order. The coordinates may be orthogonal coordinates, but the polar coordinates are more compatible with the planar sweep radar system.
As described above, the tracking extinction element is a predetermined time for which tracking information and the like are kept after the measurement point data group of the obstacle candidate to be tracked disappears. It is determined in consideration of the time from combining and separation of measurement point data groups due to traffic.

  Each time the plane sweep measurement units 11 to 13 perform measurement by plane sweep (see FIG. 2A), the data input program 21 uses the plane sweep measurement units 11 to 11 to collect polar coordinate values of the reflection positions obtained thereby. 13 is used as measurement data (see FIG. 1 (c) and FIG. 2 (b)). In this example, two sets of plane sweep measurement units 11 to 13 are simultaneously used and responded. Two sets of data obtained at the time are merged to form a set of measurement data (see FIGS. 2A and 2B).

This measurement data includes three vehicles of level crossings (FIGS. 2A and 2B) that are in the obstacle detection area 7 (see the alternate long and short dash lines in FIGS. 2A and 2B). 2) (see the black dots in FIG. 2B). However, not only that, but also the portion protruding from the obstacle detection area 7 or the black dots in FIG. Although not shown in the figure, the equipment and the like outside the obstacle detection area 7 includes the image data within the plane sweep range.
In this embodiment, the measurement data is a series of measurement point data consisting of a pair of distance and direction, and both the distance and the direction are fluctuation values obtained by measurement. For example, two rows and N columns (N is a positive integer) ) In the array region.

The data restriction program 22 executes data restriction processing and creates data for subsequent processing from the measurement data based on the detection area defining data (see FIG. 1C). At that time, the data restriction program 22 is included in the measurement data. Of the image data points (see the black dots in FIG. 2B), those located outside the obstacle detection area 7 (see the dashed line in FIG. 2B) are excluded. As a result, among the points of the measurement data, the data narrowed down to the data related to the position belonging to the obstacle detection region 7 is adopted as the subsequent processing data (see the black points in FIG. 2C).
In the case of an object straddling the boundary of the obstacle detection area 7, the subsequent processing data includes only image data of a portion in the area (refer to the left and right black dot groups in FIG. 2C).

  The grouping program 23 executes an in-image grouping process for creating an obstacle candidate measurement point data group from the subsequent processing data (see FIG. 1C), and each point of the image data of the subsequent processing data. (See the black dots in FIG. 2 (c).) The measurement point data is collected based on conditions that take into account the characteristics of crossing vehicles such as people and vehicles, by gathering short-distance objects one after another. Only the measurement point data group that can be an obstacle candidate image is identified by combining the groups (see three solid lines in FIG. 2D and two solid lines in FIG. 3B) ), The measurement point data group immediately disappears if there is no crossing vehicle that can be an obstacle candidate (see FIG. 3D).

The tracking program 24 executes a tracking process for tracking the measurement point data group obtained by the intra-group grouping process over time between a plurality of images (see FIG. 1C). If there is a plurality, the same number of tracking information is generated and held and assigned to each measurement point data group so that each measurement point data group is individually tracked.
Moreover, for each measurement point data group, for example, at least the center point of the measurement point data group is obtained by an appropriate calculation such as calculation of the center point or the center of gravity, and the inner point is adopted as the representative point of the measurement point data group. (See the black dots in FIGS. 3 (a) and 3 (c)).

  Further, the tracking program 24 performs a tracking process over time between a plurality of images related to the obstacle candidate image by estimating a next representative point position from a previous representative point position or the like by an estimation calculation such as a known Kalman filter. Specifically, for each tracking information, the next predicted position is calculated by, for example, a linear Kalman filter from the position and speed of the previous representative point, and the representative point is within a predetermined range from the predicted position. For the measurement point data group that is included, group identification information such as the representative point position is included in the corresponding tracking information.

  Here, the predetermined range is determined by a predicted radius set in advance, etc., but in the constant velocity linear motion model that is frequently used, if the predicted radius is increased, the maximum speed that can be tracked is also increased. On the other hand, there is a disadvantage that the separation performance of the tracking target deteriorates when there are multiple tracking targets, and even things that should not be tracked may enter inside the predicted radius and cause mistracking Therefore, the set value of the predicted radius is determined in advance in consideration of the balance between them.

  Furthermore, when performing such tracking processing, the tracking program 24 refers to the tracking extinction element that has been set in advance (see FIG. 1C) and tracks it for the time of the element. It has become. Specifically, when a measurement point data group to be tracked disappears by the grouping process, the tracking information related to the measurement point data group is not immediately deleted, but is continued until the tracking disappearance element expires. Yes. For example, even when three measurement point data groups to be tracked move to the right (see FIGS. 2D and 3A), the rightmost measurement point data group disappears (FIG. 3). (See (b)), the tracking information related to the representative point of the rightmost measurement point data group continues to exist before the tracking extinction time elapses (see FIG. 3C). It is to be done after elapse of.

  The final determination program 25 determines whether or not there is an obstacle in the obstacle detection area 7, and the determination process is performed according to the presence or absence of the tracking target of the tracking process (FIG. 1 (c)). reference). Specifically, if there is at least one piece of tracking information created by the tracking program 24 (see FIGS. 3A and 3C, three in the figure), it is determined that an obstacle exists in the obstacle detection area 7. If there is no such tracking information (see FIG. 3E), it is determined that there is no obstacle in the obstacle detection area 7.

  About the level crossing obstacle detection apparatus 10 of this Example 1, the use aspect and operation | movement are demonstrated referring drawings.

  FIG. 2A shows a state in which the airborne wave is swept planely toward the obstacle detection region 7 above the railroad crossing 4 by the airborne wave transmitting / receiving units 12 and 12, and FIG. ) Is an image diagram of measurement data, FIG. 2C is an image diagram of data for subsequent processing, and FIGS. 2D and 3A are image diagrams of an initial measurement point data group and tracking information. 3B and 3C are image diagrams of the subsequent measurement point data group and tracking information, and FIGS. 3D and 3E are image diagrams of the last measurement point data group and tracking information. . FIG. 4 is an image diagram in which (a) to (f) all show the measurement point data group and the tracking information superimposed, and (g) is a graph showing the speed change state when passing through the railroad crossing 4. Yes, the horizontal axis is time, and the vertical axis is speed.

In order to use the level crossing obstacle detection device 10 (see FIG. 2A), an appropriate number (two in the figure) of the aerial propagation wave transmitting / receiving units 12 and 12 are installed facing the level crossing road 4, and as much as possible. A range wider than the entire area of the obstacle detection area 7 is swept in a plane from a plurality of directions (two opposite directions in the figure).
Then, in the level crossing obstacle detection device 10, the plane sweep is repeatedly performed by the two sets of plane sweep measurement units 11 to 13 and 11 to 13, and each time there is a traveling body on the level crossing road 4 (three vehicles in the figure). Vehicle), the reflected wave is detected, the reflection position is acquired as a polar coordinate value, and measurement data relating to a region wider than the obstacle detection region 7 is obtained as a set of polar coordinate values (FIG. 2 ( In b), see the black spots on almost the whole of the three vehicles).

The determination program 20 is executed by the fail-safe computer 14 each time a set or group of measurement data obtained by such a plane sweep is obtained.
The determination procedure will be described in detail. First, the measurement data is taken into the fail-safe computer 14 by executing the data input program 21, and then the data for subsequent processing is created from the measurement data by executing the data limiting program 22. The image data to be processed is narrowed down to those belonging to the obstacle detection area 7 (in FIG. 2C, black spots relating to the rear end portion of the right vehicle, the entire center vehicle, and the front end portion of the left vehicle are determined. reference).

As described above, since the image data including the obstacle candidate is limited in the initial stage of the determination process, the load of the subsequent data processing is reduced.
Then, the in-image grouping process is performed by executing the grouping program 23, and the data of the measurement point data group of the obstacle candidate is created from the data for subsequent processing (in FIG. 2D, the rear end of the vehicle on the right side). (See the U-shaped solid line on the peripheral portion, the solid U-shaped line on the entire circumference of the central vehicle, and the U-shaped solid line on the front end peripheral portion of the left vehicle).

Furthermore, by executing the tracking program 24, representative points are selected for tracking over time for each measurement point data group, and one (a set) of tracking information used for tracking over time is created. (See three black dots in FIG. 3 (a)).
Tracking information is added each time a new measurement point data group is added to the tracking target, and each tracking information includes, for example, position data of representative points of the measurement point data group to be tracked and velocity data calculated from the change thereof. Included. When the tracking target disappears, the corresponding tracking information is deleted. Therefore, when there is no tracking target, the tracking information is completely lost (see FIG. 3 (e)).

Then, by the execution of the final determination program 25, it is determined that there is an obstacle on the railroad crossing 4 if there is even one (one set) of tracking information. The presence or absence of tracking information can be confirmed quickly and accurately by referring to any one of them, such as tracking information count data or the first tracking information presence / absence mark, in order to manage the tracking information. Can do.
Therefore, the determination of the absence of obstacles that could only be made after determining whether or not the measurement point data group to be tracked and its representative points belong to the obstacle detection region 7 for all tracking information, This level crossing obstacle detection device 10 is promptly issued with a light load. Then, this determination result is sent to a railroad crossing control device or the like and used for outputting an alarm to a special signal light emitter.

  In this way, every time the plane sweep measuring units 11 to 13 perform the plane sweep and the reflected wave measurement on the railroad crossing 4, the data limiting process, the grouping process, the tracking process, and the obstacle to the obstacle detection area 7 related to the measurement data are performed. The presence / absence determination processing is performed by the fail-safe computer 14 and further repeated at a predetermined cycle or the like. Among these processes, from the plane sweep to the grouping process, the measurement point data group of the processing result is uniquely determined by the traffic state and the image data from time to time. However, in the subsequent tracking process, the time series between multiple images is determined. At the time of tracking, since the position of the representative point of the measurement point data group is tracked over the time of the disappearance of tracking, the tracking information continues for a while after the measurement point data group disappears.

  For example, when tracking three measurement point data groups relating to three vehicles (see the black dots in the images on the right side of FIGS. 2D and 3A), the data of the measurement point data groups are also tracked. Three sets of information data are also retained (see the left block in Fig. 2 (d) and Fig. 3 (a)), but the three vehicles move from left to right and the right vehicle fails. When the object detection area 7 is exited, the measurement point data group immediately becomes two related to the central vehicle remaining in the obstacle detection area 7 and the left vehicle on which at least a part is applied (FIG. 3). (Refer to the solid line in the right image of (b)), the tracking information for the right vehicle is also maintained until the time when the tracking disappears (the black dot in the right image in FIG. 3C). And the left block).

Then, when the tracking extinction time elapses for the tracking, the tracking information also becomes two related to the two vehicles related to the obstacle detection area 7 (not shown).
Thus, it is carefully confirmed that the level crossing vehicle has left the obstacle detection area 7. In addition, although illustration is omitted, the time from merger to separation is tracked even when merger or separation occurs in multiple measurement point data groups related to multiple railroad crossings according to the traffic of the railroad crossings. If it is shorter than the extinction time, the temporal tracking between the plurality of images related to the obstacle candidate image is accurately performed without being disturbed by a large change in the image due to the merger.
After that, when all level crossing vehicles leave the obstacle detection area 7, the measurement point data group disappears (see FIG. 3D), and when the tracking extinction time passes, the tracking information also disappears (see FIG. 3). 3 (e)), it is found that no obstacle exists in the obstacle detection area 7.

  Furthermore (see FIG. 4), the use of the inner point of the measurement point data group as the representative point of the measurement point data group of the obstacle candidate tracked by the tracking process using the representative point, and the Kalman filter using the constant velocity linear motion model As for the combination with the prediction of the position of the measurement point data group, etc., the situation when only one vehicle passes through the obstacle detection area 7 from the left to the right at a constant speed V as described above. A specific example will be described (in FIGS. 4 (a) to (f), the measurement point data group and the representative point are shown superimposed on the same image image diagram). This is an intuitive image of the tracking information data).

  In this case, the measurement point data group and its representative point start from the front end of the vehicle entering from the left end of the obstacle detection area 7 (see FIG. 4A), and sequentially from the front end of the vehicle. In the obstacle detection area 7, it extends to the central portion (see FIG. 4B), relates to the entire vehicle (see FIG. 4C), and remains related to the entire vehicle. The vehicle moves from the left to the right (see FIG. 4D) and relates to the center and rear end of the vehicle that has started to advance from the right end of the obstacle detection area 7 (see FIG. 4E). Is reduced to only the rear end portion (see FIG. 4F), and finally disappears (see FIGS. 3D and 3E).

  Looking at the measurement point data group according to the movement of the vehicle and the position change and speed change of the representative point, assuming that the representative point is always at the center of the measurement point data group for the sake of simplicity ( When the vehicle is not present, the speed is “0” (see the time t1 position in the thick solid line graph of FIG. 4G), and the vehicle is approaching from there (see FIG. 4 (g)). a) and (b)), the speed of the representative point becomes the intermediate speed V / 2 between the speed “0” at the stationary left end of the obstacle detection area 7 and the front end speed V of the vehicle (see FIG. 4 (g)). While the vehicle is in the obstacle detection area 7 (see FIGS. 4C and 4D), the speed of the representative point becomes the same speed V as that of the vehicle. (See time t2 to t3 portion of the thick solid line graph in FIG. 4G).

Further, during the advancement of the vehicle (see FIGS. 4E and 4F), the speed of the representative point is an intermediate speed V between the rear end speed V of the vehicle and the speed “0” of the stationary right end of the obstacle detection area 7. / 2 (refer to the time t3 to t4 portion of the thick solid line graph in FIG. 4G), and the speed of the representative point returns to “0” after the vehicle has advanced (see the thick solid line graph in FIG. 4G). (See time t4).
Thus, even if the vehicle speed is a single speed V, the speed of the representative point is divided into the time when the area does not exist, the time when the area enters and exits, and the time when the area exists. Since the discontinuous change is kept at half V / 2, the tracking state such as the tracking speed is stabilized (see the one-dot chain line in FIG. 4G).

On the other hand, unlike the crossing obstacle detection device 10 in which the limitation in the obstacle detection area of the measurement data is completed before the grouping process in the image, it is possible to determine whether or not the limitation in the obstacle detection area of the measurement data is performed. In the case of a crossing obstacle detection device with a slow speed, the speed of the representative point and its discontinuous change in the tracking process are the same V as the vehicle speed (see the broken line graph in FIG. 4G), so the Kalman filter is used. Since the accuracy of the estimated position remains low (see the two-dot chain line in FIG. 4G), the tracking state such as the tracking speed is difficult to stabilize.
Therefore, the crossing obstacle detection device 10 has improved tracking ability in the tracking process by mitigating the adverse effect of discontinuous jumping speed change by a simple method of limiting data at an appropriate timing. I can say.

Although illustration is omitted, the level crossing obstacle detection device of the second embodiment is different from the level crossing obstacle detection device 10 described above in that the portion of the tracking program 24 that uses the tracking extinction element is changed. Is a point.
Specifically, a value obtained by converting the moving speed of an object on a cutway close to 4 km / h, which is generally the average walking speed of a healthy person, into the moving speed of a representative point in a measurement point data group in measurement data is set in advance as a predetermined speed At the time of tracking processing, when the speed related to the measurement point data group is slower than the above-mentioned predetermined speed, the tracking extinction time element is also stored after the extinction in the grouping processing of the measurement point data group to be tracked as in the first embodiment. The tracking information is continued before the lapse, but when the speed related to the measurement point data group is faster than the predetermined speed, unlike the case of the first embodiment, the tracking point disappears and the measurement point data group to be tracked is ignored. When the grouping process disappears, the corresponding tracking information is immediately disappeared.

Similarly, although the illustration is omitted, the level crossing obstacle detection apparatus of the third embodiment is different from the above level crossing obstacle detection apparatus 10 in that the tracking processing portion using the tracking disappearance element is changed.
Specifically, not only one tracking extinction element but also a plurality of elements are associated with each measurement point data group one by one. In the tracking process, the tracking extinction element is individually changed for each measurement point data group to be tracked. However, the change process is performed when the movement speed of the representative point of the corresponding measurement point data group is changed. The tracking disappearance value increases or at least keeps the current state when it slows down, and the tracking disappearance value decreases or at least keeps the current value when the movement speed of the representative point of the corresponding measurement point data group becomes faster Done. Furthermore, for each measurement point data group, when the moving speed of the representative point is faster than the predetermined speed of Example 2, by clearing the corresponding tracking disappearance value to zero time, easily and quickly, The extension of the tracking based on the disappearance of the tracking is also omitted.

[Others]
In the above-described embodiment, only one obstacle detection area 7 is defined. However, the obstacle detection area is set in a case where the number of parallel lines is large and the obstacle detection area related to one railroad crossing is very long. You may divide | segment into several partial area | regions which do not overlap or overlap.
In this case, for example, the final determination program 25 is divided into an intermediate determination program and a determination integrated program, so that it can be handled with software relatively easily.

  In the above embodiment, a situation is described in which one measurement point data group is selected for one level crossing vehicle, but it is not uncommon for a plurality of measurement point data groups to be selected for one level crossing vehicle. For example, the front and rear portions of the level crossing vehicle are clearly detected, but the middle part is unclear, or both images obtained by plane sweeping the left and right sides of the level crossing vehicle with different aerial wave transmission / reception units 12 and 12 pass Multiple measurement point data groups tend to be present, such as when they are separated at the front and back of the body. Further, there may be a case where only one measurement point data group is selected for a plurality of level crossing vehicles such as when two level crossing vehicles approach. In any case, if there is at least one measurement point data group, it is safe to determine that an obstacle exists, so there is no inconvenience apart from the large amount of data processing.

  In the above-described embodiment, the distance and direction of the measurement point data are both variable values, and the measurement data obtained by connecting the measurement point data is held in the array region of 2 rows and N columns. When the direction of sweep measurement is determined in advance, for example, when the measurement start direction is a known fixed direction and the subsequent direction difference is also known, the portion of known data related to the direction in the measurement data is one row N. In the variable data portion related to the distance in the measurement data, the distance measuring point data composed of the distance measurement values in one known direction is measured in a row in the N known directions. The distance data may be stored in an array area of one row and N columns. As a result, the measurement point data group that is data-processed as a two-dimensional matrix becomes a so-called distance measurement point group that is data-processed as a one-dimensional vector, thereby further reducing the burden on the fail-safe computer 14. It will be.

  The crossing obstacle detection device of the present invention is not limited to the detection of only obstacles, but in addition to obstacles such as detection of crossings in a situation where there is no obstacle, detection of trains, etc. The present invention can also be applied to those that detect other things.

4 ... Railroad crossing, 7 ... Obstacle detection area (monitoring area),
10 ... Crossing obstacle detection device,
11-13 ... Planar sweep measurement unit, 11 ... Rotation control unit,
12 ... aerial propagation wave transmission / reception unit, 13 ... signal processing unit, 14 ... fail-safe computer,
20 ... judgment program,
21 ... Data input program, 22 ... Data limitation program,
23 ... Grouping program (in-image grouping process),
24. Tracking program (tracking process over time),
25 ... Final judgment program

Claims (4)

  A plane sweep measurement unit that sweeps the air propagation wave to obtain measurement data of the reflection position, holds detection area definition data that defines an obstacle detection area of a crossing, and is based on the measurement data and the detection area definition data A crossing obstacle detection device including a fail-safe computer equipped with a determination means for determining whether or not an obstacle exists, wherein the determination means includes the obstacle among the measurement data based on the detection area defining data. A grouping process and a measurement for specifying a measurement point data group of obstacle candidates using the data for subsequent processing by performing data limitation processing that uses data narrowed down to data relating to positions belonging to the detection area as data for subsequent processing Tracking processing for tracking the position of the representative point of the point data group over a period of time when the tracking disappears, and the tracking Crossing obstacles, characterized in that it is as to determine the presence or absence of an obstacle depending on the presence or absence of a target object processing detection apparatus.   The level crossing obstacle detection device according to claim 1, wherein in the tracking process, the tracking disappearance element is increased or decreased according to whether a speed related to the measurement point data group is slow or fast.   3. The railroad crossing according to claim 1, wherein, in the tracking process, when the speed related to the measurement point data group is faster than a predetermined speed, the tracking disappearance time is set to zero time or ignored. Obstacle detection device.   The level crossing obstacle detection device according to claim 3, wherein the predetermined speed is set to a value corresponding to an average walking speed of a healthy person.

Images