JP2019206318A - Monitoring device - Google Patents

Abstract

To provide a monitoring device which is improved to detect obstacles accurately.SOLUTION: A monitoring device 100 comprises a first measurement part 41 of a vehicle control device 31 which functions as a first judgement part for detecting obstacles in a prescribed decision frame AR1 that is set in response to a travel position along a line RL as a track and a second measurement part 42 of the vehicle control device 31 which functions as a second judgement part for detecting obstacles in comparison with a reference distance image (reference information) regarding spatial arrangements in a prescribed recognition area AR2 including surroundings of the line RL as the track.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a monitoring device for monitoring obstacles existing around a track in a railway or other transportation facility.

  As an image processing apparatus that can be used to monitor an obstacle, a stereo image processing apparatus that calculates a parallax that gives a distance to a forward object by performing a matching process on a stereo image, and that has the same coordinates as the reference point of the reference image A point on the image is detected as a primary corresponding point, a point that is shifted by the parallax due to rail extraction on the same scanning line from the primary corresponding point is set as a secondary corresponding point, and matching processing is performed from the secondary corresponding point to the left and right of the corresponding image. There is one that calculates the correlation value by performing (see Patent Document 1).

  Also, as a railway obstacle monitoring device, the difference between the observed image data and the background image data is calculated, and a foreground region having a large degree of difference between the two pixel values is detected, and an observed image in which the foreground region is detected There is one that determines whether or not a point in the foreground area exists within the building limits by performing a trajectory detection process on the data to obtain a position or locus of the point in the foreground area with respect to the trajectory axis ( Patent Document 2).

  However, in the case of Patent Document 1, although the distance to the obstacle ahead can be determined by the parallax, it depends on only the parallax, and it is not easy to increase the detection accuracy of the obstacle. Further, in the case of Patent Document 2, it can be detected that a point in the foreground region exists within the building limit, but two-dimensional information obtained by extracting the difference between the captured images is a premise, and it is possible to improve the detection accuracy of the obstacle. Not easy.

JP 2016-192106 A JP, 2006-52849, A

  The present invention has been made in view of the above-described points, and an object thereof is to provide a monitoring device with improved obstacle detection accuracy.

  In order to achieve the above object, a monitoring apparatus according to the present invention includes a first determination unit that detects an object that exists within a predetermined determination frame that is set according to a travel position along a track, and a periphery of the track. And a second determination unit that detects an object based on reference information regarding spatial arrangement in a predetermined recognition area.

  In the monitoring apparatus, the first determination unit detects an object existing in a predetermined determination frame set according to the travel position along the track, and the second determination unit has a predetermined recognition area including the periphery of the track. Since the object is detected based on the reference information regarding the spatial arrangement in the object, the detection accuracy of the obstacle can be improved by combining the object detection in the determination frame and the change in the spatial arrangement in the recognition area.

  In a specific aspect of the present invention, in the monitoring device, the first determination unit irradiates an electromagnetic wave in a predetermined determination frame and detects a reflected wave from an object in the predetermined determination frame. In this case, simple and highly accurate measurement using radio waves and light becomes possible.

  In another aspect of the present invention, the first determination unit refers to frame data in which a predetermined determination frame is generated in association with the travel position by pre-travel, and determines a predetermined determination frame corresponding to the travel position during the main travel. Set. In this case, an appropriate determination frame can be set corresponding to the travel position.

  In still another aspect of the present invention, the predetermined determination frame is generated based on the trajectory center and the predetermined width. In this case, the determination frame is easily defined.

  In still another aspect of the present invention, the second determination unit includes measurement information related to the spatial arrangement obtained at the travel position during the main travel, and reference information related to the spatial layout obtained in association with the travel position by the preliminary travel. Extract the difference. In this case, highly accurate object detection using prior reference information is possible.

  In still another aspect of the present invention, the predetermined recognition area is set to a wider azimuth area including a predetermined determination frame. In this case, it is possible to detect an object around the upper, lower, left, and right sides of the determination frame, and the accuracy of object detection can be increased in the determination frame.

  In still another aspect of the present invention, the predetermined recognition area is set to a farther area including a predetermined determination frame. In this case, an object can be detected behind the determination frame.

  In still another aspect of the present invention, the first determination unit includes a radar device or a laser radar device for measuring the azimuth and distance of the object, and the second determination unit measures the arrangement of the object three-dimensionally. To have a camera or laser radar device. In this case, an object in the frame or an object in the recognition area can be accurately determined.

  In still another aspect of the present invention, it is determined that an obstacle exists when an object is detected by either the first determination unit or the second determination unit.

It is a conceptual block diagram explaining the monitoring apparatus of 1st Embodiment mounted in the train. It is a conceptual block diagram explaining the main body of a monitoring apparatus. (A) and (B) are a conceptual side view and a front view for explaining a determination range by the first and second determination units, and (C) and (D) are determination frames of the first determination unit. It is a figure explaining a modification. It is a flowchart explaining preparation of the reference data by the monitoring apparatus. It is a flowchart explaining the monitoring operation | movement by a monitoring apparatus. It is a conceptual plan view explaining a specific operation. (A)-(C) are the conceptual perspective views explaining a concrete operation | movement. It is a flowchart explaining the monitoring operation | movement by the monitoring apparatus of 2nd Embodiment.

  The monitoring device 100 according to the first embodiment shown in FIG. 1 is a part of the on-board device 200 incorporated in the train TR.

  The monitoring device 100 performs communication between the vehicle control device 31 that controls the operation of each unit in an integrated manner, the vehicle speed detection unit 34 that detects the current speed of the train TR, and the ground unit provided on the track RL side. Measurement that detects an object in front of the train TR, a communication unit 37 that enables communication between a vehicle upper member 35 that detects the position of the train TR on the RL, a train operation management system (not shown) or a command station. And a second measurement unit 42 that performs measurement to detect objects in front of and around the train TR. The on-board device 200 decelerates the train TR in addition to the elements 31, 34, 35, 37, 41, and 42 constituting the monitoring device 100, and a drive device 32 including a motor for accelerating the train TR. And an in-vehicle output unit 36 such as a speaker or a display unit, which are notification means for transmitting various types of information to passengers and the like.

  The vehicle control device 31 operates each part of the train TR based on instructions from the driver, etc., and enables the train TR to travel at an appropriate speed and stop at an appropriate timing, and to stop an automatic train in an emergency. It has a function. The vehicle control device 31 grasps the current traveling position of the train TR based on the accumulated distance using the vehicle speed detection unit 34 and the calibration using the vehicle upper member 35. The vehicle control device 31 constitutes a first determination unit together with the first measurement unit 41. That is, the vehicle control device 31 uses the first measurement unit 41 to detect an object that exists within a predetermined determination frame that is set according to the travel position along the track RL. Furthermore, the vehicle control device 31 constitutes a second determination unit together with the second measurement unit 42. That is, the vehicle control device 31 uses the second measurement unit 42 to detect the object based on the reference information regarding the spatial arrangement in the predetermined recognition area including the periphery of the track or the track RL.

  The current travel position of the train TR is not limited to the one using the vehicle speed detection unit 34 as described above, but the speed value and the GPS signal obtained by monitoring the Doppler effect at the time of measurement using a radar or other distance measuring device. It may be based on a speed value using the Doppler effect. In addition, position detection using RFID, position detection by Michibiki or other satellite positioning is also possible.

  As shown in FIG. 2, the vehicle control device 31 includes an arithmetic processing unit 101, a storage unit 102, an input / output unit 103, and an interface unit 104. Specifically, the vehicle control device 31 includes a computer on which a program for traveling control is mounted, and an obstacle monitoring program is added to the traveling control program in addition to the general control program. The vehicle control device 31 or the on-board device 200 has a basic role or operation for controlling the running state of the train TR. Hereinafter, the vehicle control device 31 and the like are mainly used from the aspect of the obstacle monitoring function. explain.

  The arithmetic processing unit 101 operates based on programs and data stored in the storage unit 102, performs processing based on information obtained from the input / output unit 103 and the interface unit 104, and stores the progress and results of the processing. And is presented to the input / output unit 103. Further, the arithmetic processing unit 101 operates the first measuring unit 41 via the interface unit 104 based on a program or the like, and sets a size corresponding to the gauge width along the line RL as shown in FIG. Obstacles existing within a predetermined determination frame AR1 are monitored. In addition, the arithmetic processing unit 101 operates the second measurement unit 42 via the interface unit 104 based on a program or the like, and has a wider range than the determination frame AR1 along the line RL as illustrated in FIG. Obstacles existing in a predetermined recognition area AR2 to be covered are monitored. As the obstacle, typically, an object of a certain size or larger, such as a person, a car, a rockfall, or an animal, can be considered. The arithmetic processing unit 101 is based on the detection result of the presence or absence of an object based on the measurement data in the determination frame AR1 acquired by the first measurement unit 41, or the measurement data in the recognition area AR2 acquired by the second measurement unit 42. If it is determined from the detection result of the presence or absence of an object that there is an obstacle ahead of the traveling direction, the brake device 33 is operated to decelerate or stop the vehicle, and the in-vehicle output unit 36 suddenly brakes the passenger. Informs that the stop will be performed.

  As shown in FIGS. 3A and 3B, each determination frame AR1 in monitoring using the first measurement unit 41 is a quadrangular pyramid region extending along the track RL as the track or the track center RC. It is set to a range of about the vehicle limit, and has a predetermined width as the gauge width of the track RL, and has the predetermined width and the height of the train TR. The depth distance D1 of the determination frame AR1 is set in relation to the braking distance, but can be set to 40 m, for example. The determination by the determination frame AR1 is for a medium distance. Note that the determination frame AR1 moves with the movement of the train TR, and the entire determination frame TA1 that is a set of continuous determination frames AR1 is a quadrangular columnar region. The recognition area AR2 in monitoring using the second measuring unit 42 is a quadrangular prism-shaped area extending along the track RL or the track center RC as a track, and is set to a range equal to or greater than the building limit and includes a determination frame AR1. It covers a wide range. That is, the recognition area AR2 is set to an area that is azimuthally wider including the determination frame AR1, and is set to a farther area including the determination frame AR1. The recognition area AR2 is specifically set to include the periphery beyond the gauge width or the road bed width of the track RL. The width of the recognition area AR2 can be several meters wide next to the track RL, and the depth distance D2 of the recognition area AR2 is set in relation to the braking distance, but can be set to 100 m, for example. The identification by the recognition area AR2 is intended for medium distance and long distance.

  The determination frame AR1 is set on the basis of the track center RC so that the front end FA1 of the determination frame AR1 covers the track RL as a track, and at which point on the track RL the train TR exists. Depending on the time. The determination frame AR1 is given as a function of the travel position on the track RL, and is specifically determined by running the train TR or the measurement vehicle in advance, and is stored in the storage unit 102 as a determination frame database for each distance or travel position. Stored. In actual measurement, the arithmetic processing unit 101 reads the determination frame AR1 corresponding to the travel position from the determination frame database stored in the storage unit 102 based on the travel position of the train TR on the track RL, thereby determining the travel position. A frame AR1 is set, measurement is performed using the first measurement unit 41, and it is determined whether there is an obstacle or other object of a predetermined size or more in the determination frame AR1. The range or position of the determination frame AR1 can be set at intervals or increments along the track RL such as 1 m, for example, with reference to the track center RC of the track RL and the head position of the train TR. In a curved section where the train TR bends greatly, there is a possibility that a portion that should originally be the determination frame AR1 may be hidden behind the object. In this case, the determination frame AR1 is not set and monitoring using the first measurement unit 41 is temporarily performed. Can be interrupted. When the determination frame AR1 is partially hidden behind the object, the determination frame AR1 can be partially validated.

  The recognition area AR2 is set on the basis of the track center RC so as to widely cover the track RL as a track and its periphery, and changes from moment to moment depending on where the train TR exists on the track RL. To do. The recognition area AR2 itself is not databased by itself, and a distance image is acquired in the recognition area AR2. The distance image in the recognition area AR2 is given as a function of the travel position on the track RL, and is specifically acquired by traveling the train TR or the measurement vehicle in advance, for example, and is stored as a distance image database for each distance or position. 102 etc. In actual measurement, the arithmetic processing unit 101 reads the reference distance image (reference information) in the recognition area AR2 as reference information stored in the storage unit 102 based on the travel position of the train TR on the track RL, An operating distance image (measurement information) is measured in the recognition area AR2 using the second measurement unit 42, and it is determined whether or not an obstacle or other object of a predetermined size or more has appeared in the recognition area AR2. The range of the recognition area AR2 can be set at intervals or increments along the track RL such as 1 m, for example, with reference to the track center RC of the track RL and the head position of the train TR. In a curved section where the train TR bends greatly, there is a possibility that the portion that should be the recognition area AR2 is obstructed by the near view. In this case, the close view becomes the reference distance image (reference information), and the second measurement unit 42 is used. Monitoring can be continued.

  As shown in FIG. 3C, auxiliary determination frames AR11 and AR12 may be added for the determination frame AR1. The auxiliary determination frames AR11 and AR12 are set so that the depth distance D1 is different from the basic determination frame AR1. In this way, by using the plurality of determination frames AR1, AR11, AR12 set for each distance, it is possible to determine the presence or absence of an obstacle at each position, and the collision risk determination in consideration of the speed of the train TR. Accuracy can be increased.

  The determination frame AR13 illustrated in FIG. 3D covers an area that is wider than the determination frame AR1 illustrated in FIG. 3B and extends laterally beyond the vehicle limit or the building limit.

  Returning to FIG. 1, the first measurement unit 41 includes a radar device that performs object detection using, for example, millimeter waves (that is, electromagnetic waves), and an object that exists in the front measurement region (specifically, in the determination frame AR1). Measure. The first measurement unit 41 irradiates the front of the train TR with an electromagnetic wave while scanning two-dimensionally using an antenna array or the like, and the direction and detection timing of the reflected wave that is reflected by the object and returned. The position including the depth direction of the object is measured. At this time, information on the size of the object can also be obtained from the angle range of the reflected wave. If a determination frame AR1 as shown in FIG. 3A is set when an object is detected by the first measurement unit 41, an object or an obstacle existing in the determination frame AR1 can be extracted. Since the first measurement unit 41 performs imaging and distance calculation at a high speed as the train TR travels, the first measurement unit 41 can measure a distance image in real time for the foreground that changes in front of the train TR. The first measurement unit 41 is not limited to a radar device, and for example, a laser radar device or a three-dimensional image sensor that scans with a polygon mirror, a MEMS mirror, or the like using infrared light or other laser light as electromagnetic waves can be used.

  The second measurement unit 42 includes a stereo camera that performs imaging using, for example, visible light or infrared light, and measures a distance image of the foreground in the visible region or the infrared region in cooperation with the arithmetic processing unit 101. That is, the 2nd measurement part 42 can image | photograph the front image of train TR by a stereo system, and the arithmetic processing part 101 can calculate the distance to each part in the image | photographed image. The arithmetic processing unit 101 extracts an object having a predetermined size or more from the image, and calculates the distance to each object using parallax. In this case, a distance is given to each part of the front scenery of the second measuring unit 42, and a distance image is measured. Since the second measurement unit 42 and the arithmetic processing unit 101 perform imaging and distance calculation at high speed as the train TR travels, the distance image can be measured in real time for the foreground that changes in front of the train TR. About the 2nd measurement part 42 and the arithmetic processing part 101, another optical principle, for example, a thing which moves a focus and extracts the object which focuses on each focus position may be used. About the 2nd measurement part 42 and the arithmetic processing part 101, not only a stereo camera but a laser radar apparatus or a three-dimensional image sensor can be used, for example.

  4 and 5 are flowcharts for explaining specific operations of the monitoring apparatus 100 shown in FIG.

  A reference data creation procedure will be described with reference to FIG. First, the train TR or the measurement vehicle is run in advance. During the preliminary traveling, the arithmetic processing unit 101 operates under the instruction of the operator, acquires the traveling position of the train TR using the vehicle speed detecting unit 34 and the like, and uses the second measuring unit 42 to determine the traveling position. An image is acquired for each minimum unit and stored in the storage unit 102 or the like (step S11). Next, the arithmetic processing unit 101 determines the determination frame AR1 for each minimum unit of the travel position (step S12). The determination frame AR1 can be determined using, for example, the trajectory center RC obtained from the foreground image obtained using the second measurement unit 42, and specifically, the front end FA1 is determined in the left-right direction of the trajectory center RC. It has a predetermined width and a predetermined height width above the track center RC. At this time, the size and center orientation of the front end FA1 of the determination frame AR1 are stored in the storage unit 102 in association with the travel position. Next, the arithmetic processing unit 101 performs image processing on the image obtained by using the second measurement unit 42, and obtains a distance image for each minimum unit of the traveling position (step S13). The distance image obtained in step S13 can be limited to the range of the recognition area AR2 determined based on the trajectory center RC. Next, the arithmetic processing unit 101 operates under the instruction of the operator, creates a database in which the direction of the determination frame AR1 is associated with each traveling position, and stores this determination frame database in the storage unit 102 or the like (step S102). S14). Next, the arithmetic processing unit 101 operates under the instruction of the operator, creates a database in which distance images are associated with each traveling position, and stores this distance image database in the storage unit 102 or the like (step S15). The distance image database is reference information composed of a number of reference distance images that are functions of the traveling position. Here, in the reference distance image, an area of a predetermined distance or more can be processed as a data missing point or an infinite point, but can also be set at the outer edge of the recognition area AR2.

  With reference to FIG. 5, the obstacle detection at the time of operation driving or main driving will be described. The arithmetic processing unit 101 acquires the travel position of the train TR using the vehicle speed detection unit 34 or the like (step S21). Next, the arithmetic processing unit 101 performs object measurement using the first measurement unit 41 (step S22). Thereby, when an object exists ahead of train TR, the direction (angle) and distance of the object are obtained. Next, the arithmetic processing unit 101 sets the determination frame AR1 corresponding to the travel position obtained in step S21, and determines whether the azimuth and distance of the object obtained in step S22 are within the set determination frame AR1. Judgment is made (step S23). When the arithmetic processing unit 101 determines that an object exists in the determination frame AR1 (Yes in step S23), the arithmetic processing unit 101 outputs a warning to the driver that an obstacle exists on the track via the input / output unit 103 (step S23). S24). At this time, the arithmetic processing unit 101 can cause the train TR to perform an emergency stop by appropriately operating the brake device 33. On the other hand, when it is determined that there is no object in the determination frame AR1 (No in step S23), the arithmetic processing unit 101 returns to step S21 and resumes the process of acquiring the travel position of the train TR. In parallel with the measurement of the object in the determination frame AR1 (steps S22 and S23), the arithmetic processing unit 101 performs distance image measurement using the second measurement unit 42 (step S25). Next, the arithmetic processing unit 101 extracts the difference by comparing the operating distance image obtained in step S25 with the reference distance image (reference information) corresponding to the travel position obtained in step S21 (step S26). Next, the arithmetic processing unit 101 determines whether or not the difference obtained in step S26 corresponds to a significant spatial change in the recognition area AR2 (step S27). Whether or not it corresponds to a significant spatial change is determined by using the outline, distance, area, etc. of the pixel group corresponding to the difference as a determination element. If the arithmetic processing unit 101 determines that a significant spatial change has occurred in the recognition area AR2 (Yes in step S27), the driver has an obstacle on or around the track via the input / output unit 103. A warning output is made (step S24). On the other hand, when the arithmetic processing unit 101 determines that there is no object in the determination frame AR1 (No in step S27), the arithmetic processing unit 101 returns to step S21. The above process is repeated until the operation running of the train TR is completed (No in step S28).

  FIG. 6 is a plane illustrating a specific example of the recognition area AR2, and there are a building P1, an approach prevention fence P2, a railroad crossing P3, a road P4, a traffic light P5, an overhead pole P6, a tree P7, and the like around the track RL. These objects P1, P2, P3, P4, P5, P6, and P7 are extracted as objects constituting the reference distance image associated with the travel position during the preliminary travel and the main travel. FIG. 7A is a conceptual diagram for explaining a reference distance image obtained during preliminary traveling, and an image of a crossing P3 and the like is obtained. FIG. 7B is a conceptual diagram for explaining the operating distance image obtained during the actual travel, and an image of the pedestrian P9 is obtained behind the level crossing P3. In this example, since the pedestrian P9 is waiting outside the recognition area AR2, the arithmetic processing unit 101 determines that there is no obstacle in the recognition area AR2. On the other hand, FIG. 7C is a conceptual diagram for explaining an operating distance image obtained at another timing, and an image of a pedestrian P9 is obtained inside the level crossing P3. In this example, since the pedestrian P9 has entered the recognition area AR2, the arithmetic processing unit 101 determines that an obstacle exists in the recognition area AR2.

  As described above, in the monitoring device 100 according to the present embodiment, the first measurement unit 41 and the vehicle control device 31 as the first determination unit are set in accordance with the travel position along the track RL that is a track. An object existing in the determination frame AR1 is detected, and the second measurement unit 42 and the vehicle control device 31 as the second determination unit are spatially arranged in a predetermined recognition area AR2 including the periphery of the track RL that is a track. Since the object is detected based on the reference information regarding, the detection accuracy of the obstacle can be increased by a combination of the object detection in the determination frame AR1 and the spatial arrangement change in the recognition area AR2.

[Second Embodiment]
Hereinafter, a second embodiment obtained by modifying the first embodiment will be described with reference to FIG. Although the monitoring device 100 according to the present embodiment is different in specific operation, the monitoring device 100 has the same configuration as that of the first embodiment, and thus detailed description of the entire monitoring device 100 is omitted.

  In the case of the monitoring apparatus 100 according to the second embodiment, the arithmetic processing unit 101 determines whether the spatial change detected in the recognition area AR2 corresponds to a construction limit or other limit frame that is close to the determination frame AR1 in the recognition area AR2. Is determined (step S41). If the spatial change is within the limit frame (Yes in step S41), the arithmetic processing unit 101 outputs a warning to the driver that there is an obstacle on the track via the input / output unit 103 (step S24). On the other hand, when the spatial change is outside the limit frame (No in step S41), the arithmetic processing unit 101 determines the driver via the input / output unit 103 according to the distance to the object or the target and the state of proximity to the limit frame. Is warned that there is an obstacle around the track (step S42).

[Others]
The present invention is not limited to the above embodiment, and can be implemented in various modes without departing from the scope of the invention.

  For example, in the above embodiment, a database composed of a series of reference distance images is created by associating distance images for each traveling position (step S15), but the distance along the railroad using distance images for each traveling position, etc. Image mapping can also be performed. In this case, spatial arrangement information along the line is obtained by distance image mapping, and if a traveling position is given, a reference distance image corresponding to the traveling position can be acquired from the spatial arrangement information. Thereby, the spatial change in the recognition area AR2 can be extracted based on the detailed reference distance image in which the traveling position is continuously finely changed.

  In the above description, it is assumed that the second measurement unit 42 for obtaining the operation distance image during operation travel is the same as the second measurement unit 42 for obtaining the reference distance image during preliminary travel. The second measurement unit 42 for measuring the distance image and the second measurement unit 42 for measuring the reference distance image may be different using different measurement principles. For example, when acquiring the reference distance image, a visible light stereo camera can be used, and when acquiring the operation distance image, an infrared light stereo camera or a laser radar device can be used. When comparing data from different types of measuring devices, the reliability of judgment can be increased by techniques such as calibration and normalization.

  In the above, the distance image is obtained using the second measurement unit 42, but a two-dimensional image can be obtained instead of the distance image. In this case, instead of detecting the object based on the three-dimensional information as the difference between the distance images, the spatial arrangement of the object is detected based on the two-dimensional difference information. For an object or an obstacle that does not move, the distance to the object can be determined based on difference information of two-dimensional images at different times.

  The on-board device 200 enables automatic operation of the train TR under the control of a train operation management system (not shown). The automatic operation is performed in a state where the driver is onboard, and there is no driver. It can also include unmanned automated driving. As for automatic driving, in addition to the case where the entire operation section is completely unmanned automatic driving without a driver, a part of the operation section is set to automatic driving with a driver on it, It is conceivable that the section is not automatically operated but is normally operated by a driver.

  Advance travel for acquiring a reference distance image (reference information) can be repeated as appropriate, and the reference distance image can be updated periodically.

  Information regarding the reference distance image (reference information) and the determination frame AR1 can be acquired for a while from the train operation management system via the communication unit 37.

  The monitoring device 100 can be incorporated in the remote train operation management system side instead of being incorporated in the on-board device 200.

  The monitoring target by the monitoring device 100 is not limited to the railroad track RL, but may be a tram track.

  DESCRIPTION OF SYMBOLS 31 ... Vehicle control apparatus, 32 ... Drive apparatus, 33 ... Brake device, 34 ... Vehicle speed detection part, 35 ... Vehicle upper part, 36 ... In-vehicle output part, 37 ... Communication part, 41 ... 1st measurement part, 42 ... 2nd Measuring unit, 100 ... monitoring device, 101 ... arithmetic processing unit, 102 ... storage unit, 103 ... input / output unit, 104 ... interface unit, 200 ... on-vehicle device, AR1 ... original determination frame, AR11, AR12, AR13 ... determination frame , AR2 ... recognition area, FA1 ... front end, RC ... track center, RL ... track, TR ... train

Claims (9)

A first determination unit that detects an object existing within a predetermined determination frame set according to a travel position along the track;
A monitoring apparatus comprising: a second determination unit that detects an object based on reference information regarding a spatial arrangement in a predetermined recognition area including a periphery of a trajectory.   The monitoring apparatus according to claim 1, wherein the first determination unit irradiates an electromagnetic wave in the predetermined determination frame and detects a reflected wave from an object in the predetermined determination frame.   The first determination unit sets the predetermined determination frame corresponding to the travel position at the time of the main travel with reference to the frame data generated in association with the travel position by the pre-travel to generate the predetermined determination frame. The monitoring apparatus according to any one of 1 and 2.   The monitoring apparatus according to claim 1, wherein the predetermined determination frame is generated based on a trajectory center and a predetermined width.   The second determination unit extracts a difference between measurement information related to a spatial layout obtained at a travel position during the main travel and the reference information related to a spatial layout obtained in association with the travel position by preliminary travel. The monitoring apparatus as described in any one of 1-4.   The monitoring apparatus according to claim 5, wherein the predetermined recognition area is set to an azimuthally wider area including the predetermined determination frame.   The monitoring apparatus according to claim 5, wherein the predetermined recognition area is set to a farther area including the predetermined determination frame. The first determination unit has a radar device or a laser radar device for measuring the azimuth and distance of an object,
The monitoring device according to any one of claims 1 to 7, wherein the second determination unit includes a camera or a laser radar device for measuring the arrangement of an object three-dimensionally.   The monitoring apparatus according to any one of claims 1 to 8, wherein when an object is detected by any one of the first determination unit and the second determination unit, it is determined that an obstacle exists.

Images