Obstacle detection support system and obstacle detection support method

Abstract

The purpose of the present invention is, in an obstacle detection assistance system mounted on a track transportation system, to reduce false alarms without interrupting driving assistance for a driver, In order to achieve the foregoing, this obstacle detection assistance system for assisting obstacle detection while driving is characterized by comprising: a driving assistance display unit for displaying driving assistance information constituting information for assisting driving of a vehicle; an obstacle detection unit for detecting an obstacle which is an obstacle to vehicle travel; a determination criterion determination unit for determining a determination criterion when the obstacle detection unit detects an obstacle; and an external environment recognition unit for recognizing an external environment of the vehicle. The obstacle detection assistance system is also characterized in that: the determination criterion determination unit sets a determination criterion in accordance with the external environment; and a threshold value for the number of instances of obstacle detection by the obstacle detection unit, a degree of matching of detection results by the obstacle detection unit, or both the threshold value for the number of instances of obstacle detection and the degree of matching of the detection results are set as the determination criterion / criteria.

Description

Technical Field

[0002] , ,

[0001] The present invention relates to an obstacle detection support system mounted on an orbital transport system that travels on a defined track.

Background Art

[0002] In an orbital transport system where a transport vehicle travels on a track, when there is an obstacle on the track, it cannot be avoided by steering, so detecting obstacles on the track is important for improving the safety and operability of the orbital transport system. In recent years, research has been conducted on an obstacle detection support system that uses external sensors such as millimeter-wave radar, lidar, and cameras to detect obstacles on the track and notifies the driver of the detection results to support the driver's driving. In the obstacle detection support system, it is important to reduce false detection of obstacles in order to avoid complicated notifications to the driver. Patent Document 1 discloses a technique for reducing false alarms to the driver in an obstacle detection support system.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the obstacle detection support system, it is important to reduce false detection of obstacles in order to avoid complicated notifications to the driver. Generally, when the detection rate is improved, the false detection rate also tends to increase. The technique described in Patent Document 1 interrupts the driving support to the driver when the external environment of the train meets the false detection condition that a non-obstacle is falsely detected as an obstacle by the obstacle detection unit. Therefore, although false detection to the driver is reduced, there is a problem that driving support is not performed in the section that meets the false detection condition.

[0005] To address the above-mentioned problems, the present invention aims to reduce false alarms in an obstacle detection support system installed in a rail transport system without interrupting driver assistance to the driver. [Means for solving the problem]

[0006] To solve the above problems, the obstacle detection support system according to the present invention is a system that supports obstacle detection in vehicle driving, comprising: a driving support display unit that displays driving support information which is information that supports vehicle driving; an obstacle detection unit that detects obstacles that hinder the driving of the vehicle; a judgment criterion determination unit that determines the judgment criteria when the obstacle detection unit detects an obstacle; and an external environment recognition unit that recognizes the external environment of the vehicle, wherein the judgment criterion determination unit sets judgment criteria according to the external environment, and the judgment criteria are set to be either a threshold for the number of times the obstacle detection unit detects an obstacle or the degree of agreement of the detection results by the obstacle detection unit, or both the threshold for the number of times the obstacle detection unit detects an obstacle and the degree of agreement of the detection results. [Effects of the Invention]

[0007] According to the present invention, in an obstacle detection support system installed in a rail transport system, it is possible to reduce false alarms without interrupting driver assistance to the driver.

[0008] Other issues, configurations, and effects not mentioned above will be clarified in the section on embodiments for carrying out the invention and in the drawings. [Brief explanation of the drawing]

[0009] [Figure 1] This figure shows the configuration of an obstacle detection support system in the first embodiment of the present invention. [Figure 2] This is a flowchart illustrating the processing of the decision criterion determination unit in the first embodiment of the present invention. [Figure 3] This is a functional block diagram showing the processing of the obstacle detection unit in the first embodiment of the present invention. [Figure 4]This figure shows the obstacle determination for each threshold performed by the obstacle determination unit in the first embodiment of the present invention. [Figure 5] This is a flowchart illustrating the processing of the decision criterion determination unit in the second embodiment of the present invention. [Figure 6] This figure shows the obstacle determination for each detection method performed in the obstacle determination unit in the second embodiment of the present invention. [Modes for carrying out the invention] [Examples]

[0010] Figure 1 shows the configuration of the obstacle detection support system. The present invention consists of an external sensor 110 mounted on train 100 that senses the external environment, and an obstacle detection support system 120 that notifies the driver of the presence or absence of obstacles ahead and supports the driver's operation. The obstacle detection support system 120 consists of an external environment recognition unit 121 that recognizes the external environment of train 100 based on sensor information from the external sensor 110, a judgment criterion determination unit 122 that determines the judgment criteria for obstacle detection based on the information from the external environment recognition unit 121, an obstacle detection unit 123 that detects obstacles ahead of train 100 based on the sensor information from the external sensor 110 and the judgment criteria from the judgment criterion determination unit 122, and a driving support display unit 124 that notifies the driver of support content information based on the obstacle detection result information from the obstacle detection unit 123.

[0011] The external sensor 110 senses the conditions around the train 100 (especially in front of it) and transmits the sensing data to the external environment recognition unit 121 and the obstacle detection unit 123. The external sensor 110 includes cameras, LIDAR (laser rangefinder), millimeter-wave radar, and GNSS (Global Navigation Satellite System). Cameras include monocular cameras, stereo cameras, and infrared cameras. Each sensor is generally installed in multiples for redundancy. When the train 100 is moving in the direction of travel, the external sensor 110 installed on the leading car on the direction of travel side is used.

[0012] The external environment recognition unit 121 recognizes the external environment in front of the train 100 using external sensor data from the external sensor 110. The external environment recognized here includes the route environment such as stations, level crossings, and railway bridges, the route shape environment such as straight lines and curves, the time environment such as day and night, and the train environment such as vehicle speed and self-position. The route environment is mainly recognized using a DNN (Deep Neural Network) that has been trained on the target. The route shape environment is recognized from the shape of the rails ahead detected by image processing. Specifically, methods for detecting rails by image processing include searching for rails using brightness gradient information in the image, or detecting rails using semantic segmentation with a DNN. In this invention, it is sufficient that rails can be detected, and the method is not limited. The time environment may be estimated from the average brightness value of the external sensor information (especially image data), or time information included in GNSS data may be used. Alternatively, time information may be obtained from other onboard systems (not shown in the diagram). In this invention, it is sufficient that the time can be estimated, and the method is not limited. In the train environment, the vehicle speed may be estimated from GNSS data, or the speed obtained by performing SLAM (Simultaneous Localization and Mapping) using LIDAR data may be used. Alternatively, the speed may be obtained from another onboard system (not shown). For the self-position, the latitude and longitude information obtained from GNSS data may be used, or position information starting from a certain position obtained by integrating the vehicle speed may be used. Alternatively, the self-position may be obtained from another onboard system (not shown). In this invention, it is sufficient that the vehicle speed and self-position can be estimated, and the method is not limited. The external environment recognition unit 121 transmits the recognized external environment to the judgment criterion determination unit 122. For the route environment and route shape environment, a database that records the relationship between environmental information and the location where the environment exists may be maintained in advance, and external environment information at the current position of the train 100 may be obtained from the position of the train 100 and the database.

[0013] The Judgment Criteria Determination Unit 122 uses the external environment recognized by the External Environment Recognition Unit 121, vehicle information acquired from other on-board systems (not shown), its own position, and door opening / closing information to determine the object detection criteria for the Obstacle Detection Unit 123. It determines whether the external environment is one that may cause false detection of objects and determines the object detection criteria according to the external environment. The judgment criteria include continuous detection threshold information, which specifies how many consecutive detections are required in the time-series data of the object detection result for detection to be considered detected when the Obstacle Detection Unit 123 detects an object, and detection method information, which specifies whether to determine detection by logical OR or logical AND of the object detection result. The judgment criteria determined by the Judgment Criteria Determination Unit 122 may be the continuous detection threshold information, the detection method information, or both the continuous detection threshold information and the detection method information. The Judgment Criteria Determination Unit 122 transmits the determined judgment criteria to the Obstacle Detection Unit 123.

[0014] The obstacle detection unit 123 uses external sensor data from the external sensor 110 and judgment criteria from the judgment criteria determination unit 122 to grasp the situation in front of the train 100 and detect obstacles. The object detection processing of the obstacle detection unit 123 can utilize technologies used in the automotive field. For example, one method is to create a parallax image using a stereo camera and recognize the shape and position of objects in front from the parallax image. Another method is to recognize objects on an image using a DNN from a monocular image, or to recognize objects from LIDAR point cloud data. In this case, DNN is one of the means used in machine learning, and it can detect various objects by extracting and learning the features of the target object. In this invention, it is sufficient to detect obstacles, and the method is not limited. The obstacle detection unit 123 makes a final judgment on the obstacle according to the judgment criteria from the judgment criteria determination unit 122 based on the object detection results from the stereo camera, machine learning, and LIDAR. For example, if the judgment criterion determination unit 122 transmits "2" as continuous detection threshold information, the system will not consider an object to have been detected unless two consecutive objects are detected. This reduces the possibility of misidentifying randomly occurring detected objects as obstacles, thereby reducing false detections.

[0015] The driving support display unit 124 determines the support content to be notified to the driver based on the obstacle detection result information from the obstacle detection unit 123 and the vehicle speed, and displays it on the display device. For example, scenes are defined for each environment in front of the train 100, and the support content for the driver is determined for each scene. A scene is, for example, (1) There are no obstacles in or around the track. (2) There are obstacles around the track. (3) There is an obstacle in the track, but the distance from the train 100 is far or the speed of the train 100 is low, so there is no need to perform a braking operation to avoid a collision. (4) There is an obstacle in the track, but a collision can be avoided by immediately operating the brakes. (5) There is an obstacle in the track, and there is a high possibility that a collision cannot be avoided. That's it.

[0016] For each scene, the driving support display unit 124, for example, when the scene in front of the train 100 is (3), notifies the driver only of the presence of an object. When it is (4), it determines the support content such as notifying the driver to perform a braking operation.

[0017] The driving support display unit 124 notifies the driver of the support content for the driver through the HMI (Human Machine Interface) or through voice. In the present invention, it is only necessary that the support content for the driver is transmitted to the driver, and the transmission method is not limited.

[0018] Next, referring to FIG. 2, the processing of the determination criterion determination unit 122 according to the first embodiment of the present invention will be described. In the first embodiment, an example will be described in which the determination criterion output by the determination criterion determination unit 122 is continuous detection threshold value information.

[0019] FIG. 2 is a flowchart showing the processing executed by the determination criterion determination unit 122.

[0020] Step S201: The judgment criterion determination unit 122 acquires external environment information (such as the route environment including stations, level crossings, and railway bridges; the route shape environment including straight lines and curves; the time environment including day and night; and the train environment including vehicle speed) from the external environment recognition unit 121.

[0021] Step S202: The judgment criterion determination unit 122 obtains the vehicle speed from another in-vehicle system (e.g., a vehicle information control system). The vehicle speed may be the vehicle speed estimated by the external environment recognition unit 121. In this invention, it is sufficient to obtain the vehicle speed, and the method of calculation and acquisition is not limited.

[0022] Step S203: The judgment criterion determination unit 122 acquires its own position from another in-vehicle system (for example, a vehicle information control system). The self-position may be the self-position estimated by the external environment recognition unit 121. In this invention, it is sufficient to obtain the self-position, and the method of calculation and acquisition is not limited.

[0023] Step S204: The judgment criterion determination unit 122 acquires door open / closed information indicating the open / closed state of the door from another in-vehicle system (for example, a vehicle information control system). In this invention, it is sufficient to obtain door open / closed information, and the method of calculation and acquisition of such information is not limited.

[0024] Step S205: The judgment criterion determination unit 122 determines whether the judgment criteria should be changed based on the external environment information. If the route environment is not a station, level crossing, or railway bridge, the possibility of false detection is low, so it is determined that the judgment criteria do not need to be changed. Proceed to step S208. If the route environment is a station, level crossing, or railway bridge, the possibility of false detection is high, so it is determined that the judgment criteria need to be changed. Proceed to step S206. If the route shape environment is straight, the possibility of false detection is low, so it is determined that the judgment criteria do not need to be changed. Proceed to step S208. If the route shape environment is curved, the possibility of false detection is high, so it is determined that the judgment criteria need to be changed. Proceed to step S206.

[0025] Step S206: The judgment criterion determination unit 122 determines whether or not the train 100 is in motion based on its vehicle speed. If it is in motion, it determines that the judgment criterion needs to be changed because there is a high possibility of false detection. The process proceeds to step S209. If the train is stopped, the process proceeds to step S207.

[0026] Step S207: The judgment criterion determination unit 122 determines from the door opening / closing information whether the doors of train 100 are open or closed. If the doors are closed, it determines that train 100 is about to depart and changes the judgment criteria so that obstacles near train 100 can be detected at the time of departure. Proceed to step S210. If the doors are open, it determines that the train is stopped at a station. If the driver is notified every time an obstacle is detected in front of train 100 while the train is stopped at a station, the support display may become cumbersome. Therefore, the judgment criteria are changed so that obstacles are basically not detected while the train is stopped at a station. Proceed to step S211.

[0027] Step S208: This is a case where the external environment of train 100 is judged to be an environment in which false detections are unlikely to occur. Therefore, in order to reduce the number of undetected objects, a continuous detection threshold is set that detects objects as often as possible. For example, detection is defined as two consecutive detections. The threshold does not have to be "2", and it is desirable to evaluate using external sensor data acquired in the environment in which the obstacle detection support system 120 is introduced and set the threshold so that undetected and false detections fall within the desired range. Therefore, if the false detection falls within the desired range, the continuous detection threshold may be set to "1" (i.e., detection is defined as an object detection). The continuous detection threshold determined in this step is set as the first threshold. Proceed to step S212.

[0028] Step S209: This is a case where the external environment of train 100 is judged to be an environment where false detections are relatively likely to occur. Therefore, in order to reduce false detections, a continuous detection threshold is set that avoids detecting objects as much as possible. The threshold defined in this step is a larger value than the first threshold, for example, detection is defined if four consecutive detections occur. The threshold does not have to be "4", and it is desirable to perform an evaluation using external sensor data acquired in the environment in which the obstacle detection support system 120 is introduced and set the threshold so that false detections fall within the desired range. The continuous detection threshold determined in this step is set as the second threshold. Proceed to step S212.

[0029] Step S210: This is a case where the external environment of train 100 is judged to be stationary and just before departure. In this case, it is necessary to detect obstacles near train 100 at departure, but on the other hand, because it is stationary, the external environment is such that there is a possibility of repeatedly misdetecting the same object. Therefore, in order to make it less likely to misdetect objects than when the train is in motion, a value larger than the second threshold is defined in this step. However, as mentioned above, it is necessary to detect obstacles near train 100, so it is desirable that the difference from the second threshold be a small value, for example, detection is defined if it is detected 5 times in a row. The threshold does not have to be "5", and it is desirable to evaluate using external sensor data acquired in the environment in which the obstacle detection support system 120 is introduced and set the threshold so that the misdetection falls within the desired range. The continuous detection threshold determined in this step is set as the third threshold. Proceed to step S212.

[0030] Step S211: This is a case where the external environment of train 100 is judged to be that it is stopped at a station and its doors are open. In this case, if the driver is notified every time an obstacle is detected in front of train 100, the support display may become cumbersome. Also, since train 100 is stopped, there is no possibility of collision with an obstacle even if no obstacle is detected. Therefore, a large value is defined as the threshold in this step in order to basically not detect objects. For example, the threshold is set to a value greater than or equal to the maximum number of time-series data that the obstacle detection unit 123 can hold for obstacle detection. By doing so, objects in front of train 100 will not be detected, and the support display to the driver when the train is stopped at a station and its doors are open can be minimized. The continuous detection threshold determined in this step is set as the fourth threshold. Proceed to step S212.

[0031] Step S212: The judgment criterion determination unit 122 transmits the judgment criteria to the obstacle detection unit 123.

[0032] Next, with reference to Figure 3, the processing of the obstacle detection unit 123 according to Embodiment 1 of the present invention will be described.

[0033] Figure 3 is a functional block diagram showing the processes performed by the obstacle detection unit 123.

[0034] The sensor data acquisition unit 301 of the obstacle detection unit 123 acquires sensor data from external sensors 110 such as stereo cameras, monocular cameras, LiDAR, and GNSS. The external sensors 110 may be other external sensors 110, and any type of external sensor 110 capable of recognizing the external environment is acceptable. The sensor data acquisition unit 301 transmits image data to the high visibility processing unit 302, LiDAR data to the object detection unit (LIDAR) 306, and GNSS data to the vehicle speed calculation unit 307. The high visibility processing unit 302 adjusts the brightness of the image data and corrects scenes with differences in brightness, such as tunnel entrances and exits, to generate a high-visibility image. The generated high-visibility image is transmitted to the running area detection unit 303, the object detection unit (stereo) 304, and the object detection unit (machine learning) 305. The running area detection unit 303 detects rails through image processing and defines a running area where the train 100 may run based on the detected rail detection results. The method for defining the running area may be other than the rail detection result. For example, a database that associates rail shape and position may be maintained, and the rail shape in front of train 100 may be determined by comparing the train's own position with the position in the database, and the running area may be defined from the determined shape. Alternatively, if rail detection processing is performed by the external environment recognition unit 121, the rail detection result may be obtained from the external environment recognition unit 121. In this invention, it is sufficient that the running area can be defined, and the method is not limited. The object detection unit (stereo) 304 creates a disparity image using a stereo camera and recognizes the three-dimensional shape and position of objects in front from the disparity image. It transmits the detected result to the obstacle determination unit 308. The object detection unit (machine learning) 305 recognizes objects on the image from a monocular camera image using a DNN. A DNN is one of the means used in machine learning, and it can detect various objects by extracting and learning the features of objects. The object detection unit (machine learning) 305 transmits the detection result to the obstacle determination unit 308. The object detection unit (LIDAR) 306 detects objects from point cloud data acquired from the LIDAR sensor. Point clouds located at close range are treated as a single object, and the three-dimensional shape and position of objects around the train 100 are recognized. The object detection unit (LIDAR) 306 transmits the detection results to the obstacle determination unit 308.The vehicle speed calculation unit 307 calculates the vehicle speed from the position information (latitude and longitude) and elapsed time obtained from GNSS data and transmits it to the obstacle determination unit 308. The obstacle determination unit 308 determines the obstacles to be notified to the driver based on the driving area obtained from the driving area detection unit 303, the object detection results obtained from the object detection unit (stereo) 304, the object detection unit (machine learning) 305, and the object detection unit (LIDAR) 306, the vehicle speed obtained from the vehicle speed calculation unit 307, and the judgment criteria obtained from the judgment criteria determination unit 122.

[0035] Next, referring to Figure 4, an example of the obstacle determination process for each threshold executed by the obstacle determination unit 308 in Figure 3 will be explained. Figure 4 shows the time-series results (n, n-1, n-2...) of a certain detection result and the output for each threshold when the detection result is input. In the example in Figure 4, the first threshold is "2", the second threshold is "4", the third threshold is "5", and the fourth threshold is "6 or more". In Figure 4, a circle (○) in the input means that an object was detected, and a cross (×) means that no object was detected. In the output, a circle (○) means that the obstacle detection unit 123 has determined that an obstacle exists, and a cross (×) means that it has determined that there is no obstacle. The object detection results that are input here are the object detection results obtained from the object detection unit (stereo) 304, object detection unit (machine learning) 305, and object detection unit (LIDAR) 306 in Figure 3, and are the detection results of objects that are within the running area and within the stopping distance of the train 100 determined from the vehicle speed. If, in the time-series input, an object is detected consecutively only in the two most recent instances (first row of Figure 4), the judgment criterion determination unit 122 will determine that an obstacle exists if the judgment criterion is the first threshold, but will determine that there is no obstacle if it is anything other than the first threshold. Therefore, by changing the judgment criterion (threshold) of the judgment criterion determination unit 122, it is possible to output different obstacle detection results for the same detection result input. If the judgment criterion is the second threshold, it will be determined that there is no obstacle unless the most recent four instances of time-series input do not detect an object consecutively. Therefore, even if false detections occur consecutively, notifications to the driver are suppressed, and false alarms can be reduced without interrupting driving assistance. If the judgment criterion is the fourth threshold, it will be determined that there is no obstacle regardless of the input result, so it is possible to limit notifications to the driver when the train is stopped at a station and the doors are open. This makes it possible, for example, at a station where there is a level crossing for passengers to cross directly in front of train 100, to suppress the notification of warning support content to the driver by determining each pedestrian crossing as an obstacle.

[0036] In this embodiment, the external environment to be recognized includes the railway line environment such as stations, level crossings, and railway bridges, the railway line shape environment such as straight lines and curves, the time environment such as day and night, and the train environment such as vehicle speed and self-position. However, other environments prone to false detection may also be included as targets for recognition. For example, these include weather (rain, snow), backlighting, and passing oncoming trains at night. Such conditions are particularly challenging for image recognition and increase the likelihood of false detection. Weather information may be obtained from an external system (not shown), or the system may determine that it is raining if the wipers are operating. Backlighting and passing oncoming trains at night may also be determined from the average brightness value of the entire image, or by detecting areas with extremely high brightness values ​​in only a part of the image. In this invention, it is sufficient that the external environment recognition unit 121 recognizes external conditions (weather (rain, snow), backlighting, and passing oncoming trains at night), and the method of recognition is not limited.

[0037] In this embodiment, the judgment criteria were determined based on the external environment recognized by the external environment recognition unit 121. However, if the route environment, route shape environment, time environment, and train environment can be obtained from sources other than the external environment recognition unit 121, the route environment, route shape environment, time environment, and train environment may be obtained from sources other than the external environment recognition unit 121. For example, a database that associates the route environment and route shape environment with location information may be maintained, and the route environment and route shape environment may be obtained by comparing the train 100's own position with the position in the database. Also, since the time environment and train environment are generally held by other on-board systems (e.g., vehicle information control systems), the time environment and train environment may be obtained from other on-board systems. Doing so makes it possible to reduce the processing load on the external environment recognition unit 121. [Examples]

[0038] Figure 5 is a flowchart showing the processing performed by the judgment criterion determination unit 122 in the second embodiment of the obstacle detection support system 120. In the first embodiment, a continuous detection threshold was used as the judgment criterion, and the processing of the obstacle determination unit 308 was controlled by changing the continuous detection threshold to reduce false detections. In this embodiment, however, the detection method, which determines whether the detection result of an object is judged by logical OR or logical AND, is used as the judgment criterion, and an example is described in which the processing of the obstacle determination unit 308 is controlled by changing the detection method to reduce false detections.

[0039] Referring to Figure 5, the processing of the determination criteria unit 122 according to Embodiment 2 of the present invention will be described. Figure 5 is a flowchart showing the processing performed by the determination criteria unit 122.

[0040] Step S501: The judgment criterion determination unit 122 acquires external environment information (such as the route environment including stations, level crossings, and railway bridges; the route shape environment including straight lines and curves; the time environment including day and night; and the train environment including vehicle speed) from the external environment recognition unit 121.

[0041] Step S502: The judgment criterion determination unit 122 obtains the vehicle speed from another in-vehicle system (e.g., a vehicle information control system). The vehicle speed may be the vehicle speed estimated by the external environment recognition unit 121. In this invention, it is sufficient to obtain the vehicle speed, and the method of calculation and acquisition is not limited.

[0042] Step S503: The judgment criterion determination unit 122 acquires its own position from another in-vehicle system (for example, a vehicle information control system). The self-position may be the self-position estimated by the external environment recognition unit 121. In this invention, it is sufficient to obtain the self-position, and the method of calculation and acquisition is not limited.

[0043] Step S504: The judgment criterion determination unit 122 acquires door open / closed information indicating the open / closed state of the door from another in-vehicle system (for example, a vehicle information control system). In this invention, it is sufficient to obtain door open / closed information, and the method of calculation and acquisition of such information is not limited.

[0044] Step S505: The judgment criterion determination unit 122 determines whether the judgment criteria should be changed based on the external environment information. If the route environment is not a station, level crossing, or railway bridge, the possibility of false detection is low, so it is determined that the judgment criteria do not need to be changed. Proceed to step S507. If the route environment is a station, level crossing, or railway bridge, the possibility of false detection is high, so it is determined that the judgment criteria need to be changed. Proceed to step S506. If the route shape environment is straight, the possibility of false detection is low, so it is determined that the judgment criteria do not need to be changed. Proceed to step S507. If the route shape environment is curved, the possibility of false detection is high, so it is determined that the judgment criteria need to be changed. Proceed to step S506.

[0045] Step S506: The judgment criterion determination unit 122 determines whether the train 100 is in motion or whether its doors are closed based on the vehicle speed and door opening / closing information. If the train is in motion, it determines that the judgment criteria need to be changed because there is a high possibility of false detection. Also, if the train is not in motion but its doors are closed, it determines that the train 100 is about to depart and changes the judgment criteria so that obstacles near the train 100 can be detected at the time of departure. Proceed to step S508. If the train is stopped and its doors are open, it determines that the train is stopped at a station. If the driver is notified every time an obstacle is detected in front of the train 100 while it is stopped at a station, the support display may become cumbersome. Therefore, the judgment criteria are changed so that obstacles are not detected in principle while the train is stopped at a station. Proceed to step S509.

[0046] Step S507: This is a case where the external environment of train 100 is judged to be an environment where the possibility of false detection is low. Therefore, in order to reduce the number of undetected objects, the detection criteria are set to detect objects as much as possible. For example, when determining an obstacle using the time-series results of three obstacle detection results, the logical OR of the three time-series results is set as the output for obstacle detection. That is, if an object is detected in one or more of the three time-series results, it is considered detected. In this way, it is possible to reduce the number of undetected objects. The detection method is judged based on "logical OR". Proceed to step S510.

[0047] Step S508: This is a case where the external environment of train 100 is judged to be an environment where false detections are relatively likely to occur. Therefore, in order to reduce false detections, the judgment criteria are set to avoid detecting objects as much as possible. For example, when the presence of an obstacle is judged using the time-series results of three obstacle detection results, the logical AND of the three time-series results is used as the output for obstacle detection. That is, if an object is detected in all three time-series results, it is considered detected. By doing so, false detections of randomly occurring objects can be eliminated, and the number of false detections in the driver assistance system notified to the driver can be reduced. The judgment criterion for detection is "logical AND". Proceed to step S510.

[0048] Step S509: This is a case where the external environment of train 100 is judged to be that it is stopped at a station and its doors are open. In this case, if the driver is notified every time an obstacle is detected in front of train 100, the support display may become cumbersome. Also, since train 100 is stopped, there is no possibility of collision with an obstacle even if no obstacle is detected. Therefore, the judgment criterion is set so that no object is detected. For example, if the obstacle judgment is made using the time-series results of three obstacle detection results, the output will be set to "no obstacle" regardless of the three time-series results. By doing this, objects in front of train 100 will not be detected, and the support display to the driver when the train is stopped at a station and its doors are open can be minimized. The judgment criterion for detection is "no obstacle regardless of input result". Proceed to step S510.

[0049] Step S510: The judgment criterion determination unit 122 transmits the judgment criteria to the obstacle detection unit 123.

[0050] Next, referring to Figure 6, an example of the obstacle determination process for each judgment criterion executed by the obstacle determination unit 308 in Figure 3 in the second embodiment will be explained. Figure 6 shows the time-series results (n, n-1, n-2...) of a certain detection result and the output for each judgment criterion when that detection result is input. In the example in Figure 6, there are three types of obstacle detection methods as judgment criteria: "logical OR of inputs", "logical AND of inputs", and "no obstacle regardless of input". In Figure 6, a circle in the input means that a detected object was found, and a cross means that no object was found. In the output, a circle means that the obstacle detection unit 123 has determined that an obstacle is present, and a cross means that it has determined that there is no obstacle. The object detection results that are input here are the object detection results obtained from the object detection unit (stereo) 304, object detection unit (machine learning) 305, and object detection unit (LIDAR) 306 in Figure 3, and are the detection results of objects that are within the running area and within the stopping distance of the train 100 determined from the vehicle speed. If the judgment criterion is defined as "logical OR," an obstacle is determined to be present if it is detected in at least one of the three time-series results. This reduces the rate of non-detection. On the other hand, if the external environment is judged to be one in which false detections are likely to occur, the detection method of the judgment criterion is set to logical AND. If the judgment criterion is "logical AND," an obstacle is determined to be present if it is detected in all three time-series results. Therefore, even if false detections occur randomly, if they do not occur three times in a row, the notification to the driver will be suppressed, and false alarms can be reduced without interrupting driving assistance. If the judgment criterion is "no obstacle regardless of input," an obstacle will be determined to be absent regardless of the input result, making it possible to limit the notification to the driver when the train is stopped at a station and the doors are open. For example, at a station where there is a level crossing for passengers to cross directly in front of train 100, it is possible to suppress the notification of warning assistance to the driver by identifying each pedestrian crossing as an obstacle.

[0051] Examples 1 and 2 describe cases in which either continuous detection threshold information, which specifies how many consecutive detections are required in the time-series data of object detection results to constitute detection, or detection method information, which specifies whether to determine detection by logical OR or logical AND of the object detection result, is used as the judgment criterion. The judgment criterion determined by the judgment criterion determination unit 122 may be other than the continuous detection threshold information or the detection method information, for example, both the continuous detection threshold information and the detection method information may be used as the judgment criterion. By setting the continuous detection threshold to a large value and further using logical AND as the detection method, it is possible to make it more difficult to detect objects. In this way, it is possible to further reduce false detections.

[0052] According to the embodiments described above, the obstacle detection support system 120 installed in the rail transport system can reduce false alarms without interrupting driver assistance to the driver.

[0053] The embodiments of the present invention described above can be summarized as follows.

[0054] (1) The obstacle detection support system 120 is a system that supports obstacle detection during vehicle driving and includes a driving support display unit 124 that displays driving support information which is information that supports vehicle driving, an obstacle detection unit 123 that detects obstacles that hinder the driving of the vehicle, a judgment criterion determination unit 122 that determines the judgment criteria for when the obstacle detection unit 123 detects an obstacle, and an external environment recognition unit 121 that recognizes the external environment of the vehicle. The judgment criterion determination unit 122 sets judgment criteria according to the external environment, and the judgment criteria are set to a threshold number of times the obstacle detection unit 123 detects an obstacle, or the degree of agreement of the detection results by the obstacle detection unit 123, or both the threshold number of times the obstacle detection detects and the degree of agreement of the detection results. In this way, the obstacle detection support system 120 can reduce false alarms without interrupting driving support to the driver.

[0055] (2) The judgment criterion determination unit 122 includes one or more of the following as external environment when changing the judgment criterion: route environment such as station areas, level crossings, and railway bridges; route shape environment such as straight lines and curves; time environment such as day and night; and train environment such as vehicle speed and self-position.

[0056] (3) When the judgment criterion determination unit 122 determines that the external environment is one in which false detections are likely to occur, it changes the threshold value for the number of obstacle detections to a larger value.

[0057] (4) When the judgment criterion determination unit 122 determines that the external environment is an environment in which false detections are likely to occur, it changes the method for detecting the degree of agreement of the detection results from logical OR to logical AND.

[0058] (5) When the judgment criterion determination unit 122 determines that the external environment is such that the vehicle is stopped and the doors are open, it sets the threshold for the number of obstacle detections to be greater than or equal to the maximum value of past detection result data that the system can retain.

[0059] (6) When the determination criteria unit 122 determines that the external environment is such that the vehicle is stopped and the doors are closed, it sets the threshold for the number of obstacle detections to less than the maximum value of past detection result data that the system can retain.

[0060] (7) The obstacle detection unit 123 determines that only detected objects that are within the driving area and within a distance from which the vehicle can stop are to be detected as obstacles.

[0061] Although embodiments of the present invention have been described above, the present invention is not limited to the embodiments described above, and various modifications are possible without departing from the spirit of the present invention.

[0062] In the diagrams above, the control lines and information lines shown are those deemed necessary for explanation and do not necessarily represent all control lines and information lines in actual implementation. For example, it can be assumed that almost all components are interconnected in practice.

[0063] Furthermore, the arrangement of each functional component of the obstacle detection support system 120 described above is merely an example. The arrangement of each functional component can be changed to the optimal arrangement from the perspective of the performance, processing efficiency, and communication efficiency of the hardware and software provided by the obstacle detection support system 120. [Explanation of symbols]

[0064] 100 trains 110 External Sensors 120 Obstacle Detection Support System 121 External environment recognition department 122 Judgment criteria determination section 123 Obstacle detection unit 124 Driver support display unit

Claims

1. An obstacle detection support system that assists in the detection of obstacles during vehicle driving, comprising: a driving support display unit that displays driving support information which is information that assists in vehicle driving; an obstacle detection unit that detects obstacles that obstruct the driving of the vehicle; a judgment criterion determination unit that determines the judgment criteria for when the obstacle detection unit detects an obstacle; and an external environment recognition unit that recognizes the external environment of the vehicle, wherein the judgment criterion determination unit sets judgment criteria according to the external environment, and the judgment criteria are set to be either a threshold for the number of times the obstacle detection unit detects an obstacle or the degree of agreement of the detection results by the obstacle detection unit, or both the threshold for the number of times the obstacle detection unit detects an obstacle and the degree of agreement of the detection results.

2. An obstacle detection support system according to claim 1, wherein the determination criterion unit includes one or more of the following as external environments when changing the determination criterion: a route environment such as a station, level crossing, or railway bridge; a route shape environment such as a straight line or a curve; a time environment such as day or night; and a train environment such as vehicle speed or self-position.

3. An obstacle detection support system according to claim 1, characterized in that when the judgment criterion determination unit determines that the external environment is an environment in which false detections are likely to occur, it changes the threshold value for the number of obstacle detections to a larger value.

4. An obstacle detection support system according to claim 1, characterized in that when the judgment criterion determination unit determines that the external environment is an environment in which false detections are likely to occur, the method for detecting the degree of agreement of the detection result is changed from logical OR to logical AND.

5. An obstacle detection support system according to claim 1, characterized in that when the determination criterion determination unit determines that the external environment is such that the vehicle is stopped and the doors are open, it sets the threshold for the number of obstacle detections to be greater than or equal to the maximum value of past detection result data that the system can retain.

6. An obstacle detection support system according to claim 1, characterized in that when the determination criterion determination unit determines that the external environment is such that the vehicle is stopped and the doors are closed, it sets the threshold for the number of obstacle detections to less than the maximum value of past detection result data that the system can retain.

7. An obstacle detection support system according to claim 1, characterized in that the obstacle detection unit determines only detection objects that are located within the driving area and within a distance at which the vehicle can stop as obstacle detection targets.

8. An obstacle detection support method for assisting obstacle detection during vehicle driving, characterized in that a computer comprising at least a processor and a memory device displays driving support information which is information that assists vehicle driving, detects obstacles that hinder the driving of the vehicle, determines criteria for determining the obstacle, recognizes the external environment of the vehicle, sets criteria according to the external environment when determining the criteria, and sets a threshold for the number of obstacle detections when detecting the obstacle, or the degree of agreement of the detection results when detecting the obstacle, or both the threshold for the number of obstacle detections and the degree of agreement of the detection results as the criteria.

Images