JP7256699B2 - train control system - Google Patents

Description

TECHNICAL FIELD The present invention relates to a train control system, and is suitable for application to, for example, a train control system that controls a train running on a track.

In a rail transportation system, if there is an object around the track, it cannot be evaded by steering, so object detection is important to improve train safety and operability.

In railways, which is one of rail transportation systems, a device for detecting obstacles has been proposed as a technique for detecting objects around a track (see Patent Literature 1). The obstacle detection device described in Patent Document 1 emits radio waves from a radar mounted on a train, measures the reflected waves, and confirms the difference from a prerecorded reference signal to detect the obstacle on the track and the Detect objects around the orbit.

JP 2015-34793 A

In railways, the general public is prohibited from entering the vicinity of the tracks, but workers frequently work in the vicinity of the tracks during maintenance work, maintenance inspections, and the like. However, with the technology described in Patent Literature 1, the train may determine that a worker who is working normally is an obstacle. For example, when a train determines that a worker who is working normally is an obstacle, the train brakes and decelerates even in situations where deceleration is unnecessary.

SUMMARY OF THE INVENTION The present invention has been made in consideration of the above points, and aims to propose a train control system or the like that can more appropriately recognize objects and control trains.

In order to solve such problems, the present invention provides a train control system for controlling a train running on a track, comprising: a storage unit for storing control information relating to stop control of the train for each attribute of an object; a detection unit that detects an object from sensor information obtained by a sensor that acquires information related to a trajectory; a specification unit that specifies an attribute of the object detected by the detection unit from the sensor information obtained by the sensor; a control unit for controlling the train according to control information stored in a storage unit and corresponding to the attribute of the object identified by the identification unit.

According to the above configuration, the train can be controlled to stop according to the attribute of the object, so that necessary braking can be applied while unnecessary braking can be reduced. For example, according to the above configuration, energy consumption can be reduced by reducing unnecessary brakes. Further, for example, according to the above configuration, by reducing unnecessary brakes, it is possible to suppress the disturbance of the timetable, so that the train can be operated stably. Further, for example, according to the above configuration, when the worker is in the area where the worker is permitted to enter, the vehicle travels without applying the brake or applying the brake with a low deceleration. If there are ordinary people in the area, it is possible to control the speed of the train, such as applying brakes for high deceleration and running, so that safety can be ensured.

According to the present invention, a highly reliable train control system can be realized.

It is a figure showing an example of composition concerning a train control system by a 1st embodiment. It is a figure which shows an example of the train control rule by 1st Embodiment. It is a figure which shows an example of the flowchart regarding a train control process by 1st Embodiment. FIG. 7 is a diagram showing an example of a flowchart relating to object attribute identification processing according to the first embodiment; It is a figure which shows an example of a structure concerning the train control system by 2nd Embodiment.

One embodiment of the present invention will be described in detail below with reference to the drawings. This embodiment relates to a technology that enables safe and stable train control based on the attributes of objects existing around the track. For example, the train control system according to the present embodiment records in advance the attribute of an object that has the possibility of intruding into the monitored area of the route, and the control information that defines the control to be performed when the object with the attribute is detected. Equipped with a database. In such a train control system, when an object is detected in the monitoring area provided in front of the train, the attribute of the object is specified, and the train is controlled based on the attribute of the specified object based on the database. characterized by performing According to such a train control system, for example, it is possible to avoid erroneous detection of a worker working at an appropriate location on the railway line, so that it is possible to appropriately control the train.

In the following description, "storage device" includes one or more memories. At least one memory in the storage device may be a volatile memory or a non-volatile memory. The storage device is RAM (Random Access Memory), ROM (Read Only Memory), HDD (Hard Disk Drive), or the like.

Also, in the following description, a "controller" is one or more processors. The at least one processor is typically a microprocessor such as a CPU (Central Processing Unit), but may be another type of processor such as a GPU (Graphics Processing Unit). At least one processor may be a broadly defined processor such as a hardware circuit (for example, FPGA (Field-Programmable Gate Array) or ASIC (Application Specific Integrated Circuit)) that performs part or all of the processing.

Also, in the following description, the function may be described using the expression “kkk unit”, but the function may be realized by executing one or more computer programs by the control device, or may be realized by executing one or more computer programs. may be realized by a hardware circuit of The description of each function is an example. A plurality of functions may be combined into a single function, or a single function may be divided into a plurality of functions.

Also, in the following description, a "train" is composed of one or more train cars (hereinafter referred to as "vehicles").

(1) First Embodiment In FIG. 1, 100 indicates a train control system according to the first embodiment as a whole.

FIG. 1 is a diagram showing an example of the configuration of a train control system 100. As shown in FIG.

A train control system 100 includes a train 110 . Train 110 runs along a fixed track. The train 110 is a vehicle that transports passengers, freight, and the like. The train 110 is provided with a braking device that reduces the speed (running speed) of the train 110 based on instructions from a control unit 117, which will be described later.

Also, the train 110 includes a control device 111 , a storage device 112 and a sensor 113 . The functions of the train 110 (the detection unit 114, the identification unit 115, the calculation unit 116, the control unit 117, etc.) may be realized by, for example, the control device 111 reading a program to the storage device 112 and executing it (software). However, it may be implemented by hardware such as a dedicated circuit, or may be implemented by combining software and hardware. Also, some of the functions of train 110 may be implemented by other computers that can communicate with train 110 .

The sensor 113 is mounted in front of the train 110 and acquires information (sensor information) obtained by sensing the front of the train 110 . As the sensor 113, various means such as a camera, millimeter wave radar, LIDAR (Light Detection and Ranging), ultrasonic sensor, beacon radio wave receiver, etc. can be employed. Moreover, as the sensor 113, a plurality of sensors 113 may be used in combination.

The detection unit 114 monitors the area around the track in front of the train 110 using the sensor 113, and determines (detects) the presence or absence of an object around the track.

When the detection unit 114 determines that an object exists around the trajectory, the detection unit 114 transmits the sensor information of the sensor 113 obtained from the object to the identification unit 115 and the calculation unit 116 . The sensor information differs depending on the type of the sensor 113 to be used. For example, when a camera is used, it becomes image information of the object. Become. In addition, the worker holds a beacon that periodically emits radio waves, and when the detection unit 114 receives radio waves from the beacon via the beacon radio wave receiver, it determines that the worker is present in the surrounding area. You may Image recognition technology, laser detection technology, or the like may be used as a means for detecting an object from sensor information, and the method does not matter in this embodiment.

Further, the detection unit 114 acquires track information such as the radius of curvature and gradient of the track corresponding to the current position of the train 110 estimated by the self-position estimation means from a track information database (not shown). A surveillance area for detecting objects may be set around the forward orbit. Computation of the current position of train 110 is typically based on an integral of the speed of train 110, although other methods may be used. For example, a global positioning system (GPS) or the like may be used. In the present embodiment, any method is acceptable as long as the current position of train 110 can be uniquely determined. As another means of setting the monitoring area, the track shape in front of the train 110 may be estimated from the image information, and the monitoring area may be derived based on the track shape. For example, it is conceivable to set a certain range from the construction gauge set for the route on which the train 110 runs as the monitoring area.

The identifying unit 115 evaluates the degree of matching between the feature amount of the sensor information received from the detecting unit 114 and the feature amount of the attribute type information storage unit 118, and identifies the attribute of the object detected by the detecting unit 114. As a means for evaluating the degree of conformity, for example, a predetermined feature amount (for example, a threshold for saturation, reflection intensity, etc.) may be compared with the feature amount of the sensor information received from the detection unit 114. , image recognition, and machine learning generally used for data classification may be applied. The specifying unit 115 transmits attribute information indicating attributes of the specified object to the control unit 117 .

Here, the attribute type information storage unit 118 is a database in which feature amounts of sensor information are registered in advance for each attribute of an object that may exist around the trajectory. As an example, a case of registering a worker will be described. For example, when image information is used as sensor information, information such as the shape of clothes worn by the worker, the color of the clothes, and the predetermined gesture of the worker when the train 110 approaches (raising hand, raising flag, etc.) is registered. Moreover, when using the reflection intensity of a millimeter wave radar, LIDAR, etc. as sensor information, the reflectance of the clothes which a worker wears is registered. For example, the worker may wear a reflective material having a high-luminance reflectance, and the reflectance may be registered in the attribute type information storage unit 118 . By wearing reflective material that is significantly stronger than general clothing, it is expected to improve the accuracy of distinguishing between ordinary people and workers.

For the sake of simplification, the example of registering workers has been described above, but similar registration processing is performed when registering lineside facilities such as electrified poles and tool boxes, oncoming trains, maintenance vehicles, and the like.

Further, the identifying unit 115 may identify an attribute of an object from one piece of sensor information, or from a plurality of pieces of sensor information. For example, when image information cannot be sufficiently obtained by a camera at night or in bad weather, the reflection intensity measured by the millimeter wave radar matches the reflection intensity of the worker registered in the attribute type information storage unit 118. In this case, the attribute of the object detected by the detection unit 114 may be identified as the worker. Further, for example, when it is determined that an object is a worker based on the image information but is determined not to be a worker based on the reflection intensity, the attribute of the object may be specified as not being a worker. Even if an ordinary person wearing clothing similar to that of a worker is detected, it can be identified with high accuracy by identifying it based on a combination of conditions including reflection intensity. In this way, even in an environment where it is difficult to acquire sensor information by a specific sensor 113 and recognize the sensor information, it is possible to specify the attribute of an object with high reliability.

In addition, when one object exhibits a certain degree of matching with the feature values of multiple types (for example, when the object matches both workers and general people), the identifying unit 115 determines that the matching is higher. You may select an attribute, or select an attribute that provides safer control. It is also possible to select a "general person" who slows down from "workers").

The calculation unit 116 calculates a point where the object detected by the detection unit 114 exists (hereinafter referred to as object point information) from the sensor information received from the detection unit 114 . Calculation section 116 transmits the object point information to control section 117 .

The object point information is information including distance information indicating the distance from the center of the trajectory. As a method of calculating the object point information, for example, the distance from the head of the train 110 and the distance between the train 110 and the object calculated based on the time and angle at which the emitted radio wave or emitted light from millimeter wave radar, LIDAR, or the like hits the object and is reflected. The distance from the center of the trajectory may be calculated from the angle, or the distance from the center of the trajectory may be calculated using the parallax image of the stereo camera. . Further, the object point information may include distance information indicating the distance from the head of the train 110 to the object.

Note that the object point information may be current position information indicating the current position calculated from GPS or the like. Also, the object point information may be coordinate information on a map calculated from the current position information and the map information.

The control unit 117 selects a corresponding train control rule from the control information storage unit 119 based on the attribute information received from the identification unit 115 and the object point information received from the calculation unit 116 . The controller 117 controls the speed of the train 110 according to the selected train control rule. Examples of specific devices for the braking/driving means of the train 110 include inverters, motors, and friction brakes.

Here, the control information storage unit 119 is a database in which train control rules (an example of control information) corresponding to attribute information of objects detected in front of the train 110 and object point information are registered in advance. An example of the train control rule will be described later with reference to FIG.

For example, when the control unit 117 selects a train control rule for decelerating the train 110, the timing to start deceleration control and the applied deceleration are the current speed of the train 110 and may be determined according to the distance of For example, when an object exists far away, the control unit 117 may gradually decelerate with a low deceleration brake, or may perform control so that deceleration does not start until the object approaches.

In addition, the detection unit 114 may detect the presence or absence of an object using not only newly acquired sensor information but also past sensor information. Further, the identifying unit 115 may identify the attribute of an object using not only newly acquired sensor information but also past sensor information. Further, the calculation unit 116 may calculate the object point information using not only newly acquired sensor information but also past sensor information. Each of the detection unit 114, the identification unit 115, and the calculation unit 116 may utilize a technique such as a Kalman filter, and may also use past sensor information to estimate the motion state of the object. For example, if the detection unit 114 has detected an object moving linearly from past sensor information, even if sensor information cannot be acquired at a certain moment, the object is located at a position predicted from the past sensor information. may be assumed to exist and the subsequent processing may be executed.

Further, the detection unit 114 may calculate accuracy information regarding the presence or absence of the detected object and transmit the information to the control unit 117 . The identifying unit 115 may calculate accuracy information for the attribute of the identified object and transmit the information to the control unit 117 . The calculation unit 116 may calculate accuracy information for the calculated object point information and transmit it to the control unit 117 . As the accuracy information, the fitness calculated by the identifying unit 115 may be used, or the probability distribution information of the data acquired by each sensor 113 may be used. For example, when the sensor 113 is a camera, it exhibits a probability distribution with a small variance of reliability during the daytime and a probability distribution with a large variance of reliability during the nighttime. When the presence/absence of an object is detected by the sensor, the accuracy information regarding the presence/absence of an object detected during the daytime is transmitted to the control unit 117 as the first value, and the accuracy information regarding the presence/absence of the object detected during the nighttime is higher than the first value. It is transmitted to the control unit 117 as a low second value.

If the accuracy information is lower than a predetermined value, the control section 117 may adopt a safer train control rule to control the train 110 . For example, even if the attribute of the object is identified as a worker doing a predetermined gesture, if the accuracy of the result is lower than a predetermined value, the train An operation such as controlling the speed of 110 is assumed.

In addition, as a method for controlling the speed of the train 110, the control unit 117 may, for example, transmit the generated braking/driving command to an ATO (Automatic Train Operation) device (automatic train operation device) for automatic control, An alert may be presented to the driver to prompt manual brake control. Whether to apply automatic control or manual control may be determined before the train 110 runs, or may be manually determined by the driver after the train 110 starts running. Further, when receiving the accuracy information, the control unit 117 may determine which of the automatic control and the manual control should be applied based on the accuracy information. For example, normally, the train 110 is controlled by automatic control, but if the accuracy information is lower than a predetermined value, it is determined that there is a possibility that the judgment by the system is erroneous, and the system manually controls the driver. It is conceivable to operate such as encouraging switching to

FIG. 2 is a diagram showing an example of a train control rule. In this example, train control rules are defined according to the attributes of the object and the distance from the center of the track.

The control information storage unit 119 stores, for example, a train control rule for decelerating the train 110 when a person is present within "2 m" from the center of the track, regardless of the attribute of the person. Also, for example, when a person exists within "2m" to "5m" from the center of the track, if the attribute of the person is a worker and the worker is making a predetermined gesture, the train If the person's attribute is a worker and the worker is not making a predetermined gesture, the train 110 is decelerated, and if the person's attribute is an ordinary person, A train control rule for slowing the train 110 is stored. Also, for example, a train control rule is stored that, when a person is more than "5 m" away from the center of the track, the train 110 continues to run regardless of the attribute of the person.

Also, in FIG. 2, a person is taken as an example, but the same applies to objects. Although illustration is omitted, in the control information storage unit 119, for example, if an object exists within “5 m” from the center of the track, and if the attribute of the object is equipment for work, the train 110 will not run. Continuing, a train control rule is stored that decelerates the train 110 when the attribute of the object is an object other than equipment for work. Further, for example, a train control rule is stored that, when an object is more than "5 m" away from the center of the track, the train 110 continues running regardless of the attributes of the object.

In this example, for the sake of simplicity, the deceleration is shown in two stages, "continue running" and "deceleration". deceleration may be defined by

In addition, the control information storage unit 119 may store train control rules including information on inspection work, maintenance work, and construction work (hereinafter referred to as work information) on the route as conditions. The work information includes information on one or more of a place (for example, section, work position, etc.) and time (for example, time zone, period, etc.) where the worker works. As an example of a train control rule using work information, for example, in a section and time zone where inspection work is being performed, if a worker is detected, the speed of the train 110 is adjusted according to the train control rule shown in FIG. However, if a worker is detected outside the section, deceleration control may be performed as an unplanned intrusion.

The above is the description of the main configuration and each component of the train control system 100 . Next, processing related to the train control system 100 will be described.

FIG. 3 is a diagram showing an example of a flowchart relating to train control processing performed by the train 110. As shown in FIG. The train control process is executed at regular intervals. Steps S<b>300 to S<b>303 are executed by the detection unit 114 . Step S<b>304 is executed by the identification unit 115 . Step S<b>305 is executed by the calculation unit 116 . Steps S<b>306 and S<b>307 are executed by control unit 117 .

In step S<b>300 , the detection unit 114 acquires current sensor information from the sensor 113 .

In step S301, the detection unit 114 sets a monitoring area at the current position of the train 110 (location on the train) based on sensor information.

In step S302, the detection unit 114 determines whether an object exists in the monitoring area based on the sensor information (object presence/absence determination).

In step S303, the detection unit 114 determines the next process to be executed based on the result of determining whether or not an object exists within the monitoring area in step S302. If the detection unit 114 determines that an object exists in front of the train 110, the process proceeds to step S304, and if it determines that there is no object in front of the train 110, the train control process ends.

In step S304, the identification unit 115 performs object attribute identification processing. The specifying unit 115 evaluates the degree of matching between the feature amount of the sensor information for the object (object to be controlled) detected by the detection unit 114 and each feature amount stored in the attribute type information storage unit 118 . The identifying unit 115 identifies the attribute of the object to be controlled based on the evaluated degree of compatibility, and transmits attribute information indicating the attribute of the object to the control unit 117 . Note that an example of object attribute identification processing will be described later using FIG.

In step S<b>305 , the calculation unit 116 calculates object point information of the object based on the sensor information of the object to be controlled, and transmits the calculated object point information to the control unit 117 .

Although FIG. 3 shows an example in which step S305 is performed after step S304, these steps may be processed independently. For example, step S305 may be processed before step S304, or step S304 and step S305 may be processed in parallel.

In step S306, the control unit 117 selects from the control information storage unit 119 a train control rule corresponding to the attribute information of the object to be controlled and the object point information.

At step S307, the controller 117 controls the speed of the train 110 based on the selected train control rule.

FIG. 4 is a diagram showing an example of a flowchart relating to object attribute specifying processing performed by the specifying unit 115. As shown in FIG. In the object attribute identification process, a case will be described in which a plurality of sensors 113 are used to identify the attributes of the object to be controlled. As the plurality of sensors 113, sensor A (for example, camera) and sensor B (for example, millimeter wave information) will be described as an example.

In step S<b>400 , the identification unit 115 acquires a feature amount (threshold value) for each attribute of the object related to the sensor A from the attribute type information storage unit 118 .

In step S401, the identification unit 115 calculates the goodness of fit of the sensor A. FIG. For example, the identifying unit 115 uses the feature amount (eg, shape information) of sensor information of sensor A and the acquired feature amount (eg, threshold value of shape information) related to sensor A, for each attribute, A degree of similarity between the shape information and the threshold of the shape information is calculated. The identifying unit 115 normalizes the highest degree of similarity among the calculated degrees of similarity to obtain the degree of suitability of the sensor A. FIG.

In step S<b>402 , the identification unit 115 acquires a feature amount (threshold value) for each attribute of the object related to sensor B from the attribute type information storage unit 118 .

In step S<b>403 , the identification unit 115 calculates the goodness of fit of the sensor B. FIG. For example, the specifying unit 115 uses the feature amount (for example, reflectance) of sensor information of sensor B and the acquired feature amount (reflectance threshold value) related to sensor B to determine the reflectance for each attribute. and the reflectance threshold. The specifying unit 115 normalizes the smallest difference among the calculated differences and uses it as the goodness of fit of the sensor B. FIG.

In step S404, the identifying unit 115 identifies attributes of the object to be controlled. For example, the identification unit 115 compares the degree of suitability of the sensor A and the degree of suitability of the sensor B, refers to the attribute type information storage unit 118, and uses the feature amount of the sensor information with the higher degree of suitability to determine the control target identify attributes of objects in For example, when the attribute of the object to be controlled determined from the sensor information of sensor A and the attribute of the object to be controlled determined from the sensor information of sensor B are different, the identifying unit 115 determines that the deceleration of the train 110 is The attribute with the higher stopping control may be specified as the attribute of the object to be controlled.

In step S405, the identification unit 115 determines whether the attribute of the object to be controlled is a person. If the identification unit 115 determines that the person is a person, the process proceeds to step S407, and if it determines that the person is not a person, the process proceeds to step S406.

In step S406, the specifying unit 115 outputs the attribute of the object to be controlled to the control unit 117 as "object".

In step S407, the identification unit 115 determines whether or not the attribute of the object to be controlled is worker. If the identification unit 115 determines that the person is a worker, the process proceeds to step S409, and if it determines that the person is not a worker, the process proceeds to step S408.

In step S408, the specifying unit 115 outputs the attribute of the object to be controlled to the control unit 117 as "general person".

In step S409, the identifying unit 115 determines whether or not the attribute of the object to be controlled is that of a worker raising his hand. If the identifying unit 115 determines that the worker is raising his hand, the process proceeds to step S411. If the identifying unit 115 determines that the worker is not raising his hand, the process proceeds to step S410.

In step S410, the specifying unit 115 outputs the attribute of the object to be controlled to the control unit 117 as "worker (no raised hand)".

In step S<b>411 , the identifying unit 115 outputs the attribute of the object to be controlled to the control unit 117 as “worker (hand raised)”.

The above is the description of the train control process executed by the train control system 100 .

As described above, the train control system according to the present embodiment stores in advance attribute information indicating the attributes of objects that may enter the monitored area of the railway line, and train control rules for when an object with the attributes is detected. database. Such a train control system is characterized in that when an object is detected within a monitored area, the attribute of the object is specified based on the database, and the train is controlled based on the train control rule corresponding to the attribute. By controlling the train based on the attributes of the object, for example, it is possible to avoid disruption of operation due to unnecessary deceleration, warnings, etc. for workers performing normal construction work, inspection work, and the like. Also, for example, when an ordinary person, an unplanned worker, or the like enters the vicinity of the track, it is possible to perform control to avoid a collision by decelerating the vehicle.

For the sake of simplicity, the example of registering workers has been mainly described, but as described above, the same applies to registering objects other than people, such as electrified poles, equipment along railway lines such as tool boxes, oncoming trains, maintenance vehicles, etc. effect is obtained.

(2) Second Embodiment In the train control system 100 of the first embodiment, an object is detected using sensor information acquired by the sensor 113 mounted on the train 110 (on the train). The train control system 500 of the second embodiment is characterized in that an object is detected using sensor information acquired by the sensor 513 installed along the railway. In this embodiment, a configuration different from that of the first embodiment is mainly described.

FIG. 5 is a diagram showing an example of the configuration of a train control system 500. As shown in FIG.

Train control system 500 includes ground equipment 510 . The ground equipment 510 comprises a control device 511 , a storage device 512 , a sensor 513 and a communication device 514 .

The sensor 513 acquires sensor information by sensing the surroundings of the installed location. The sensor 513 may be configured using existing equipment such as a device for detecting an obstacle at a railroad crossing, a monitoring camera provided in the premises of a station, or the like. Further, for example, as for the sensor 513, a new sensor 513 (for example, camera, millimeter wave radar, LIDAR, ultrasonic sensor, etc.) may be installed along the railway.

The communication device 514 is an interface device for performing wireless communication with the communication device 521 provided on the train 110 .

The functions of the ground equipment 510 (for example, the ground monitoring unit 515, the information transmission unit 516, etc.) may be realized by, for example, the control device 511 reading a program to the storage device 512 and executing it (software), or by using a dedicated program. It may be realized by hardware such as a circuit of , or may be realized by combining software and hardware. Also, some of the functions of ground equipment 510 may be implemented by other computers capable of communicating with ground equipment 510 .

Ground monitoring unit 515 transmits sensor information acquired by sensor 513 to train 110 via information transmitting unit 516 . The information transmission unit 516 controls the communication device 514 and transmits sensor information to the information reception unit 522 that controls the communication device 521 provided on the train 110 .

Note that the processing in the train control system 500 is basically the same as in the first embodiment, so the description thereof will be omitted.

Although the present embodiment is characterized in that the sensor 513 is provided along the railroad, the sensor information from the sensor 113 provided on the vehicle may be used together, or the sensor 113 may not be provided on the vehicle. .

Further, the detection unit 114, the identification unit 115, the calculation unit 116, the control unit 117, the attribute type information storage unit 118 and the control information storage unit 119 are configured to be mounted on the train 110, but some or all of these functions are It may be installed in ground equipment (eg, ground equipment 510, server equipment, etc.). For example, when all the functions are installed in ground equipment, after the ground equipment acquires sensor information and detects an object, the train 110 receives a control instruction for the train 110 from the ground equipment, and based on the control instruction An operation of controlling the train 110 by using the

As described above, according to the present embodiment, by utilizing the sensors installed on the ground, it is possible to detect objects even in areas beyond the line of sight from the train, such as distant locations and curved sections, thus improving the safety of train operations. can contribute to

(3) Other Embodiments In the above-described embodiment, the case where the present invention is applied to a train control system has been described, but the present invention is not limited to this, and various other systems, It can be widely applied to devices, methods, programs, recording media, and the like.

In the above-described embodiment, the calculation unit 116 calculates the distance from the center of the trajectory as the distance from the trajectory, but the present invention is not limited to this. As the distance from, the distance from the construction gauge may be calculated.

Also, in the above-described embodiments, the configuration of each table is an example, and one table may be divided into two or more tables, or all or part of the two or more tables may be one table. may

In the above description, information such as programs, tables, and files that implement each function is stored in a memory, hard disk, SSD (Solid State Drive), or other storage device, or recorded on an IC card, SD card, DVD, or the like. You can put it on the medium.

The embodiments described above have, for example, the following characteristic configurations.

A train control system (e.g., train control system 100, train control system 500) that controls a train (e.g., train 110) that travels on a track determines attributes of an object (e.g., person, object, etc.) (e.g., person, General people, workers, workers who recognize the train 110, workers who do not recognize the train 110, objects, facilities along the railway line, oncoming trains, maintenance vehicles, equipment related to work, equipment not related to work, etc.) A storage unit (e.g., storage device 112, control information storage unit 119) that stores control information (e.g., train control rules) related to the stop control of the train, and information related to the track (e.g., image information, information from an object sensor information obtained by a sensor (e.g., sensor 113, sensor 513, camera, millimeter wave radar, LIDAR, ultrasonic sensor, beacon radio wave receiver, etc.) that acquires reflection intensity and distance information, radio waves from beacons, etc. a detection unit (e.g., detection unit 114) that detects an object from the The train is controlled according to the control information stored in the storage unit and corresponding to the attribute of the object specified by the specifying unit (for example, brake control according to the speed of the train 110, control according to the distance between the train 110 and the object). A control unit (for example, control unit 117) that transmits a corresponding brake control and drive command to the ATO device, presents an alarm to the driver, and performs manual brake control, etc.

According to the above configuration, the train can be controlled to stop according to the attribute of the object, so it is possible to apply the necessary brakes and reduce the unnecessary brakes. For example, according to the above configuration, energy consumption can be reduced by reducing unnecessary brakes. Further, for example, according to the above configuration, by reducing unnecessary brakes, it is possible to suppress timetable disturbances, so that trains can be operated stably. Further, for example, according to the above configuration, when the worker is in the area where the worker is permitted to enter, the vehicle travels without applying the brake or applying the brake with a low deceleration. If there are ordinary people in the area, it is possible to control the speed of the train, such as applying brakes with high deceleration and running, so that safety can be ensured.

The storage unit stores control information related to stop control of the train for each attribute of the object and for each distance from the track (for example, see FIG. 2), and sensor information obtained by the sensor and The distance between the object detected by the detection unit and the track (for example, track distance from the center, distance from the construction gauge, etc.), and the control unit is stored in the storage unit, and the object specified by the specifying unit and the control information corresponding to the distance calculated by the calculation unit, the train is controlled.

According to the above configuration, the train can be controlled to stop according to the attribute of the object and the distance of the object from the track. For example, if the object is located far from the track (for example, if the object is more than a predetermined value away from the center of the track), the brakes with low deceleration are applied and the object is positioned close to the track. (For example, when the object is located within a predetermined value range from the center of the trajectory), it becomes possible to apply the brakes more appropriately, such as driving with high deceleration braking. Therefore, according to the above configuration, it is possible to further reduce energy consumption, to operate trains more stably, and the like.

A second storage unit (for example, storage device 112, The attribute type information storage unit 118) is provided, and the specifying unit refers to the second storage unit and uses the feature amount of sensor information acquired by the sensor to determine the attribute of the object detected by the detection unit. Specify (for example, calculate the degree of matching between the feature amount stored in the second storage unit and the feature amount of the sensor information, and specify the attribute of the feature amount with the highest degree of matching as the attribute of the object).

According to the above configuration, the attribute of an object can be identified from, for example, the color, shape, gesture, reflectance, or any combination of these of the object. For example, by using multiple types of feature amounts, it is possible to more accurately identify the attributes of an object.

The storage unit stores control information related to a first stop control corresponding to the attribute of a worker who does not recognize the train, and stores control information related to the first stop control corresponding to the attribute of a worker who recognizes the train. Stores control information related to a second stop control in which the deceleration is relatively lower than that of the first stop control (see, for example, FIG. 2), and the identifying unit recognizes that the train is approaching. When it is determined that the object is making a gesture indicating that the According to the control information stored in the storage unit and corresponding to the attribute of the worker who recognizes the train specified by the specifying unit (for example, a train control rule corresponding to raised hand), the train is selected. Control.

According to the above configuration, for example, the train can be controlled depending on whether or not the worker recognizes the train, so that the brakes can be applied more appropriately.

The sensors include a first sensor (e.g., sensor A) and a second sensor (e.g., sensor B), and the identification unit includes sensor information of the first sensor and information of the second sensor The attribute of the object detected by the detection unit is specified from the sensor information (see FIG. 4, for example).

According to the above configuration, for example, the attribute of the object is specified using the sensor information of the first sensor and the sensor information of the second sensor, so the attribute of the object can be specified more accurately.

A third storage unit (for example, storage device 112, attribute type information storage unit 118) that stores the feature amount of the first sensor and the feature amount of the second sensor for each attribute of the object, and the specifying unit compares the feature amount of the sensor information acquired by the first sensor and the feature amount of the first sensor stored in the third storage unit to calculate a goodness of fit; The feature amount of the sensor information of the sensor and the feature amount of the second sensor stored in the third storage unit are compared to calculate the degree of conformity, and the sensor information of the sensor with the higher degree of conformity is calculated. is used to identify the attributes of the object (see, for example, FIG. 4).

In the above configuration, for example, the attribute of the object is specified using the sensor information of the sensor with the higher degree of conformity, so the attribute of the object can be specified more accurately.

When the attribute of the object determined from the sensor information of the first sensor is different from the attribute of the object determined from the sensor information of the second sensor, the specifying unit is configured to stop the train at a high deceleration. The attribute of the control side is specified as the attribute of the object.

In the above configuration, for example, even if the attribute of the object cannot be uniquely specified, the train can be operated safely by specifying, as the attribute of the object, the attribute of the stop control with the high deceleration of the train. can be done.

A fourth storage unit (for example, storage device 112) for storing work information (for example, work information) indicating a place where work is performed by the worker is provided, and the control unit is stored in the fourth storage unit. When it is determined that the current position of the train is the place where the work is to be performed by referring to the work information stored therein, the attribute of the object specified by the specifying unit, which is stored in the storage unit, corresponds to The train is controlled according to the control information.

With the above configuration, for example, it is possible to control the stop of the train in consideration of the place where the work is scheduled, so that it is possible to apply the brakes more appropriately.

A fourth storage unit (e.g., storage device 112) that stores work information (e.g., work information) indicating the time at which work is performed by the worker is provided, and the control unit is stored in the fourth storage unit. When it is determined that the current time is the time when the work is to be performed by referring to the work information stored in the , to control the train above.

With the above configuration, for example, the train can be controlled to stop in consideration of the scheduled work time, so that the brakes can be applied more appropriately.

The control unit controls the accuracy of the result of detection by the detection unit (for example, accuracy information regarding the presence or absence of a detected object), the accuracy of the result of calculation by the calculation unit (such as accuracy information regarding the calculated object point information), and the above If it is determined that one or more of the accuracy of the result of identification by the identification unit (for example, accuracy information for the attribute of the identified object) is lower than the threshold value, the identification by the identification unit stored in the storage unit The train is controlled according to control information with a higher deceleration than control information corresponding to the attribute of the object obtained and the distance calculated by the calculation unit.

In the above configuration, for example, when the accuracy of any of object detection, object position calculation, and object attribute identification is low, the train is operated safely by performing stop control with a higher deceleration of the train. be able to.

The sensor is provided on the train.

According to the above configuration, for example, as the train approaches the object, the accuracy of detecting the object and the accuracy of specifying the attribute of the object improve, so that the brakes can be applied more appropriately.

The sensor is provided on the ground.

According to the above configuration, for example, it is possible to detect an object even in a range where visibility from the train is poor, such as a distant place or a curved section, so that it is possible to apply the brakes more appropriately.

Moreover, the above-described configurations may be appropriately changed, rearranged, combined, or omitted within the scope of the present invention.

The present invention is not limited to the above-described embodiments, and includes various modifications. The above embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the described configurations. Also, part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. .

100...Train control system, 110...Train.

Claims (10)

A train control system for controlling a train running on a track,
a storage unit that stores control information related to stop control of the train for each attribute of an object;
a detection unit that detects an object from sensor information obtained by a sensor that acquires information related to the trajectory;
a specifying unit that specifies an attribute of an object detected by the detecting unit from sensor information obtained by the sensor;
a control unit that controls the train according to control information stored in the storage unit and corresponding to the attribute of the object identified by the identification unit;
a second storage unit that stores information on gestures of the object and information on the color of the object, the shape of the object, and/or the reflectance of the object, as feature amounts indicating attributes of the object;
with
The specifying unit refers to the second storage unit and specifies the attribute of the object detected by the detection unit using the feature amount of sensor information acquired by the sensor,
The storage unit stores control information related to a first stop control corresponding to an attribute of a worker who does not recognize the train, and stores control information related to a first stop control corresponding to an attribute of a worker who recognizes the train. Stores control information related to second stop control in which the deceleration is relatively lower than that of the first stop control,
When the identifying unit determines that the object is making a gesture indicating that the train is approaching, the attribute of the object is a worker who recognizes the train. identified as an attribute,
The control unit operates the brake according to the control information related to the second stop control, which is stored in the storage unit and corresponds to the attribute of a worker who recognizes the train identified by the identification unit. controlling the speed of said train without applying or applying brakes with low deceleration;
train control system. The storage unit stores control information related to stop control of the train for each attribute of the object and for each distance from the track,
a calculating unit that calculates the distance between the object detected by the detecting unit and the trajectory based on the sensor information obtained by the sensor and the position information of the sensor;
The control unit controls the train according to control information stored in the storage unit and corresponding to the attribute of the object identified by the identification unit and the distance calculated by the calculation unit.
The train control system according to claim 1. the sensors include a first sensor and a second sensor;
The identifying unit identifies an attribute of the object detected by the detecting unit from the sensor information of the first sensor and the sensor information of the second sensor.
The train control system according to claim 1. A third storage unit that stores the feature amount of the first sensor and the feature amount of the second sensor for each attribute of an object,
The specifying unit compares the feature amount of sensor information acquired by the first sensor and the feature amount of the first sensor stored in the third storage unit to calculate a goodness of fit, The feature amount of the sensor information of the second sensor is compared with the feature amount of the second sensor stored in the third storage unit to calculate the degree of adaptation, and the sensor having the higher degree of adaptation is selected. identifying attributes of the object using sensor information;
The train control system according to claim 3 . When the attribute of the object determined from the sensor information of the first sensor is different from the attribute of the object determined from the sensor information of the second sensor, the specifying unit determines that the train stops at a high deceleration. Identifying an attribute of the controlling party as an attribute of the object;
The train control system according to claim 3 . A fourth storage unit that stores work information indicating a place where work is performed by the worker,
The control unit refers to the work information stored in the fourth storage unit, and when determining that the current position of the train is the place where the work is to be performed, the work information stored in the storage unit , controlling the train according to control information corresponding to the attribute of the object identified by the identifying unit;
The train control system according to claim 1. A fourth storage unit that stores work information indicating the time when the work is performed by the worker,
The control unit refers to the work information stored in the fourth storage unit, and when determining that the current time is the time when the work is to be performed, the specific work information stored in the storage unit. controlling the train according to the control information corresponding to the attribute of the object identified by the unit;
The train control system according to claim 1. When the control unit determines that one or more of the accuracy of the result of detection by the detection unit, the accuracy of the result of calculation by the calculation unit, and the accuracy of the result of identification by the identification unit is lower than a threshold value, Controlling the train in accordance with control information stored in the storage unit that has a higher deceleration than control information corresponding to the attribute of the object identified by the identification unit and the distance calculated by the calculation unit;
The train control system according to claim 2. The sensor is provided on the train,
The train control system according to claim 1. The sensor is provided on the ground,
The train control system according to claim 1.

Images