WO2022172728A1

WO2022172728A1 - Train control system

Info

Publication number: WO2022172728A1
Application number: PCT/JP2022/002208
Authority: WO
Inventors: 健二今本; 潤小池; 泰志近江; 敬一勝田
Original assignee: 株式会社日立製作所
Priority date: 2021-02-15
Filing date: 2022-01-21
Publication date: 2022-08-18

Abstract

The present invention invention provides a train control system that controls the travel of a train traveling on a track, the train control system comprising: a sensor unit that acquires information about the environment around the train; an obstacle determination unit that, on the basis of sensor information received from the sensor unit, evaluates a degree of reliability quantifying the certainty of an obstacle determination result pertaining to whether obstacles are present around the track on which the train is traveling, the degree of reliability being evaluated as a relationship of magnitude compared with a determination reference; a control instruction storage unit that stores, as a control instruction rule, train control details associated with the result of evaluating the degree of reliability; and a train control unit that controls the speed of the train on the basis of the control instruction rule, the obstacle determination result, and the degree of reliability.

Description

train control system

The present invention relates to a train control system that detects foreign matter existing around the track on which the train runs and ensures the safety of the train based on the detection results.

In a track transportation system in which trains, etc., run on the track, if there is an obstacle around the track, it is not possible to steer the approaching train, etc. to avoid the obstacle from the track. It is important for improving the safety and operability to continuously monitor the road ahead and detect the presence or absence of obstacles and the like. Sensors such as cameras, LIDAR (Light Detection and Ranging), and millimeter wave radars are used as means for detecting obstacles. Based on sensor information obtained by digitizing these sensor outputs, it is conceivable to determine the presence or absence of an obstacle with a threshold value.

Therefore, in the railroad, which is one of the railroad transportation systems, an obstacle detection device in railroad crossings described in Patent Document 1 has been proposed as a technology for detecting objects around the railroad tracks. Patent Literature 1 discloses a technique for determining the presence or absence of an obstacle based on a result of comparison between an image in a pre-recorded image database and an image captured by a camera mounted on a train. In the technique of Patent Document 1, an imaging device is provided at the head of a train, and an image recognition database is provided on the train in which the track images captured by the imaging device are recorded in synchronization with absolute kilometers. It is characterized by comparing the image recognized by the device with the image recognition database, and actuating the brake when a different image is detected.

Japanese Unexamined Patent Application Publication No. 2010-063260

However, in the technique of Patent Document 1, when comparing an image recognition database with an image taken while driving, if the comparison and determination processing is performed to determine that the image is the same image only when the images match completely, a slight amount of dirt may be detected. and vegetation, etc., may cause frequent and excessive detection of obstacles. On the other hand, if the comparative judgment process allows a certain amount of difference, the sensitivity to obstacles will decrease. In some cases, we have no choice but to limit the operation of trains, such as by controlling speed. SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a train control system capable of improving the sensitivity for detecting obstacles while preventing the operation of trains from being impaired.

The present invention for solving the above problems is a train control system for controlling the running of a train running on a track, comprising a sensor unit for acquiring information on the surrounding environment of the train, and based on the sensor information received from the sensor unit. , a trouble judgment unit that evaluates the reliability, which is a numerical value of the likelihood of the trouble judgment result of the presence or absence of trouble around the track on which the train runs, as a magnitude relationship compared with the judgment standard, and the reliability evaluation result. A control instruction storage section for storing train control contents as control instruction rules, and a train control section for controlling train speed based on the control instruction rules, trouble determination results, and reliability.

The present invention provides a train control system that eliminates oversight of obstacles and appropriately reflects even sensor information suspected of excessive detection in control based on a comprehensive viewpoint of safety and operability. can provide.

BRIEF DESCRIPTION OF THE DRAWINGS It is a functional block diagram which illustrates the structure of the train control system (henceforth "this train control system" or "this system") which concerns on embodiment of this invention. In addition to the graph in which the curve of the judgment reliability of the sensor is superimposed on the operation curve by this system of FIG. . FIG. 3 is a graph and a table in which the determination criteria, which are fixed values in FIG. 2, are changed according to the detection distance and the train speed. 3A and 3B are graphs and a table in which the determination criterion, which is set to one fixed value in FIG. 2, is changed stepwise; FIG. 3 is a graph and a table to which a highly accurate sensor characteristic interval is applied to FIG. 2; FIG. 6 is a flow chart showing processing of the system shown in FIGS. 1 to 5; FIG.

First, using Fig. 1, the configuration of this system and the role of each component will be explained. FIG. 1 is a functional block diagram illustrating the configuration of this system. The train 10 is composed of one or more train cars (hereinafter referred to as cars). This system can basically be configured simply with only the on-board equipment of the train 10 .

The on-board device of the train 10 includes a sensor section 101, a trouble determination section 102, a control instruction storage section 103, and a train control section 104. Although not shown, the train 10 is a vehicle that transports passengers and freight traveling along a railroad track, and is equipped with a braking device that reduces the running speed of the train 10 based on instructions from the train control unit 104 .

Although the system shown in FIG. 1 is basically composed of only the on-board equipment of the train 10, each component may be distributed and installed within the ground equipment. Ground facilities can be enumerated along railway lines, stations, control rooms, and the like. For example, the sensor unit 101 may be installed at a railroad crossing, and the function of the control instruction storage unit 103 may be provided in the form of a cloud service on a network. When each unit is installed outside the train 10, the train 10 is configured to separately include a communication unit for communicating with each unit. In the present invention, any specific system configuration may be used as long as the configuration realizes the functions of each unit. Therefore, detailed description related to the communication unit is omitted.

The sensor unit 101 uses sensors mounted on the train 10 to acquire sensor information when monitoring the area around the train 10 . It is desirable that the sensor used has performance suitable for object detection. For example, various means such as cameras, millimeter wave radar, LIDAR, and ultrasonic sensors can be applied, and multiple sensors can be used in combination. .

The sensor unit 101 transmits sensor information obtained during forward monitoring to the trouble determination unit 102 . The sensor information differs depending on the type of sensor used. For example, when a camera is used, it is image information of an object, and when a millimeter wave radar or LIDAR is used, it is reflection intensity and distance information from an object.

Based on the sensor information received from the sensor unit 101, the obstacle determination unit 102 determines whether there is an obstacle around the track on which the train 10 runs, and secondarily evaluates the reliability of this primary determination result. When it is determined that there is an obstacle, the distance from the train 10 to the obstacle is also estimated. The trouble determination unit 102 transmits the trouble determination result, the distance estimation result, and the reliability evaluation result to the train control unit 104 . Note that the primary determination process and the secondary evaluation process are distinguished for convenience of explanation, and do not have to be strictly distinguished. In addition to sensor information, this system adds a lot of objective information and uses the results of AI processing in multiple stages to ensure the safety of trains.

Technology for object detection used in other fields such as automobiles can be used for the judgment of the presence or absence of obstacles by the obstacle judgment unit 102 and the process of estimating the distance from the train 10 to obstacles. For example, there is an obstacle recognition method of creating parallax images using a stereo camera and recognizing the shape and position of an object in front from the parallax images.

There are also obstacle recognition methods that use a DNN (Deep Neural Network) to recognize objects on images from monocular images, and obstacle recognition methods that detect the presence and distance of objects from LIDAR point cloud data. At this time, DNN is one of the means used for machine learning, and by extracting and learning the features of the object, it is possible to recognize various objects and improve the accuracy of obstacle detection. . In the present invention, any method is acceptable as long as an obstacle in front of the train can be detected.

Technology for object detection used in other fields such as automobiles can also be used for the process of secondary evaluation of the reliability of the primary judgment result of the presence or absence of trouble by the trouble determination unit 102 . For example, when using a camera, there is a reliability calculation method of calculating reliability based on the degree of similarity between an acquired camera image and pre-learned data. Further, when LIDAR or millimeter wave radar is used, there is a reliability calculation method of calculating the reliability based on the number of point cloud data obtained by reflection from an object determined as an obstacle and the reflection intensity.

Also, based on the estimated distance between the train 10 and the obstacle, for example, the reliability of the primary determination result may be secondarily lowered as the distance increases. In addition, the results of determining whether there is an obstacle for the sensor information received at different timings are checked, and if an obstacle is continuously detected (that is, if the obstacle can be tracked as the same object), the reliability is evaluated highly. can be

In addition, when the sensor unit 101 uses a plurality of sensors, sensor fusion technology can be used. A secondary evaluation for reliability may be performed. In the present invention, it is sufficient that the reliability of the primary determination result of the presence/absence of trouble can be evaluated secondarily, and the method is not limited.

Up to this point, the primary judgment processing by the obstacle judgment unit 102 and the secondary evaluation processing of the reliability of the primary judgment result have been described. was described as an example. Not limited to this, the same evaluation process can be performed even if a function to confirm that there is no obstruction and set "no obstruction" is assumed. For example, if known trackside structures are properly detected, with a high degree of confidence "no trouble" (i.e., there is no object between the known trackside structures and the train 10) and artificial intelligence (AI : Artificial Intelligence).

In this case, when the reliability based on the acquired sensor output (hereinafter referred to as "sensor information") is above a certain criterion (threshold), if it is not judged as "no problem", a gray zone that cannot be said either way should be provided. It is also possible to make a logical determination that "there is a problem" by interpreting it in the opposite direction. Similarly, when the reliability is equal to or higher than a certain criterion, if it is not determined as "problem", it may be logically determined as "no problem".

In this system, secondary evaluation results are weighted and reflected in train control. When the reliability of the primary judgment processing of the presence or absence of trouble by the obstacle judgment unit 102 and the reliability of the primary judgment result is below a certain judgment standard, the judgment result is not reflected in the train control and is rejected or taken lightly. It makes sense to have a gray zone. For example, if a known trackside structure is not properly detected due to bad weather or nighttime, it is inappropriate to judge that there is a problem, so the reliability should be reduced or set to zero. be.

The trouble determination unit 102 may determine the presence or absence of obstacles and position information using not only newly acquired sensor information but also past sensor information. For example, a method such as a Kalman filter may be used to estimate the operating state of the obstacle using past sensor information. For example, when past sensor information detects an object moving in a straight line, even if the sensor information cannot be obtained at a certain moment, the object still exists at the position predicted from the past sensor information. , and the subsequent processing may be executed.

Information other than sensor information may be used in the process of secondary evaluation of the reliability of the primary determination result of the presence or absence of a failure by the failure determination unit 102 . For example, the reliability may be evaluated based on environmental information such as weather information, travel time zone and travel location.

In other words, information that generally lowers the detection reliability compared to normal times includes weather information that changes over time during bad weather such as rain and snowfall, nighttime, early morning, and evening (hours affected by backlight), etc. There is information on the running time zone of In addition to visibility information around stations, which does not change in a short period of time but needs to be updated as appropriate, there is also travel point information such as in tunnels and mountainous areas. It is preferable that the present system evaluates the reliability while adjusting the reliability based on such factors of deterioration of the detection reliability.

Only when the driving route is confirmed in advance, it is possible to improve the detection reliability of this system by preparing and storing environmental information that depends on the driving location, etc., such as that used by the car navigation system. It is also possible to use it as a reference for evaluation. In addition to such stored information, the present system may evaluate the reliability based on the latest real-time environmental information acquired during driving.

The control instruction storage unit 103 stores a combination of a determination criterion of reliability for the primary trouble determination result and a train control content when a secondary evaluation of reliability exceeding or falling below the criterion is obtained. , are stored as control instruction rules. An example of a train control method (hereinafter also referred to as "this method") according to an embodiment of the present invention will be described using FIG.

Fig. 2 shows a graph in which the sensor's determination reliability curve is superimposed on the operation curve of this system in Fig. 1. Below the graph in FIG. 2, there is also shown a table of control instruction rules in which control details are associated with comparison results between the fixed criterion A and the determination reliability. For example, a train control method is conceivable in which the brake is operated in the case of a failure determination result with a reliability exceeding a certain criterion A, and coasting is performed in the case of a reliability lower than that.

　Criteria A shown in the graph of Fig. 2 functions as a threshold for numerically fluctuating reliability. In this system, if the judgment criterion (threshold) A is high, judgment results with low reliability are ignored or taken lightly. Conversely, if the criterion (threshold value) A is low, even low-reliability determination results are emphasized and reflected in the control. In this system, the criterion A acts like the acuity and strictness of the control rule, so it is necessary to determine whether the result of applying it tends to over-detect or miss an obstacle.

In addition, it is desirable to set the criterion A to an optimum value based on the degree of impact on the operation delay time as a result of applying this system, as well as the presence or absence of backup by the driver. For example, even if this system detects excessively, if the driver confirms and properly cancels the reflection in the operation control, ideal safe driving may be possible.

In addition, the reliability of the failure determination result by the failure determination unit 102 is expected to vary depending on the route environment and operating conditions, such as the difference in the frequency of false detections due to the number of structures along the route. It is desirable to set the determination criterion A in consideration of the acquisition status of the reliability in normal times based on the driving results.

The control instruction storage unit 103 stores the distance from the train 10 to the obstacle and the train speed in order to start braking based on the stoppable distance of the train 10 when setting the reliability criterion A for the obstacle determination result. should be set to the optimum value based on Another example of this method will be described with reference to FIG.

FIG. 3 is a graph and a table showing determination criteria F that are changed according to the detected distance and train speed, instead of the determination criteria A that are fixed values in FIG. The detection distance here is the distance from the sensor unit 101 mounted on the train 10 to the obstacle detected by it. change to

In other words, the detection distance is the distance to the object while the sensor is detecting the object. Even an object that should be detected is outside the range of the maximum detection distance if it is too far away for the sensor to detect. That is, here, the detection distance within the range that can actually be detected is used, and is distinguished from the maximum detection distance, which means the capability of the sensor.

In this way, sensor performance specifications (specs) are represented by the maximum detection distance based on range resolution, the maximum detection speed based on velocity resolution, and the maximum detection angle based on angle resolution. Therefore, the reliability of sensor information obtained by a sensor with constant distance resolution and speed resolution decreases as the detection distance increases and as the train speed increases. On the other hand, from the viewpoint of safety, which is a different dimension from the reliability of sensor information, the longer the detection distance and the slower the train speed, the higher the safety.

The reliability of the sensor information output in a quantifiable manner can be determined as either high or low depending on the magnitude relationship compared to a certain threshold. If the reliability is high, the sensor information is weighted so as to be emphasized and reflected in the control, but if not, it is neglected or ignored. A threshold for judging reliability in this manner is called a judgment criterion F. FIG. This criterion F may be varied stepwise or steplessly according to the situation.

For example, when the distance from the train 10 to the obstacle is long, or when the train speed is slow, a high criterion F is applied. If the criterion F is high, it is not strongly reflected in the control unless the information is more reliable. If the brake is applied from that point, it is possible to stop at a point far enough away from the obstacle, so at best, continue powering suppression or coasting until the AI can determine whether the obstacle is a person, a vehicle, or a small animal. It is a judgment that is also acceptable.

Conversely, if the distance is short or the train speed is fast, apply the low criterion F. If the criterion F is low, even if the reliability of the sensor information is so low that it falls below it, it is reflected in the control unless it is clearly determined that it is a false alarm. For example, when an obstacle is obtained as sensor information at a close distance before a railroad crossing, even if the AI cannot determine whether the obstacle is a person, a vehicle, or a small animal, the emergency brake is activated for the time being.

If the brakes are applied from that point, there is a possibility of colliding with an obstacle unless the brakes are applied strongly, so this decision was made to avoid that situation. Considering safety, the above-described control mode in which the weight for reflecting sensor information in control is changed depending on whether the reliability determination result for sensor information is high or low can be rephrased as follows.

When it is presumed that the operating state of the train 10 is clearly secured with a high degree of safety, and the criterion F for reliability is set high, the failure determination sensor with a degree of reliability higher than that is set. Do not apply the brakes unless information is available. As a result, the system is less susceptible to over-controlling. Conversely, when it is estimated that the safety level based on the current operating state of the train 10 is low and the reliability criterion F is set low, the sensor information of the obstacle is obtained with a correspondingly low level of reliability. The brakes are activated quickly even in the state of being caught. As a result, the system is easy to secure.

As shown in Figure 3, this system can continuously change the reliability criterion F according to the distance to obstacles and the train speed. By doing so, even in a running environment where it is difficult to obtain a highly reliable judgment, if there is a risk of a collision due to approaching, this system will lower the judgment criterion F to ensure safe train control. It becomes possible.

When setting the reliability criterion F, this system considers the acquisition status of reliability in normal times based on the driving results on the actual route. It is desirable to In other words, this system uses train speed, vehicle performance (deceleration, vehicle weight, running resistance, etc.) and line alignment information (slope, curve, etc.) as parameters for calculating the possible stopping distance for each point. , a criterion F may be set.

When the control instruction storage unit 103 optimally sets the contents of train control based on the distance from the train 10 to the obstacle and the train speed, the stop possible distance and the necessary deceleration of the train 10 are calculated using the formulas (1) to Calculate in (3).

Ls = Li + Lb (1)
Li＝v×ti …(2)
Lb= ^v2 /2β (3)

where Ls[m] is the possible stopping distance, Li[m] is the idling distance, Lb[m] is the braking distance, v[m/s] is the train speed, ti[s] is the idling time, β[ m/s ² ] is the deceleration. The idling time refers to the time from when the train control unit 104 outputs a control instruction to when the train 10 actually starts decelerating. Based on the calculated required deceleration, the strength of the braking force or corresponding braking notch is selected.

When the control instruction storage unit 103 sets the reliability criterion F for the failure determination result, the control instruction storage unit 103 sets the strength of the braking force according to the reliability of the failure determination result. Also good. Another example of this method will be described with reference to FIG. FIG. 4 is a graph and a table showing determination criteria A to C that are changed stepwise instead of the single fixed determination criteria A in FIG.

　The criteria for determination reliability are fixed at A in FIG. 2, F is variable in FIG. In this system, as shown in the table of control instruction rules in FIG. remembered. In this system, the control instruction storage unit 103 changes the stored control content according to the result of comparing the determination reliability based on the sensor output (sensor information) of the sensor unit 101 with the determination reliability A to C. Create a run curve to read and run.

That is, for each setting of criteria A to C classified into three ranks illustrated in Fig. 4, in the case of a trouble determination result with reliability exceeding criteria A, which is the highest, the content of train control is emergency braking. Similarly, when the second criterion B is exceeded, the train control content is set to the regular brake. Similarly, when the third criterion C is exceeded, the content of train control is coasting. Similarly, if the value falls below the third criterion C, train control content is set to power running suppression.

This method is similar to the concept of "possible driving" that is commonly performed in manual driving of a car. This "driving with possible danger" means to drive while always paying attention to the surrounding situation and foreseeing danger. 4, the sensor unit 101 detects some obstacle in front of the train 10 and the train 10 is less than the criterion C at the initial detection point, power running is suppressed.

As the train 10 approaches the obstacle point, if the criterion C is exceeded, the contents of the train control are changed to coasting. When the train 10 further approaches the trouble point and exceeds the criterion B, the contents of the train control are changed to the regular brake. Furthermore, when the train 10 is closest to the obstacle point and exceeds the criterion A, the contents of the train control are changed to an emergency brake.

In this way, even if only a low reliability is obtained from the judgment results regarding the judgment reliability, by starting to reduce the speed at an early stage within the range where the influence on the operation can be suppressed, it is possible to approach the estimated trouble point at a low speed. However, a large amount of sensor information can be obtained, and reliable train control with higher reliability is possible.

The control instruction storage unit 103 uses a method of setting the optimum value A based on the distance from the train 10 to the obstacle and the train speed when setting the reliability criterion for the obstacle determination result, and two or more determination methods. It may be set in combination with the method of setting the criteria A to C.

The control instruction storage unit 103 also sets, as the control contents to be recorded as the control instruction rule, the control contents in the case where only the judgment result of reliability less than a certain criterion can be obtained regardless of the trouble judgment result. is desirable. For example, even if it is determined that there is no problem, if the reliability of the determination result is low, power running suppression or coasting may be considered. In addition, when it is determined that "there is a problem" in the previous determination, even if it is determined that the reliability is low and "there is no problem", it is conceivable that the control contents immediately before are retained.

In the control instruction storage unit 103, the control content recorded as the control instruction rule may be a "multi-stage brake control method" that relaxes the brake control when the train speed drops below a predetermined value after a predetermined brake control. In addition, without being limited to such a multi-stage control method, a target point and a target speed are designated as control contents, and the brake start point and braking force are optimized so that the train control unit 104 can decelerate to the point. It is good also as a "single step brake control system" which determines.

In this system or method, it suffices if appropriate train control can be selected based on reliability, and it does not matter whether the control content recorded as the control instruction rule is a "multi-stage brake control system" or a "single-stage brake control system." do not have. That is, the first feature of the present invention is to optimally select various brake control methods according to reliability. Another example of this method will be described with reference to FIG. In relation to this, the second feature of the present invention is to obtain good results by setting a sensor characteristic interval (FIG. 5) with good sensor characteristics and high reliability so as to take advantage of the goodness of the sensor.

FIG. 5 is a graph and table to which highly accurate sensor characteristic intervals are applied to FIG. In order to make the best use of the sensor, the control instruction storage unit 103 uses a running curve based on the sensor characteristics (detection distance, viewing angle, train speed, processing time, etc.) mounted on the train 10 as the control contents to be recorded as control instruction rules. can be specified. If the vehicle is driven so that the driving curve falls within the "sensor characteristic interval" shown in FIG. 5, the object detection and trouble determination are highly reliable, and safety can be easily ensured. Therefore, the distance range in which the sensor characteristics are optimal is set as the sensor characteristic section, which is the optimal environment assumed by this system. That is, the train 10 runs in an appropriate speed range, making it easier to ensure safety.

The sensor characteristic interval can be set based on the performance specifications (specs) of the sensor such as maximum detection distance and maximum detection speed. For example, if a camera is used as a sensor in this system, only low-reliability detection is possible at a distance of around 1 km based on the performance specifications of the camera. Suppose it can be detected with high reliability. In such a system, after detecting at a point about 1 km from the obstacle, it is conceivable to specify a running curve that approaches the point at a distance of 500 m from the obstacle at a train speed of 70 km/h or less.

In addition, even if this system is combined with a camera with the above-mentioned performance specifications and multiple sensors with different specifications, a driving curve can be specified in a similar manner. For example, when using LIDAR, which can detect objects within a detection distance of 300 m and at a speed of 80 km/h or less with high reliability, it is preferable to specify a driving curve that approaches the point at a distance of 300 m at 70 km/h or less.

This system, which is equipped with such a combination of sensors, can detect objects and determine obstacles with higher reliability by combining both sensors. It is desirable to set such sensor characteristics and their combinations in consideration of performance specifications by sensor manufacturers, performance evaluations based on driving results, and the like.

The train control unit 104 controls the train 10 based on the trouble determination result received from the trouble judgment unit 102, the distance estimation result, the reliability evaluation result, the control instruction rule received from the control instruction storage unit 103, and the running speed of the train 10. Train control instructions are generated to control the train 10 . As a method of controlling the brake, for example, the generated train control instruction may be sent to an ATO device (automatic train operation device) to automatically control, or the driver may be provided with driving support information such as an alarm. Manual brake control may be prompted.

Whether to apply automatic control or manual control may be decided before the train runs, or the driver may decide manually after the train starts running. Further, when receiving the reliability information, the train control unit 104 may determine whether to apply automatic control or manual control based on the reliability.

For example, train control is normally performed by automatic control, but if the reliability of the failure determination result is lower than a predetermined value, it is determined that the system may have made an error, and the system manually instructs the driver to Operation such as prompting switching to control is conceivable. Examples of specific devices for the braking/driving means of the train 10 include inverters, motors, and friction brakes.

When the obstacle determination unit 102 determines that there is no obstacle in front of the train 10, the train control unit 104 generates a brake release instruction to control the train 10 based on the train running speed and the reliability evaluation result. good. According to such processing of the train control unit 104, the effect of being able to quickly return to normal running from the suppressed operation state can be expected.

For example, after the train control unit 104 determines that there is a "problem" in the previous processing cycle and issues a brake instruction, the obstacle is removed and it is confirmed that there was a temporary sensor error detection. If it is confirmed that the train is running safely, it is expected that the brakes will be released to avoid unnecessary deceleration and return to normal running as soon as possible. The above is the configuration of the train control system and the description of each component.

Next, the processing flow of the train control system will be explained using FIG. FIG. 6 is a flowchart showing the processing of the system shown in FIGS. 1-5. The processing shown in FIG. 6 is executed at regular intervals. Step 900 is executed by the sensor unit 101 , steps 901 to 904 are executed by the trouble determination unit 102 , and steps 905 to 906 are executed by the train control unit 104 .

At step 900, current sensor information is obtained from the sensor unit 101. At step 901 , based on the sensor information acquired by the sensor unit 101 , it is primarily determined whether there is an object that hinders the running of the train 10 . In step 902 , based on the primary determination result in step 901 , if “problem” is determined, the process proceeds to step 903 , and if “no problem” is determined, the process proceeds to step 904 .

In step 903 (determining that there is an obstacle), the distance between the train 10 and the detected obstacle is estimated based on the sensor information, and the process proceeds to step 904. In step 904, the reliability of the trouble determination result is secondarily evaluated based on the sensor information and more objective criteria.

In step 905, a train control instruction for the train 10 is generated based on the trouble determination result, the distance estimation result, the reliability evaluation result, the control instruction rule received from the control instruction storage unit 103, and the running speed of the train 10. to control. In the case of manned operation, instead of directly controlling the train 10 by train control instructions, the results of intrusion determination and brake instructions are presented to the driver by screen display, voice information, alarms, etc. on the cab. Also good.

At step 906, the running speed of the train 10 is controlled based on the contents of the train control instruction (braking instruction or brake release). The above is the description of the flow example of the train control operation executed by the present train control system. This system realizes safe and stable train control according to the operating environment and sensor characteristics by controlling the brakes according to the reliability of the trouble determination result in such a procedure.

A train control system (this system) according to an embodiment of the present invention can be summarized as follows.
[1] This system includes a train 10 , a sensor section 101 , a trouble determination section 102 , a control instruction storage section 103 and a train control section 104 . A train 10 runs on a track. The sensor unit 101 acquires information on the surrounding environment of the train 10 . Based on the sensor information received from the sensor unit 101 , the obstacle determination unit 102 primarily determines whether there is an obstacle around the track on which the train 10 runs, and stores the result in the control instruction storage unit 103 . Further, the trouble determination unit 102 considers the certainty of the primary trouble judgment result obtained in this way, and considers the environment such as the weather and the running position of the train, etc., and determines the reliability based on more objective judgment criteria. Secondarily evaluate the degree.

　Reliability can be digitized and processed. Further, the control instruction storage unit 103 optimally sets train control details based on the detected distance from the train 10 to the obstacle and the train speed. The failure determination unit 102 secondarily evaluates the reliability of the primary failure determination result read from the control instruction storage unit 103 based on whether it exceeds or falls below a more objective criterion. , is stored in the control instruction storage unit 103 .

Also, the combination of the reliability evaluation and the train control content to be applied according to the evaluation is set as the control instruction rule. This control instruction rule is stored in the control instruction storage unit 103 . Also, the train control unit 104 controls the train speed based on the control instruction rule, the obstacle determination result, and the reliability of the obstacle determination result.

This system configured as described above can prevent obstacles from being overlooked and over-detected on a practical level in a well-balanced manner. For example, in this system, when the image recognition database and the image taken while driving are compared and judged to obtain a trouble judgment result, by appropriately setting the judgment criteria, if the judgment criteria are not completely matched, is too strict, making it easier to avoid over-detection defects. Conversely, the system tends to be too lenient to tolerate substantial differences, thus avoiding the problem of missing actual obstacles.

　According to this system, the reliability of the trouble determination result by the external sensor is appropriately evaluated based on the current situation, enabling the operation of the optimum braking force according to the driving environment and sensor characteristics. By doing so, this system eliminates the overlooking of obstacles, and on the contrary, even sensor information that is suspected of being overdetected is appropriately reflected in control based on a comprehensive viewpoint of safety and operability. Let As a result, this system can realize safe and stable train control.

[2] In the present system of [1] above, based on the sensor information received from the sensor section 101, the obstacle determination section 102 estimates the detected distance between the train 10 and obstacles and the train speed. In addition, the control instruction storage unit 103 sets and stores in a readable manner a reliability criterion for the primary failure determination result based on the detected distance between the train 10 and the obstacle and the train speed. In this system, the threshold for secondarily judging reliability is called a criterion.

With many sensors, the detection performance and reliability of detection results fluctuate due to the distance to obstacles, bad weather, and changes in the surrounding environment such as at night. If a fixed threshold value is applied to threshold determination based on sensor information on the premise that the reliability of detection results fluctuates, it is difficult to obtain an ideal determination result that does not interfere with train running. Therefore, it is not necessarily appropriate to apply a uniformly fixed threshold value (A in FIG. 2) as a reliability criterion for the trouble determination result.

Therefore, in this system, necessary information is stored in the control instruction storage unit 103 so that the determination criterion (threshold value) can be varied stepwise or steplessly according to the situation. The stored contents are updated accordingly.

As shown in FIG. 3, in the control instruction storage unit 103, the criterion F for the reliability of the obstacle determination result is the distance from the train 10 to the obstacle, and the optimum value F should be set based on the train speed. If the reliability of the failure determination result is high, the sensor information is weighted so as to be emphasized and reflected in the train control, but if not, it is neglected or ignored.

For example, if the distance from the train 10 to the obstacle is long, or if the train speed is slow, a high criterion is applied. If the criterion is high, it is not strongly reflected in the control unless the information is more reliable than the criterion. Conversely, when the distance to the obstacle is short or the train speed is high, a low criterion is applied. If the criterion is low, even if the reliability of the sensor information is so low that it falls below the criterion, it is reflected in train control unless it is clearly determined that it is an erroneous alarm.

When the train 10 operates the brakes for the first time at a point near the obstacle point, the system may collide with an obstacle unless the brakes are strongly applied. do. In this way, the present system changes the weighting for reflecting the trouble determination result in the control depending on whether the reliability of the sensor information and the trouble judgment result based on the sensor information is high or low.

As shown on the left side of the graph in FIG. 3, in this system, when it is estimated that the operating state of the train 10 clearly ensures a high degree of safety, the reliability criterion is set high. . Therefore, the system will not apply the brakes unless sensor information is obtained for fault determination with a higher degree of reliability than that. As a result, the system more reliably reduces the problem of excessive control.

Conversely, as shown on the right side of the graph in FIG. 3, when the safety level is estimated to be low based on the current operating state of the train 10, the reliability criterion is set low. Therefore, the system will not miss even if the sensor information of the obstruction is obtained with a correspondingly low level of confidence, and will actuate the brakes in an agile manner. As a result, the system is easy to secure.

As shown in Fig. 3, this system can continuously change the reliability criteria according to the distance to obstacles and train speed. By doing so, even in a running environment where it is difficult to obtain a highly reliable judgment, if there is a risk of collision due to approaching, this system can control the train while ensuring safety by lowering the judgment criteria. becomes.

[3] In the present system of [1] above, the secondary evaluation of the reliability of the primary failure determination result is output by the failure determination unit 102 . Based on the output reliability evaluation, the control instruction storage unit 103 preferably sets the strength of the braking force according to the reliability level.

The contents of train control are reflected as follows for each setting of the criteria A to C of the three ranks illustrated in Fig. 4. That is, when the trouble determination result exceeds the highest criterion A, the emergency brake is applied. If the trouble determination result exceeds the second criterion B, the brake is used for normal use. If the trouble determination result exceeds the third determination criterion C, the content of the train control is coasting operation. If the trouble determination result is less than the third determination criterion C, the content of train control is to suppress power running.

In this way, even if only a low degree of reliability is obtained for the trouble determination result, this system can speed up the speed control at an early stage with train control contents such as power running suppression, coasting operation, and regular braking in a range where the influence on operation can be suppressed. By starting to drop , a lot of sensor information can be obtained while approaching the probable obstruction point at a slow speed. As a result, this system enables reliable train control, such as actuating the emergency brake only after a highly reliable trouble determination result.

[4] In the system of [1] to [3] above, in addition to the primary sensor information received from the sensor unit 101, the trouble determination unit 102 receives at least one of weather information, time information, or travel location. Based on the above information, the presence or absence of obstacles around the track on which the train 10 runs is determined, and the reliability of the obstacle determination result is secondarily evaluated. Such a system is more realistic and practical. For example, it is possible to effectively adopt a method of judging "no problem" when a known structure along the railroad is properly detected, and judging "has a problem" in the opposite case. In other words, a trouble determination method using sensor information whose reliability fluctuates based on weather information or time information can also be effectively adopted in this system.

[5] In the present system of the above [1] to [4], the control instruction storage unit 103, as the control content to be recorded as the control instruction rule, detects a detectable distance or viewing angle assumed by the sensor unit 101 or train speed Alternatively, the running curve of the train 10 may be designated based on at least one piece of information of the processing time. In other words, the control instruction storage unit 103 may designate a running curve based on the characteristics of the sensors mounted on the train 10 (detection distance, viewing angle, train speed, processing time, etc.).

For example, the distance range in which the sensor characteristics are optimal is set as the sensor characteristic section, which is the optimal detection range assumed by this system. Then, in this system, the sensor characteristic section shown in FIG. 5 is easy to take advantage of the sensor, so the train 10 can be controlled with high accuracy. In this system, if the driving curve is kept within the "sensor characteristic section" shown in FIG. 5, the object detection and trouble determination are highly reliable, and safety can be easily ensured. That is, according to this system, if the train 10 is run in an appropriate speed range in the sensor characteristic section, it is easy to perform precise control, and it is easy to ensure safety.

The train control method (this method) according to the embodiment of the present invention can be summarized as follows.
[6] This method is a train control method for controlling running of the train 10 running on the track, and has the following processes. First, the sensor unit 101 acquires information on the surrounding environment of the train 10 (step 900). Next, based on the sensor information received from the sensor section 101, the obstacle determination section 102 primarily determines whether or not there is an obstacle around the track on which the train 10 runs (step 901).

Next, the trouble determination unit 102 evaluates the degree of reliability of the certainty of the trouble judgment result regarding the presence or absence of a primary trouble by comparing the magnitude relation with a more objective secondary judgment criterion (step 904). The control instruction storage unit 103 stores, as a control instruction rule, a combination of the judgment criteria, the evaluation obtained by the comparison, and the train control content to be applied based on the evaluation. Next, the train control unit 104 reads out the control instruction rule according to the reliability of the trouble determination result from the control instruction storage unit 103, and controls the train speed based on the control instruction rule (step 906).

On railway lines, the acquisition of sensor information may become unstable at night or in bad weather. In that case, based on the acquired sensor information, an appropriate braking force according to the reliability of the trouble determination is set for the judgment results regarding the presence or absence of trouble in steps 900 to 901. A safe and stable train control method can be realized. Based on the detected distance and train speed obtained from sensor information, the evaluation of the reliability of the determination result of the presence or absence of trouble is changed, and the reflection in train control is weighted according to the evaluation.

With this method, if there is a high level of safety, only the content with a high degree of reliability is reflected in the control to avoid obstruction to normal operation due to excessive detection. Conversely, if the safety level is low and the situation is tense, even if the contents of the determination result of the presence/absence of obstacles are reflected in the control even if the reliability is low, it is possible to avoid overlooking obstacles. In this way, the present method eradicates oversight of obstacles, and on the other hand, even sensor information suspected of overdetection can be used for control based on a comprehensive viewpoint of safety and operability. Reflect as appropriate. As a result, according to this method, safe and stable train control can be achieved.

[supplement]
The sensor system (a combination of the sensor unit 101 and the trouble determination unit 102), which is also applied to this train control system, detects people and dangerous objects, outputs a primary trouble judgment result, and puts the controlled object in a safe state. The purpose is to protect people by doing so, and the sensor function and accuracy are secondarily objectively evaluated in accordance with the sensor system standard IEC62998. In other words, the sensor system that acquires information on the surrounding environment of the train 10 uses the sensor information received from the sensor unit 101 to determine not only the accuracy of the primary sensor functions, but also Regarding the reliability of the obstacle determination result by detecting people and dangerous objects, we will conduct a secondary objective evaluation, taking into consideration the environment and train conditions. This system operates trains according to control details that are appropriately weighted according to this evaluation.

In order to achieve the purpose of the sensor system standard IEC62998, it is preferable to use "YOLO (You-Only-Look-Once)" to detect people and dangerous objects. "YOLO" is a relatively new image recognition algorithm and has the following advantages. First, high precision and fast processing. Secondly, since the information of the entire image is captured and predicted, the problem of erroneously detecting the background as an object can be reduced. In other words, YOLO can learn and verify from the information of the entire image including the background, so it can rationally recognize the relationship between the background and the object based on learning.

The reliability of the sensor changes depending on the environmental conditions. Here, the sensor characteristics mainly refer to the performance in a range in which the sensor can easily demonstrate its detection ability with high accuracy, that is, in a range of high utility value. For example, areas of strength related to detection distance and speed are defined as sensor characteristics. A characteristic technique of the present invention is a technique for activating the optimum braking force according to the driving environment of the vehicle and the sensor characteristics based on the reliability of the information detected by the sensor according to the sensor characteristics. It should be noted that the control in which the braking intensity is changed in stages to obtain the optimum braking force is suitable for railway vehicles because it is easy to adapt to the driving operation involving the number of notches.

In other words, this train control system recognizes sensor characteristics that are not necessarily compatible with all environments, and then controls the train using the area in which the sensor is good at detecting distances from obstacles and train speed. The contents of control for the train are, for example, contents such as emergency braking, normal braking, coasting, and power running suppression. At that time, the stop possible detection distance and the necessary deceleration of the train 10 are calculated by the above formulas (1) to (3). Prior to formulating these formulas, sensors are objectively evaluated secondarily in terms of sensor functions and trouble determination accuracy based on the sensor information that they primarily output, as well as the environment and train conditions. If the sensor information and trouble determination result are highly reliable, they are weighted and quickly reflected in the control.

This train control system is basically premised on unmanned operation, but it is not necessarily limited to that, and a driver may coexist in addition to fully automated operation. In that case, the content of control is presented as driving assistance information to the driver who is intervening on the route through which the content of control is transmitted to the traveling device.

In the configuration of this system, if there is a function equivalent to a "storage unit", it should include one or more memories. At least one memory in the storage unit may be a volatile memory or a non-volatile memory. The "control instruction storage unit 103" is a computer-equivalent unit formed by combining a storage unit storing programs with a processor unit (CPU) and an input/output interface. This computer-equivalent part may be realized by a one-chip microcomputer, may be used as part of another computer, or may be realized by hardware circuits alone.

In addition, the functions have been described using the expression “kkk unit”, but these functions may be realized by executing one or more computer programs by the processor unit, or may be realized by one or more hardware circuits. May be. Note that the description of each function is an example, and a plurality of functions may be combined into one function, or one function may be divided into a plurality of functions.

The present invention is not limited to the above-described embodiments, and includes various modifications. These embodiments are 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.

DESCRIPTION OF SYMBOLS 10... Train 101... Sensor part 102... Trouble determination part 103... Control instruction storage part 104... Train control part 900-906... Each step

Claims

A train control system for controlling running of a train running on a track,
a sensor unit that acquires information about the surrounding environment of the train;
Based on the sensor information received from the sensor unit, a trouble judgment unit that evaluates the degree of reliability, which is a numerical value of the likelihood of the trouble judgment result of the presence or absence of trouble around the track on which the train runs, as a magnitude relationship compared with a judgment criterion. When,
a control instruction storage unit that stores, as a control instruction rule, train control content associated with the reliability evaluation result;
a train control unit that controls a train speed based on the control instruction rule, the trouble determination result, and the reliability;
A train control system with
The obstacle determination unit estimates the detected distance from the train to the obstacle and the train speed based on the sensor information,
The control instruction storage unit sets a criterion for determining the reliability based on the detected distance and the train speed.
The train control system according to claim 1.
The control instruction storage unit sets the strength of the braking force according to the reliability level.
The train control system according to claim 1.
In addition to the sensor information received from the sensor unit, the obstacle determination unit determines whether or not there is an obstacle around the track on which the train runs, based on at least one or more of weather information, time information, and travel point information. , and assessing the reliability;
The train control system according to any one of claims 1 to 3.
The control instruction storage unit records control details as the control instruction rule, and is based on information on at least one or more of detection distance, viewing angle, train speed, and processing time assumed by the sensor unit. Prescribing a running curve of the train so as to execute the control content,
The train control system according to any one of claims 1 to 4.
A train control method for controlling running of a train running on a track,
A sensor unit acquires information on the surrounding environment of the train,
A trouble determination unit primarily determines whether there is a trouble around the track on which the train runs based on the sensor information received from the sensor unit,
Evaluating the degree of reliability that quantifies the certainty of the primary failure determination result regarding the presence or absence of the failure based on a magnitude relationship compared with a more objective secondary determination criterion,
A control instruction storage unit stores, as a control instruction rule, a combination of the criterion, the evaluation of the reliability, and train control details associated with the evaluation,
The train control unit controls the train speed based on the control instruction rule according to the evaluation;
train control method.