Detailed Description
Various examples of the present disclosure are discussed below. While specific embodiments are discussed, it should be understood that this is done for illustrative purposes only. A person skilled in the relevant art will recognize that other components and configurations may be used without parting from the spirit and scope of the disclosure.
The present disclosure presents an on-board system of a train. An on-board system is a system to be installed on a train and travel with the train. The on-board system includes a sensor system for detecting an object on a railroad track ahead of the train. In some examples, the sensor system may include a camera and a LiDAR (LiDAR) device, while in other examples, the sensor system may include other types of sensors. If the sensor system detects an object on the railroad track, the driver may be notified or the control system of the train activated to apply the brakes, or otherwise adjust the speed of the train. If the sensor system does not detect an object on the railroad track, the speed of the train can be set to a safe speed.
The safe speed refers to a speed at which the train can be stopped by braking before colliding with an obstacle. The safe speed may be based on a detection range of the sensor system. For example, the safe speed may be set such that the braking distance of the train is within the detection range of the sensor system. The detection range of the sensor system is a range in which the sensor system can detect an object. The braking distance of a train refers to the distance the train travels after the on-board system determines that braking is required before the train comes to a stop. The braking distance may be calculated, for example, based on the speed of the train, the braking deceleration of the train, and the reaction time of the driver or the reaction time of the automatic braking system.
The detection range may vary at different locations on the railway depending on the characteristics of the railway track ahead of the train. Thus, in some examples, the detection range may be determined by: the present invention relates to a method for detecting a train position in a railway, and more particularly to a method for detecting a train position in a railway, which includes determining a current position of a train on a digital map of a railway including geometric information of the railway, and calculating a detection range based on a geometry of a railway track ahead of the train.
Fig. 1 illustrates an example method 100 of determining a safe speed of a train.
At block 110, a current location of the train on the railway is determined. For example, the current location may be determined from Global Positioning System (GPS) signals, location signals received wirelessly from trackside devices, or by identifying landmarks detected by sensor systems.
At block 120, a detection range of a sensor system of the train is determined based on a current location of the train and a data store that includes information from which detection ranges at a plurality of locations on the railway may be determined. In one example, the data store is a digital map of the railroad that includes geometric information of the railroad, such as track curvature and grade.
At block 130, a safe speed of the train is determined based on the detection range of the sensor system. For example, the safe speed may be set such that the braking distance of the train is within the detection range of the train, i.e., equal to or less than the detection range of the train.
In this way, an appropriate safe speed of the train can be determined taking into account the characteristics of the railroad track ahead of the train and the resulting detection range of the sensor system at the current location of the train. In some examples, this determination of safe speed may be performed autonomously by the on-board system without input from an external off-train device or system. In such an example, the system does not rely on a central signal server or an external communication network to determine the safe speed of the train.
Fig. 2 shows an example of an on-board system 200 of a train. The on-board system 200 may implement a method for determining a safe speed of a train, such as the method 100 described above with reference to fig. 1.
The in-vehicle system 200 includes: a sensor system 210 for detecting objects on a railroad track ahead of the train, a positioning system 220 for determining a current location of the train, a data store 230, and a speed determination module 240.
The sensor system 210 may include, for example, sensors such as, but not limited to, cameras and LiDAR (LiDAR) devices. In some examples, the camera may be configured to generate a video stream including 2D images of the environment of the train, and the LiDAR device may be configured to generate a 3D point cloud of the environment of the train. The sensor system may be capable of detecting obstacles, such as other trains, pedestrians, vehicles, or obstacles on the track in front of the train. If an obstacle is detected, an alert may be generated to the driver of the train, or the speed of the train may be automatically adjusted, for example, by activating the braking system of the train.
The positioning system 220 is configured to determine a current location of the train on the railway. In one example, the positioning system 220 includes logic for determining a current location of the train based on data from the sensor system 210. In other examples, the positioning system 220 may include a Global Positioning System (GPS) or may wirelessly receive location signals from trackside devices.
The data store 230 stores information from which the detection ranges of the sensor system at various locations on the railroad can be determined. The data store may, for example, include data stored on a non-transitory machine readable medium, such as a hard disk, a solid state drive, read-only memory, or random access memory, among others.
In one example, the data store is a digital map that includes information relating to the geometry of the railroad track, including track grade and track curvature. The detection range may be calculated in real-time based on the geometry of the railway track in one or more upcoming sections of the railway ahead of the current location of the train along the train route. Real-time computation means computation on demand. In this manner, the detection range may be calculated at frequent intervals (e.g., once per second) as the train is operated and the current position of the train changes. In another example, the data store 230 includes a database or look-up table with detection ranges for multiple locations on a railroad so that real-time calculations are not required, but such an approach may require large storage capacity for the on-board system in order to store detection ranges for a large number of locations on the railroad.
The speed determination module 240 is configured to determine a safe speed of the train. The speed determination module may include: a module 242 for receiving a current location of the train from the positioning system, a module 244 for determining a detection range of the sensor system based on the current location of the train and information in the data store, and a module 246 for determining a safe speed of the train based on the determined detection range of the sensor system.
In one example, the speed determination module 240 may be implemented by a processor, such as a Central Processing Unit (CPU) or microprocessor, and a non-transitory storage medium, such as a memory, hard disk, or solid state drive, that stores machine-readable instructions executable by the processor. Thus, it will be understood that modules 242, 244 and 246 may be implemented as machine readable instructions executable by a processor.
Fig. 3 shows an example of a train 310 traveling in a direction from the left to the right of the figure over a section of a railway 300. An on-board system 320 including a sensor system 325 is mounted on the train. An in-vehicle system may be, for example, as shown in FIG. 2 or other examples in this disclosure. The sensor system 325 may be directed toward the front of the train and may have a field of view 330 as shown in fig. 3.
The sensor system 325 has a detection range 340, which is the maximum distance that the on-board system can detect and classify objects on the railroad track. Detection and classification means detecting and classifying objects as a specific object type, e.g. an obstacle such as another train, a pedestrian, a car, a station buffer, etc., and distinguishing these objects from other visual features such as railway tracks, etc. For example, in some examples, the detection range may be a value between 200m and 300m ahead of the train, depending on the design of the sensor system. In the example of fig. 3, the detection range is 240m in front of the train. The obstacle 370 is located 300m forward of the front of the train and is therefore outside the detection range 340 of the sensing system 325. The braking distance of the train is the distance the train will travel if it is decided at this point to apply the brakes.
If the braking distance is greater than the detection range 340, for example greater than 240m in this example, there is a risk that the train may detect the obstacle 370 too late. In such a case, the train may not be able to brake to a stop before a collision with the obstacle occurs. For example, if the braking distance is 260m but the detection range is 240m, the on-board system may detect an obstacle at a distance of 240m, but may not stop the train until 260m later (i.e., 20 m beyond the position of the obstacle). Conversely, if the braking distance is within the detection range, for example 240m or less in this example, the sensing system may detect an obstacle and may stop the train before a collision occurs. Fig. 3 shows the stopping distance 360 within the detection range.
The braking distance of the train depends inter alia on the speed of the train. Thus, the safe speed of the train can be set such that the braking distance of the train is within the detection range of the sensor system. That is, the braking range is less than or equal to the detection range. In this way the train will be able to brake to a stop after an obstacle has been detected on the track and before a collision with the obstacle has occurred.
Fig. 4 illustrates an example method 400 of setting a safe speed of a train based on a detection range in such a way that a braking distance of the train is within the detection range of the sensor system. This method may be employed, for example, by block 130 of fig. 1 or module 246 of fig. 2.
At block 410 of fig. 4, a detection range of the sensor system is determined. This may be done, for example, as described above in block 120 of fig. 1.
At block 420, a first speed at which a braking distance of the train is within the detection range is determined. For example, the braking distance may be set to a value within the detection range, and the maximum speed at which the train has the braking distance may be calculated.
At block 430, the first speed is set to a safe speed for the train.
The braking distance of the train may be calculated based on the speed of the train, the braking deceleration of the train, and the reaction time. The reaction time may be the reaction time of the driver in case the driver is to activate the brake or the reaction time of the on-board system in case the on-board system is to activate the brake automatically. In one example, the braking distance may be calculated according to the following formula:
D=v T+(v 2 /2 a) (Eq.1)
Wherein:
d = the braking distance of the brake pad,
v = the speed of the train(s),
t = reaction time, and
a = braking deceleration of train
In one example, the first speed may be a maximum speed of the train with a braking distance within a detection range. In this case, the braking distance may be set equal to the detection range, and the speed may be found by finding the maximum speed by solving equation 1.
Once the safe speed is determined, the safe speed may be used in various ways. In one example, the safe speed may be displayed to a driver of the train. For example, the safe speed may be displayed on a display panel of the on-board system together with the current speed of the train. The driver of the train can take action accordingly.
In some examples, as shown in the method 500 of fig. 5, the on-board system may determine a current speed of the train at block 510 and compare the current speed of the train to a safe speed at block 520. At block 530, in response to determining that the current speed is greater than the safe speed, an alert may be generated or the speed of the train may be automatically adjusted. For example, the driver may be notified of a visual and/or audible alert through a display panel or speaker in the cab of the train. In some embodiments, the on-board system may automatically adjust the current speed of the train, for example, by activating the train's braking system or reducing the engine speed, etc.
Fig. 6 illustrates a further example of an on-board system 600 for a train in accordance with the present disclosure. The in-vehicle system 600 includes a host computer 620, a display 680, and a plurality of sensors including a camera 610, a LiDAR device 612, and a speed sensor 614.
The camera 610 is configured to generate a video stream of an environment of the train, and may for example generate a video stream comprising a plurality of 2D images. The LiDAR device 612 is configured to generate LiDAR data for the environment of the train and is capable of determining the distance to an object. The LiDAR device may, for example, generate a 3D point cloud of the environment of the train. The cameras 610 and LiDAR devices may be directed toward the front of the train so that they may detect objects on the railroad track in front of the train. In some examples, the cameras 610 and/or LiDAR devices may be rotatable to scan different areas in the environment of the train. The cameras 610 and LiDAR devices 612 may together form a sensor system for detecting objects in front of a train, and may perform functions similar to those of the sensor system 210 of fig. 2.
The speed sensor 614 is configured to determine the speed of the train. The speed sensor may be, for example, an odometer, a gyroscope device, or a device for detecting the rotational speed of an axle of the train, or a radar or LiDAR device. In one example, the speed sensor is a millimeter-wavelength radar device. Radar or LiDAR speed sensing devices may be configured to determine the speed of a train using the doppler effect by: the method includes transmitting radar or LiDAR waves, detecting the radar or LiDAR waves reflected back to the train, and determining a frequency shift between the transmitted and reflected waves. Radar or LiDAR speed sensing devices may be independent of other systems of the train and, therefore, are readily implemented on a variety of different types of trains.
Host computer 620 may include a single computer or multiple computer systems. The host computer may include at least one processor 622, such as a Central Processing Unit (CPU) or microprocessor, etc., and a non-transitory machine readable storage medium 624, such as a hard disk drive, disk array, solid state drive, memory, etc. Storage medium 624 may include a module having machine-readable instructions executable by processor 622 to perform any of the methods described in the present disclosure.
Modules stored on storage medium 624 having machine-readable instructions may include a vision analysis module 630, an integration module 632, an obstacle detection module 634, a position detection module 640, a velocity determination module 650, and a measure module 670. The storage medium 624 may also store a digital map 660 of the railroad. While the modules are shown in fig. 6 as being implemented on a single computer and residing on the same storage medium, it should be understood that in other examples, the modules may be distributed among multiple computers and/or storage media of an in-vehicle system.
The vision analysis module 630 is configured to detect objects in 2D images in the video stream from the camera 610. The visual analysis module may detect and classify objects, for example, using machine learning. For example, the visual analysis may be trained to recognize portions of the image corresponding to railroad track and detect any obstructions on the track. The visual analysis may be trained to classify the obstacle by type, such as another train, a station buffer, a pedestrian, a car, or other foreign object.
The integration module 632 is configured to integrate data from the visual analysis module with 3D point cloud data from the LiDAR device. For example, objects detected and classified by visual analysis may be mapped to features in the point cloud data. In this manner, the distance to the detected object may be determined because the point cloud includes 3-dimensional information and is able to determine the distance based on the time of flight of the transmitted and received LiDAR pulses. The point cloud may also include information about finer features of the detected object. By combining visual analysis with the 3D point cloud, the integration module can form a digital model of the environment of the train. The obstacle detection module 634 is configured to determine whether an obstacle is present on the track according to the digital model. The obstacle detection module 634 may be configured to: the boundary of the track area in front of the train is determined and then obstacles within the track area are detected while obstacles outside the track area are ignored. The obstacle detection module may be configured to determine whether the obstacle is likely to collide with the train based on the position of the obstacle, the distance to the obstacle, and the speed and direction of travel of the obstacle (if the obstacle is moving). If there is a risk of collision, the obstacle detection module may notify the action module 670 so that appropriate actions may be taken, such as notifying the driver, generating an alarm, or automatically adjusting the speed of the train.
The digital map 660 includes information from which the detection ranges of the sensor system at various locations on the railroad can be determined. For example, the digital map may include geometric data of the track, including track grade, track curvature (e.g., radius of a curve), tunnel width, and whether various sections of the track are straight, curved, horizontal, inclined upward or downward. The digital map may also include image data, data relating to the shape of the track, inter-station distance, trackside infrastructure at a particular location, and the like. The digital map may also include information about landmarks such as stations or other trackside infrastructure, including identifying features and known locations of stations.
By way of non-limiting example, the digital map may include one or more of the following:
a) Image data of landmarks
b) Point cloud data of landmarks
c) Information about station
-map of stations
The length of each section
d) Information about switches
Position of the switch
Type of switch
e) Track section information
Straight track (length of straight section)
Curved track (radius of curve, location of entrance of curved track section and location of exit of curved track section)
Inclined track (whether the slope is concave or convex, slope gradient, entry and exit positions of inclined track sections, junction between sections of different gradients and/or characteristics of nominal slope approximating the actual slope)
Whether the sections of track are open or tunneled
The position detection module 640 is configured to determine the position of the train. In some examples, the location detection module may determine the location based on receiving GPS signals or wireless location signals from trackside devices. In other examples, the location detection module may be configured to match the digital model of the environment of the train generated by the integration module 632 to a location on a digital map. For example, the location detection module may determine the location of the train based on identifying locations on a digital map that have features that match features detected by the cameras and LiDAR devices. For example, the location detection module 640 may determine the location of a train by comparing landmarks in a digital map to landmarks detected by cameras and LiDAR devices. A further example of such a method of determining the location of a train using landmarks is depicted in fig. 10 and 11.
The speed determination module 650 is configured to determine a speed of the train. The speed determination module 650 includes: a module 652 for receiving a current location of the train from the location detection module 640, a module 654 for determining a detection range of the sensor system at the current location of the train, and a module 656 for determining a safe speed of the train based on the detection range. Modules 650, 652, 654 and 656 may perform the same functions as modules 240, 242, 244 and 246 of fig. 2 and may perform the method of fig. 1 to determine a safe speed of the train.
The on-board system includes a display 680 that can display information about the train and the train environment to assist the driver. The display may be part of a user interface through which the driver may control the train. The display may display the current speed of the train and the safe speed of the train determined by the speed determination module. The display may further display and/or highlight the obstacle detected by the obstacle detection module. For example, the display may highlight a car stop, train, or pedestrian when the obstacle detection module 634 detects a corresponding obstacle.
The action module 670 is configured to determine whether an action needs to be taken based on the output of the speed determination module 650 and/or the object detection module 634. The action module 670 may compare the safe speed determined by the speed determination module 650 to the current speed of the train determined by the speed sensor 614. If the current speed is greater than the safe speed, the action module 670 may generate an alert or automatically adjust the current speed of the train. In some examples, the action module 670 may generate an alert or automatically adjust the current speed of the train if an obstacle (e.g., other trains, vehicles, pedestrians, and car stops, etc.) is detected in front of the train.
For example, the alert may be displayed on the display 680 or user interface of the in-vehicle system. The alert may indicate that the speed of the train should be reduced. Instead of generating an alert, or if the alert is not active within a certain period of time, the action module may automatically adjust the speed of the train by sending a command to the control system of the train to brake the train or adjust the engine revolutions of the train.
The obstacle detection module 634 may notify the action module 670 of the risk of collision so that the action module may take any of the actions described above and/or cause the detected obstacle to be displayed on the display 680.
The on-board system may further include a communication interface 690 for wirelessly communicating with a remote computer at a control center of the railroad or for sending communications to other trains on the railroad. The communication interface may, for example, send information (such as the location of the train, the speed or status of the train) and/or alerts over a telecommunications network, such as a 3G, 4G or 5G network.
As discussed above, the detection range is determined based on characteristics of the track ahead of the train. For example, determining the detection range of the train may include: the calculation is performed in real time based on the geometry of the railway track on the upcoming section of the railway ahead of the current location of the train along the train route. The geometry of the railway track can be found by reference to a digital map. The railway track may be divided into sections, each section being straight, curved or inclined. One, two or more upcoming segments may be considered when determining the detection range. An upcoming section may be defined as a section within a predetermined length of railroad track ahead of the current location of the train. The predetermined length may be a length equal to the maximum detection range when the front rail is straight and horizontal, for example, 240m. Since the track ahead of the train may depend on the route of the train in the presence of a junction or a turnout, the digital map may store the route of the train so that an upcoming section may be determined.
The detection range will depend on whether the track in front is straight or curved, horizontal or inclined. When the upcoming segment includes only a straight segment, the detection range will be longer than if the upcoming segment includes a curved segment or an inclined segment. Therefore, the calculation of the detection range may include: it is determined from the digital map whether the upcoming section of the railroad comprises a straight, curved, horizontal, upwardly inclined or downwardly inclined section.
FIG. 7 illustrates an example method 700 for determining a detection range of a sensor system.
At block 710, it is determined whether the upcoming segment includes a curved segment or an inclined segment.
At block 720, in response to determining that the upcoming segment does not include a curved segment or a sloped segment, the detection range is set to a first predetermined distance X. X is the maximum detection range of the sensor system. In one example, X may be set to 240m.
At block 730, in response to determining that the upcoming segment comprises a curved segment, a detection range is calculated based at least in part on a radius of the curved segment. If the curved section is in a tunnel rather than open air, the detection range may depend on the width of the tunnel and the radius of the curve.
At block 740, in response to determining that the upcoming segment comprises a sloped segment, a detection range is calculated based at least in part on a gradient of the sloped segment. If the sloped segment is in a tunnel rather than open air, the detection range may depend on the height of the tunnel and the gradient of the sloped segment.
In some cases, two or more consecutive track sections (e.g., curved section-straight section-curved section) may be considered to allow for smoothing of the detection range curve so that the detection range does not jump when a train enters or leaves a curved section or a sloped section of the railway track. Otherwise, if the detection range is abruptly changed, for example, from 240m to 150m when entering a curved section or from 150m to 240m when leaving a curved section, the safety speed is also abruptly changed, which may cause sudden braking or sudden acceleration of the train, which may be an undesirable result.
Thus, calculating the detection range may include: the change in detection range is smoothed such that the detection range gradually decreases from a first predetermined distance X to a second predetermined distance Y, where X is the detection range for a straight section of the railway track and Y is the minimum detection range for a curved section or a slanted section of the railway track, and gradually increases from the second predetermined distance Y to the first predetermined distance X.
An example of the way in which detection ranges for different sections of the track can be calculated will now be described.
Straight track section
When the forward section is straight (e.g., as shown in fig. 3), the detection range will not be limited by the track, but rather by the inherent detection capability of the sensor system. Thus, the detection range is X, which is the maximum distance that the sensor system can reliably detect an obstacle on the railway track, and can be determined by testing the system. To prioritize security, a conservative value may be chosen for X. Since the value X depends on the characteristics of the train, the value may be set to the first predetermined distance stored in the on-board system. The value of X can be determined by testing the on-board system on the train and the railroad track. The value X may be set and stored in the on-board system as part of the on-board system setup and calibration. In one example, X is a value between 200m and 300 m. In one example, X equals 240m.
Curved track section
Fig. 8A shows an example of a train T on a portion of a railway 800. The upcoming section of the railroad ahead of the train includes a straight section 810 and a curved section 820. Position a is at some point before the entrance 815 of the curved section, position B is inside the curved section, and position C is at the exit of the curved section. When the train is on a straight section 810 of track and X or more distance ahead of a curved section 820, then the detection range will be equal to X. When the train is less than X from the curved section, then the detection range will be less than X because it will be reduced by the curved section.
Figure 8B shows an example where the train T is in a curved section 820 of a tunnel (e.g., at a location such as location B in figure 8A). It can be seen that the field of view 830 of the sensor system of the train is reduced due to the curvature of the tunnel. Thus, the detection range of the sensor system is reduced and the sensor system cannot detect an obstacle (such as the second train T2) that is less than X meters away because the second train T2 is out of the field of view of the sensor system.
Fig. 8C shows a curved section 820 of a railway tunnel with a radius R and a tunnel width DI. Assuming that the train is in the middle of the track and the radius of curvature is measured from the origin O at the center of curvature to the inner edge of the curved section, the distance of the train from the origin O is R-DI/2. Therefore, the detection range at position B inside the curved track section can be calculated according to the following formula:
wherein:
-R is the radius of the curved track section
DI is the width of the tunnel
The value Y may be regarded as the minimum value of the detection range in the curved section. For example, it is applicable at the entrance 815 of the bend section, and may be applicable at position B in the middle of the bend section as shown in fig. 8C.
Figure 8D shows the train at position a on the straight section 810 but approaching the entrance 815 to the curved section 820. In this case, the detection range Y' at the position a depends on the distance DS of the train from the entrance 815 of the curve section and the minimum detection range Y in the curve section.
Therefore, the detection range Y' at the position a near the entrance of the curved section can be calculated according to the following formula:
wherein:
DS is the distance between the train and the entrance of the curved section;
-R is the radius of the curved section;
DI is the width of the tunnel
A similar calculation may be applied when the train is inside the curved section but approaching the exit of the curved section (e.g., beyond the midpoint).
With the above calculation method, the detection range curve can be visualized as shown in fig. 8E. It can be seen that the detection range gradually decreases from X to Y at a position near the beginning of the curved section and gradually increases from Y to X at a position near the end of the curved section; where Y is the minimum detection range for the curved section.
In summary, when the train is approaching the curve section, assuming that the maximum detection range of the sensor system is X, the detection range is set to X when the distance between the train and the entrance of the curve section is greater than-Y/2. When the distance between the train and the entrance of the curved section is less than X-Y/2, the detection range is changed based on the above equation 3. The calculation method can also be applied to the case where there are continuous curved track sections.
The above calculations assume that the curved section of track is in a tunnel of width DI. If the curved track is in a surface section, the detection range will theoretically be longer, since there is no line-of-sight obstruction of the sensor system by the tunnel walls. However, the above equation can also be used for open sections of track when conservative methods are used. For example, a nominal tunnel width based on the width of the track may be assigned to the open-air section of the track, or the same nominal tunnel width as the tunnel width of the railway section with the tunnel may be assigned. This method is safe because it will determine for the open air segment a detection range that is the same as or smaller than the actual detection range for the open air segment.
Convex slope track section
In determining the detection range, the relevant consideration is not the absolute gradient or gradient, but whether the gradient of the current section of the track is different from the gradient of the next section of the track. This reduces the detection range when there is a change in the gradient between sections of the track, as the ground of the track (and also the top wall if in a tunnel) may obstruct the field of view of the sensor system. In this regard, there are two types of ramps: a convex ramp and a concave ramp. Thus, the onboard system may first determine whether the upcoming incline segment is a convex or concave ramp, and then determine the detection range for that incline segment.
Fig. 9A shows a length of track 900 comprising a horizontal section 910 of the track followed by an inclined section 920 forming a convex ramp. The solid lines show the top wall and the floor of the tunnel. As can be seen, in this example, the sloped section 920 includes a first section 920A and a second section 920B having different gradients. However, for modeling purposes, these may be converted to a nominal ramp profile shown in dashed lines, which includes a horizontal portion 921A and an incline or decline portion 921B. Thus, the nominal ramp curve includes two portions: a flat first portion 921A and a second portion 921B corresponding to the last portion of the ramp or the steepest gradient. Point 925 is defined as the intersection between the first and second portions of the nominal ramp curve, and is referred to hereinafter as the intersection of the sloped sections. The figure shows the train in four possible positions T1, T2, T3 and T4. The sensing system may have a field of view that covers a range of angles (e.g., 20 degrees on each side of a horizontal plane). The dotted line shows the longest line of sight of the sensing system within the field of view.
The train at location T1 is a first distance Z before the junction 925 and the train at a second location T2 is a second distance W before the junction 925. The train at location T3 is a third distance T after the intersection 925. Since the line of sight of the sensor system is blocked by the ramp, the detection range is at a minimum at a location between Z meters and W meters (i.e., between locations T1 and T2 shown in fig. 9A) before the intersection 925 of the inclined section. In the following discussion, the coordinate of the intersection 925 of the sloped section is taken to be 0, so, for example, the coordinate at a distance Z before the intersection 925 is-Z.
Therefore, the detection range may be set to the minimum value Y at a position between the first distance (-Z) and the second distance (-W) in front of the intersection point of the inclined sections. At a position before-Z and after-W, the detection range may gradually increase to the maximum value X.
The values Z, W and T may be calculated based on the characteristics of the oblique section. In one example, values Z, W, and T may be calculated according to the following formulas:
z ≈ 4.5/relative gradient
Wherein:
z, W and T depend on the steepness or gradient of the ramp. The larger the steepness, the smaller will be Z, W and T, and vice versa.
H is the height of the train relative to the railway track
HC is the height of the sensor system relative to the railway track
DH is the height of the tunnel
The relative gradient is a variable that depends on the characteristics of the ramp, such as the gradient or steepness of the ramp.
The relative grade value may be unique for each inclined track segment and may be determined based on field testing of the detection range at the grade.
In one example, the detection range may be calculated from the location of the intersection 925 of the train relative to the incline zone according to the following three scenarios:
scene 1:
-if the train is at a distance between-X + T and-Z from the intersection of the ramps, the detection range will gradually decrease from X to Z + T, given by the following equation:
detection range = -1 x + T (equation 5)
Wherein:
-X is the distance of the train from the intersection of the inclined sections, and-X + T ≦ X ≦ -Z
Scene 2:
-if the train is at a distance between-Z and-W from the intersection of the ramp, the detection range is given by the following equation:
detection range = Z + T (equation 6)
This is the minimum value of the detection range for the above-mentioned ramp, and may be represented by the variable Y. Therefore, in this example, the minimum value Y of the detection range of the convex slope is equal to Z + T.
Scene 3:
if the distance of the train exceeds position-W (i.e. less than W before or on the slope), the detection range is not limited by the slope track section, but by the next track section. If there is no factor limiting the line of sight on the next track segment, the theoretical detection range will be X. However, for the purpose of smoothing and to avoid abrupt changes, the detection range beyond (-W) may be gradually increased from Z + T to X. For smoothing purposes, the rate of increase may be limited to a certain value, for example not exceeding 22 meters per second. In some cases, if there is a limit to the detection range due to the characteristics of the next track section (such as an additional slope or curve), the detection range may be increased to a value less than X.
A visual example of the curve of the detection range at a position near and in the convex slanting section is shown in fig. 9B. It can be seen that the detection range decreases from X to Z + T as the train travels from a distance X-T ahead of the intersection of the incline section to a distance Z ahead of the intersection of the incline section. The detection range is then maintained at Z + T until the train reaches a distance W in front of the intersection of the incline section. After exceeding the distance W in front of the intersection point of the slanted section, the detection range gradually increases back to X or a value determined by the characteristics of the next track section in front of the current slanted section.
Although fig. 9A and 9B show a convex ramp that slopes upward, the same method and equation can also be used for a convex ramp that slopes downward.
Concave track section
Fig. 9C shows an example where an inclined section 930 of track and a subsequent horizontal section 940 of track form a concave ramp 901. The solid lines show the top wall and the floor of the tunnel. As can be seen, in this example, the sloped section 930 includes a first section 930A and a second section 930B having different gradients. However, for modeling purposes, these may be converted to a nominal curve shown in dashed lines having a horizontal portion 931B and an inclined or declined portion 931A. Intersection 935 is defined as the point of connection between the two parts. The train is shown at a first location T1 and a subsequent second location T2, wherein the dash-dot lines indicate the line of sight of the sensor system of the train at the respective locations. It will be understood that the detection range at the position T1 is smaller than the detection range at the position T2.
In the following discussion, the connection point or intersection 935 between the first and second portions of the nominal ramp curve is considered to have a 0 coordinate. The in-vehicle system may define a first distance Z and a second distance W forward of the intersection. At a position between-Z and-W, the detection range will have a minimum value of Y. At a position before-Z and after-W, the detection range may gradually increase to X (which is the detection range for a horizontal straight section of the track).
The first distance Z and the second distance W may be defined as follows:
z ≈ 4.5/relative gradient
Wherein:
z, W depend on the gradient or steepness of the slope. The larger the steepness is, the smaller Z, W and vice versa.
-H is the height of the train
HC is the height of the sensor system relative to the railway track
DH is the height of the tunnel
The relative gradient is a variable that depends on the characteristics of the ramp, such as the gradient or steepness of the ramp.
The relative grade value may be unique for each inclined track segment and may be determined based on field testing of the detection range at the grade.
In one example, the detection range may be calculated based on the location of the train relative to the junction 935 according to the following three scenarios.
Scene 1:
-if the train is within the range (-X, -Z), the detection range will gradually decrease from X to Z, the detection range = -1X
Wherein:
x is the distance of the train from the connection point
--X≤x≤-Z
Scene 2:
-if the distance of the train is within the range (-Z, -W), then
Detection range = Z
This is the minimum value of the detection range for the above-mentioned ramp, and may be represented by the variable Y. Therefore, in this example, the minimum value Y of the detection range for the concave slope is equal to Z.
Scene 3:
if the distance of the train is in the range exceeding (-W), the detection range is not limited by the ramp track section but by the next track section. If there is no factor limiting the line of sight on the next track section, the theoretical detection range will be X. However, for the purpose of smoothing and in order to avoid abrupt changes, the detection range beyond (-W) may be gradually increased from Z to X. For smoothing purposes, the rate of increase may be limited to a certain value, for example not exceeding 22 meters per second. In some cases, if there is a limit to the detection range due to the characteristics of the next track section (such as an additional slope or curve), the detection range may be increased to a value less than X.
Although fig. 9C shows an upwardly inclined concave ramp, the same method and equations may be used for a downwardly inclined concave ramp.
Although fig. 9A to 9C show one example of a convex slope and one example of a concave slope, the same method may be applied to other convex slopes or concave slopes having different curves. In general, the actual ramp profile can be converted into a nominal ramp profile comprising the following two parts: a first portion having a gradient corresponding to the gradient of the first portion of the ramp; and a second portion having a gradient corresponding to the gradient of the last portion of the ramp. For example, fig. 12A, 12B, and 12C show an example of an actual ramp 1200 in solid lines and an example of a corresponding nominal ramp including a first portion 1210 and a second portion 1220 in dashed lines. The above calculation of the detection range may be applied by calculating T, W and Z based on the gradient of the second portion 1220 of the ramp and the location of the intersection 1215 between the first and second portions of the nominal ramp.
The above calculation assumes that the inclined section of track is in a tunnel of height DH. If the inclined track is in an open air section, the detection range will theoretically be longer, since the tunnel ceiling or the ground will not cause a sight obstruction. However, the above equation can also be applied to open sections of track when a conservative approach is used. For example, a nominal tunnel height may be assigned to a surface segment of track, e.g., using the same tunnel height as a railway segment with a tunnel. This method is safe because it will determine for the open air segment a detection range that is the same as or smaller than the actual detection range for the open air segment.
Setting a safe speed based on a detected obstacle
When a sensor system of an on-board system of a train detects an obstacle (e.g., a pedestrian or another train) on a track in front of the train, the distance between the object and the train may be obtained by the sensor system (e.g., by using LiDAR). In this case, the action module 670 may generate an alert to the driver of the train or take automatic action to adjust the speed of the train. In some examples, the on-board system may determine a new safe speed for the train based on the detected obstacle. Thus, when no obstacle is detected, the safe speed may be determined based on a detection range derived from a digital map or data store; and when an obstacle is detected on the railway track in front of the train, the safe speed may be determined based on the detected obstacle.
One method for determining a new speed for the train based on the detected obstacle is to redefine the detection range as the distance of the obstacle from the train. In this case, the method of fig. 4 and the adjusted detection range may be used to determine a new speed.
Train position detection based on sensor system
In some examples, the positioning system may include a GPS receiver for receiving GPS location signals. However, GPS signals cannot be received in tunnels and underground portions of railways. In other examples, the positioning system may include a wireless device (such as an RFID reader) for receiving location signals from trackside devices. However, installing trackside equipment along the entire length of a railway can be expensive. Further, even if trackside equipment has been installed, there will still be a need to adapt the on-board system to be compatible with such trackside equipment, which may be rail-specific.
Accordingly, one aspect of the present disclosure proposes determining a current location of a train based on information from a sensor system. For example, a landmark detected by a camera and LiDAR device may be matched to a landmark having a known position, and the current position determined based on the known position of the landmark, the speed of the train, and the period of time elapsed since the landmark was detected. In this manner, the on-board system does not rely on external systems (such as GPS or trackside equipment) to determine the location of the train. This method can also work underground and in tunnels and does not require expensive trackside equipment installation.
Fig. 10 illustrates a method 1000 of determining a location of a train on a railway.
At block 1010, a video stream of an environment of the train is generated by a camera installed on the train.
At block 1020, liDAR data for an environment of a train is generated by a LiDAR (LiDAR) system installed on the train.
At block 1030, a digital model of the environment of the train is generated based on the video stream and the LiDAR data.
At block 1040, a landmark is detected in the digital model, the landmark having identifying features corresponding to identifying features of landmarks having known locations in a digital map of the railway.
The landmark may be, for example, a particular station or arrangement of trackside infrastructure. Each landmark has unique identifying features such as the arrangement and shape of various trackside infrastructures, the length of the station platform, the size and shape of the buildings or other infrastructures near the platform, and the like. The images, liDAR data, and/or identifying features of the landmarks, as well as the locations of the landmarks, may be stored in a digital map so that machine learning may be used by the in-vehicle system to identify the landmarks.
At block 1050, the speed of the train is determined.
For example, the speed of the train may be determined by using an odometer, a gyroscope device, or a radar or LiDAR speed sensor. If a radar or LiDAR speed sensor is used, it may be independent of the train's existing control system, which may simplify the installation of the on-board system. This makes the on-board system highly adaptable and easier to migrate to different types of trains.
At block 1060, a current location of the train is determined based on the known location of the landmark, the speed of the train, and the time elapsed since the landmark was detected.
The known locations of landmarks may be stored in a digital map. Thus, at the point in time when a landmark is detected, the known location of the train near the landmark can be determined. In some examples, the location of the train may be further refined based on LiDAR data that indicates the distance of the train from a landmark. If the train continues to move after the landmark is detected, the location may be updated based on the speed of the train as determined in block 1050 and the time elapsed since the landmark was detected.
The method may further comprise: a second landmark is detected in the digital model, and a position calibration is performed in response to detecting the second landmark. In this way, the position can be corrected based on detecting the second landmark at a later time after the first landmark is detected.
In some examples, the position calibration may include: a discrepancy between the determined current position and the known position of the second landmark is detected, and the determination of the current position is adjusted to correct the discrepancy.
For example, the position calibration may include resetting the current position to a known position of the second landmark (which may be stored in a digital map). In this way, the position of the train can be reset each time a landmark is passed so that no cumulative error is introduced into the position calculation.
Generating a digital model of the environment of the train in block 1030 may include: information from a visual analysis system that detects objects in 2D images in a video stream is combined with information about the objects from a 3D point cloud generated based on LiDAR data.
FIG. 11 shows an example of an in-vehicle system 1100 for performing the method of FIG. 10.
The on-board system includes cameras 1110 for generating a video stream of the environment of the train, and LiDAR means 1120 for generating LiDAR data for the environment of the train.
The video stream may include a plurality of 2D images, and the LiDAR data may include a 3D point cloud. The in-vehicle system may include a vision analysis system that uses machine learning to detect objects in 2D images in the video stream, and may be configured to combine the vision analysis with information about the objects from a 3D point cloud generated from LiDAR data to form a digital model of the environment of the train.
The in-vehicle system further includes a digital map 1130 that includes identifying characteristics 1132A of a plurality of landmarks 1132 of the railway and known locations 1132B of the plurality of landmarks. The landmark may be, for example, a particular station or a particular arrangement of trackside infrastructure. Each landmark has unique identifying features such as the arrangement and shape of various trackside infrastructures, the length of the station platform, the size and shape of the buildings or other infrastructures near the platform, and the like. The images, liDAR data, and/or identifying features of the landmarks, as well as the locations of the landmarks, may be stored in a digital map so that machine learning may be used by the in-vehicle system to identify the landmarks. The digital map may have any of the features of the digital maps and data stores described in the present disclosure.
The in-vehicle system further includes a landmark detection module 1140 configured to detect landmarks in the video stream and LiDAR data having features that match the identifying features of landmarks in the digital map. The landmark detection module may use machine learning to detect 1142 landmarks in the data from the camera and LiDAR and match 1144 the detected landmarks in the digital map. The landmark detection module may then determine 1146 the location of the landmark from the digital map.
The on-board system further includes a speed determining device 1150 for determining the speed of the train. For example, the speed determining means may be an odometer, a gyroscope device, or a radar or LiDAR speed sensor.
The on-board system further includes a position determination module 1160 for determining the current location of the train based on the known location of the detected landmark, the speed of the train, and the elapsed time since the landmark was detected. The speed determination module may, for example, perform the process described above with respect to block 1060 of fig. 10.
The in-vehicle system may include one or more non-transitory machine-readable storage media, such as a hard disk drive, a solid state drive, a disk array, read-only memory, or random access memory, etc., and one or more processors, such as a Central Processing Unit (CPU), microprocessor, etc. The digital map may be stored on a non-transitory machine-readable storage medium. The modules 1140 and 1160 may be implemented as machine-readable instructions stored on a non-transitory machine-readable storage medium and executable by one or more processors of an in-vehicle system.
The on-board system of fig. 11 may have any of the features of the on-board system described above with respect to fig. 1-9, and may be configured to perform any of the methods described herein, including those described with reference to fig. 1-10.
Thus, the on-board systems and methods described herein may be readily modified for use with a variety of different trains and railways. This is particularly true where the speed sensor is a LiDAR or radar sensor and is not tied to an existing train system. In some examples, the on-board system may be independent of any existing signal system or monitoring system of the train or railway. When introducing an on-board system into a new railway system, a digital map of the new railway can be installed on the on-board system. When introducing an on-board system to a new train, certain parameters, such as brake deceleration, may be set by testing the train or based on the specified values of the train manufacturer or the railway operator.
In some examples, the on-board system may provide a display panel for a driver of the train to inform the driver of the determined safe speed of the train, to alert the driver of any detected obstacles, and/or to generate an alert if adjustment of the speed of the train is required. In this case, the on-board system does not need to interface with the control system of the train. In other examples, the on-board system may interface with a control system of the train to facilitate automatic braking and/or automatic driving of the train. In still other examples, the on-board system may both provide a display to the driver and interface with the control system of the train so that the train may be driven in a manual mode, an assisted and/or an autonomous mode depending on the circumstances or the driver's selection.
In the above examples, in some cases, the present techniques have been presented as including functional blocks that include the following: these functional blocks include devices, device components, blocks or routines in methods implemented in software or a combination of hardware and software.
The methods according to the examples described above may be implemented using computer-executable instructions stored on or otherwise available from computer-readable media. Such instructions may include, for example, instructions and data which cause or otherwise configure a general purpose computer, special purpose computer, or special purpose processing device to perform a certain function or group of functions. Part of the computer resources used may be accessible via a network. The computer-executable instructions may be, for example, binaries, intermediate format instructions, such as assembly language, firmware, or source code. Examples of computer readable media that may be used to store instructions, information used, and/or information created during a method according to the described examples include magnetic or optical disks, flash memory, universal Serial Bus (USB) devices provided with non-volatile memory, networked storage devices, and so forth.
An apparatus implementing methods in accordance with these disclosures may include hardware, firmware, and/or software, and may take any of a variety of forms. Typical examples of such forms include laptop computers, smart phones, small form factor personal computers, personal digital assistants, and the like. The functionality described herein may also be implemented in a peripheral device or add-on card. As another example, such functionality may also be implemented on circuit boards of different chips, or may be implemented on different processes executing in a single device.
Instructions, media for communicating such instructions, computing resources for performing them, and other structures for supporting such computing resources are means for providing the functionality described in these disclosures.
Although various examples and other information may be used to describe aspects within the scope of the appended claims, no limitation to the claims should be implied based on the particular features or arrangements in such examples because one of ordinary skill in the art would be able to derive numerous embodiments using these examples. Further, and although certain subject matter may have been described in language specific to examples of structural features and/or methodological steps, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the described features or acts. For example, such functionality may be distributed differently or performed in components other than those identified herein. Rather, the above-described embodiments are described by way of example only, and many variations are possible without departing from the scope of the present disclosure as defined in the appended claims.