Auxiliary driving system and method for rail train based on three-dimensional modeling

Abstract

The present application discloses an assisted driving system and method for a rail train based on three-dimensional modeling. An assisted driving system and method for a rail train based on three-dimensional modeling, including laser radars, multiple of which are arranged along both sides of the track, and the maximum detection ranges of two adjacent laser radars on the same side of the track intersect in the extension direction of the track; multiple cameras, multiple of which are arranged along both sides of the track, and the field of view of two adjacent cameras on the same side of the track intersect in the extension direction of the track; a three-dimensional model construction device, which is connected to the laser radar and camera signals respectively, and constructs the original three-dimensional model of both sides of the track based on the point cloud data and image data uploaded by the laser radar and the camera; an early warning device, which is arranged on a train for assisted driving and is connected to the three-dimensional model construction device signal. The beneficial effect of the present application is that it can assist the driver in driving the train and reduce safety accidents.

Description

Technical Field

[0001] The present application relates to the technical field of rail trains, and in particular to an auxiliary driving system and method for rail trains based on three-dimensional modeling. Background Art

[0002] The current rail trains have fixed running routes, fixed running stations and specified running speeds, so the running safety and reliability of rail trains are high. Because rail trains have fast running speeds, large mass and large inertia, the response time of rail trains to emergencies is longer than that of ordinary cars.

[0003] In order to ensure the safety of rail trains, rail trains are usually set in a closed environment to prevent outside vehicles from entering the track. As traffic conditions become more and more congested, the closed level of rail transportation is actually decreasing, and some private vehicles and wild animals may suddenly enter the track. In addition, in harsh natural environments, there are also sudden natural conditions such as road collapse and landslides.

[0004] If an object that suddenly enters the track is not discovered in time, it will cause a very serious traffic accident. Therefore, when driving a rail train, the driver needs to concentrate and observe whether there are any emergencies in the field of vision. If an emergency occurs, emergency braking should be performed. This method has great defects in rail train driving. The driver's field of vision will be affected by factors such as rainy weather, terrain obstruction, and haze. Therefore, in practice, it is easy for the driver to discover the emergency when the train cannot slow down, which leads to traffic accidents. Summary of the invention

[0005] The content of this application is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this application is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0006] As a first aspect of the present application, in order to solve the technical problem that the driver cannot detect the emergency situation in time, the present application provides an auxiliary driving system of a rail train based on three-dimensional modeling, including:

[0007] There are multiple laser radars arranged along both sides of the track, and the maximum detection ranges of two adjacent laser radars located on the same side of the track intersect in the extension direction of the track;

[0008] A plurality of cameras are arranged along both sides of the track, and the field of view of two adjacent cameras located on the same side of the track intersects in the extension direction of the track;

[0009] A 3D model building device is connected to the laser radar and camera signals respectively, and builds the original 3D model of both sides of the track based on the point cloud data and image data uploaded by the laser radar and camera;

[0010] An early warning device, which is installed on the train for assisting driving and is connected to the three-dimensional model building device by signal;

[0011] The early warning device updates and displays the three-dimensional models of both sides of the track in real time.

[0012] In the technical solution provided by the present application, a complete three-dimensional model of both sides of the track is constructed by setting a large number of laser radars and cameras on both sides of the track. Therefore, when the driver is driving the train, in the case of low visibility and small field of view, the three-dimensional model provided by the early warning device can be used as an auxiliary reference, so that when an emergency occurs on the track, the driver can promptly discover and take relief measures. In addition, compared with the solution of setting the laser radar and camera on the train, on the one hand, the influence of the obstruction of the object and the detection accuracy on the distant objects can be avoided, and on the other hand, the detection accuracy will not be affected by the movement and vibration of the train during driving, so as to ensure the accuracy of the three-dimensional model. In addition, although the cost of setting a large number of laser radars and cameras on the track is relatively high, all trains running on the railway can play an early warning role. Therefore, compared with the large number of trains running on the track, the average cost of this solution is not high.

[0013] Furthermore, the laser radar is used to obtain point cloud data of areas on both sides of the track, and the camera is used to obtain image data on both sides of the track.

[0014] In this solution, laser radar and camera are used to obtain point cloud data and image data respectively, so the point cloud data can be fused with the image data to construct a high-precision three-dimensional model.

[0015] In the rail transportation network, there are a lot of tracks and a lot of trains. If the early warning device needs to receive the three-dimensional model information of all tracks, it will result in a large amount of information received and a long time to build the three-dimensional model, which will reduce the timeliness of the early warning. To solve this problem, this application provides the following technical solutions:

[0016] The three-dimensional model building device includes:

[0017] An information collection unit connected to each laser radar and camera signal;

[0018] A track model building unit is provided, wherein a plurality of track model building units are provided, each track corresponds to a track model building unit, and each track model building unit is used to build a three-dimensional model of the corresponding track;

[0019] The information distribution unit is connected to the signals of each early warning device. The early warning device is used to upload the track of the train it is on to the information distribution unit. The information distribution unit sends the corresponding three-dimensional model to the early warning module according to the track where the train is located.

[0020] In the technical solution provided by the present application, each track model construction unit is used to construct a three-dimensional model of the track corresponding to itself. In this way, the information collected by the information collection unit can be forwarded to the corresponding track model construction unit in an orderly manner according to the track it is in, thereby increasing the efficiency of information distribution. At the same time, for the early warning module, by sending the track it needs to enter to the information distribution unit, it can directly match the corresponding track model construction unit, so that the early warning module does not need to collect the three-dimensional model information of all tracks, which reduces the amount of information transmission and increases the timeliness of the early warning.

[0021] When the train is running, it is necessary to provide the driver with sufficient information so that he can make an early warning as soon as possible. Therefore, in practice, it is necessary not only to obtain the three-dimensional information of all objects on the track, but actually to obtain the three-dimensional information of all objects on both sides of the track and on the track. For example, there is an object in front of the train moving toward the track, but it has not yet reached the track. If the object is shown to the driver after it reaches the track, the distance between the object and the train may be less than the shortest braking distance of the train. Therefore, in order to increase the timeliness of the early warning, it is necessary to send the three-dimensional object information on both sides of the track to the early warning module, thereby playing an early warning role. Under this scheme, the scope that the early warning module needs to pay attention to will be too large, and the amount of information that needs to be received and processed will increase, which will reduce the efficiency of establishing the three-dimensional model and reduce the timeliness of the early warning. In response to this problem, the present application provides the following technical solutions:

[0022] The early warning module comprises:

[0023] A speed acquisition unit, used to acquire the running speed and the shortest stopping distance of the train, and send the running speed and the shortest stopping distance to the information distribution unit;

[0024] A model information acquisition unit, signal-connected to the information distribution unit;

[0025] The model information acquisition unit acquires all three-dimensional model information of the track before the train enters the corresponding track, and constructs a static three-dimensional model;

[0026] An information distribution unit obtains the travel speed and current position of the train in real time, obtains the observation range of the train according to the travel speed and current position of the train, and sends update information of changes in the observation range compared with the static three-dimensional model to the model information acquisition unit;

[0027] The height of the observation range relative to the ground is a fixed value. The horizontal distance between any point L on the edge of the observation range and the center line of the train is L1. The distance between the current position of the train and the observation range along the train's forward route is L2. L1 = (V1 / L0) V2; L2 = S v +λS0V1, where V1 is the current train speed, V2 is the average moving speed of the moving objects around the track, L0 is the length from the train’s forward direction to the plane where the train is located, S v is the shortest braking distance at the current speed of the train, S0 is the preset compensation distance, λ is the compensation coefficient, 0<λ<1.

[0028] In the solution provided by the present application, the observation range is related to the speed of the train and the shortest braking distance. If the speed of the train is fast and the braking distance is long, the corresponding observation range will be farther, and if the speed of the train is slow and the shortest braking distance is short, the observation range will be relatively shorter. At the same time, for trains with different speeds, the faster the speed, the larger L2, and the longer the length of the observation range required, but the same L1 will be smaller, so in practice, the observation range of trains with different speeds and different masses will be dynamically adjusted to ensure the safety of train travel while minimizing the observation range of each train.

[0029] In actual outdoor environments, there are many emergencies and the weather will also change. Therefore, the original information captured by the lidar and camera is very complex and changeable. If the changes in the original information captured by the lidar and camera are directly used as the basis for sending the three-dimensional model update information to the train warning module, the warning module will need to receive too much information and will transmit a lot of information that is irrelevant to the assisted driving, affecting the timeliness of the assisted driving response. To address this problem, this application provides the following technical solutions:

[0030] Furthermore, the track model building unit includes:

[0031] Point cloud data preprocessor, used for preprocessing point cloud data;

[0032] An image data preprocessor, used for preprocessing image data;

[0033] An image fusion module is used to fuse point cloud data and image data to obtain fused data;

[0034] The information discriminator is used to receive the fused data and compare the fused data received by each detection group with the previous fused data. If the fused data has not changed, it does not belong to updated information. If it has changed, it belongs to updated information.

[0035] In this solution, after receiving the point cloud data and image data, the track model construction unit will pre-process the point cloud data and image data, and then generate fused data. Whether the fused data belongs to update information is determined based on whether the three-dimensional model generated by the fused data is the same as the original three-dimensional model. Therefore, in practice, the fused data does not need to be deeply analyzed and constitute a three-dimensional model. In this way, the fused data does not require complex calculations, but only needs to represent its corresponding characteristics. When the early warning module of the train needs updated information for the area, the fused data with changes is directly sent as update information. In this way, the early warning module needs to receive less data and needs to calculate less data, which increases the timeliness of the early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings constituting a part of this application are used to provide a further understanding of this application, so that other features, purposes and advantages of this application become more obvious. The illustrative embodiment drawings and their descriptions of this application are used to explain this application and do not constitute an improper limitation on this application.

[0037] In addition, throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the components and elements are not necessarily drawn to scale.

[0038] Figure 1 A schematic diagram of the structure of the auxiliary driving system of a three-dimensionally modeled rail train.

[0039] Figure 2 Schematic diagram of the observation range.

[0040] Figure 3 Schematic diagram of the direction of key information transmission in the assisted driving system of a three-dimensionally modeled rail train.

[0041] Figure 4 A flow chart of an assisted driving method for a three-dimensionally modeled rail train. DETAILED DESCRIPTION

[0042] Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms, and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0043] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0044] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0045] refer to Figure 1 Embodiment 1: The present application provides an assisted driving system for rail trains based on three-dimensional modeling. The assisted driving system mainly installs a large number of visual sensors on both sides of the track and then constructs a dynamic three-dimensional model. When driving a train, the driver can roughly understand the situation in front of the train based on the dynamic three-dimensional model and understand the changes in the situation outside the driver's field of view. It is convenient for the driver to discover sudden problems early and use intervention measures to deal with dynamic areas earlier.

[0046] The auxiliary driving system of rail trains based on three-dimensional modeling includes: laser radar, camera, three-dimensional model building device and early warning device. Among them, there are multiple laser radars arranged along both sides of the track, and the maximum detection range of two adjacent laser radars on the same side of the track intersects in the extension direction of the track; there are multiple cameras arranged along both sides of the track, and the field of view of two adjacent cameras on the same side of the track intersects in the extension direction of the track. Both laser radar and camera are visual sensors. Laser radar is used to capture the motion information and position information of objects, and the position of objects can be calculated. Cameras are used to capture the color information and shape information of the surface of objects. In three-dimensional model construction technology, laser radar and cameras are generally used as the main information acquisition methods. Among them, the information obtained by laser radar is point cloud data, and the information obtained by the camera is image data.

[0047] In this solution, a laser radar and a camera are set as a detection group, and the detection range of the laser radar and camera in a detection group is the same. Because 3D image data on both sides of the track is required when building a 3D model, many detection groups are arranged on both sides of the track. These detection groups are used to obtain point cloud data and image data to complete the construction of the 3D model along both sides of the entire track.

[0048] In practice, since rail trains are generally powered by electricity, a large number of detection groups can be installed on both sides of the power facility fixed frame.

[0049] For each detection group, the center line of its detection is perpendicular to the track, and the detection ranges of two adjacent detection groups overlap. Therefore, after collecting information from all detection groups on the track, point cloud data and image data of all areas along both sides of the entire track can be obtained. Using the collected point cloud data and image data, a three-dimensional model can be constructed along both sides of the track.

[0050] The 3D model building device is connected to the laser radar and camera signals respectively, and builds the original 3D model on both sides of the track based on the point cloud data and image data uploaded by the laser radar and camera. Among them, the 3D model building device is actually a cloud service device. All laser radars and cameras will transmit the collected data to the 3D model building device, and then the 3D model building device will send different original 3D models or some update information to the train. Therefore, the 3D model building device can be understood as the cloud service platform of the entire assisted driving system, which needs to handle data collection, processing and distribution on the platform.

[0051] The early warning device is arranged on the train for assisting driving and is connected to the three-dimensional model building device by signal; the early warning device updates and displays the three-dimensional models on both sides of the track in real time. The early warning device is arranged on each train.

[0052] The 3D model building device includes: an information collection unit, a track model building unit and an information distribution unit. The information collection unit is connected to the laser radar and camera signals. The information collection unit is mainly used to collect point cloud data and image data, and can transmit all the information collected by the laser radar and camera in real time.

[0053] There are multiple track model construction units, one for each track, and each track model construction unit is used to construct a three-dimensional model of the corresponding track. In actual track transportation, there are a large number of tracks, which will form a complete transportation network. Therefore, a track model construction unit is configured for each track to establish a three-dimensional model of the track, which is actually the initial three-dimensional model around the track. In this way, the track model construction unit is a processor with integrated data calculation functions. In practice, because the environment around the track will change, the track model construction unit will update the initial three-dimensional model of the track once or several times at a fixed time every day.

[0054] The information distribution unit is connected to the signals of each early warning device. The early warning device is used to upload the track of the train it is on to the information distribution unit. The information distribution unit sends the initial three-dimensional model of the track where the train is located to the early warning module. The early warning device is actually set on each train to provide auxiliary guidance to the train driver. The train schedule and operation plan will be specified before the train runs, so before the train departs, the early warning device of the train will send the track it needs to select to the information distribution unit, and then the information distribution unit will find the corresponding track model construction unit based on the track, obtain the initial three-dimensional model of the track from the track model construction unit, and then send it to the early warning device.

[0055] The early warning module includes: a speed acquisition unit, a model information acquisition unit and a display unit. The speed acquisition unit is used to acquire the train's running speed and the shortest braking distance, and send the running speed and the shortest braking distance to the information distribution unit; the model information acquisition unit is connected to the information distribution unit signal; the model information acquisition unit acquires the initial three-dimensional model information of the track before the train enters the corresponding track, and constructs a static three-dimensional model; the display unit is used to display the three-dimensional model.

[0056] In this solution, the speed acquisition unit will directly connect to the train's speed monitor signal to obtain the train's speed and the train's shortest braking distance. The shortest braking distance can generally be obtained from the train's control center, which is an important parameter in train operation management. The specific calculation method or acquisition method will not be further elaborated here.

[0057] The model information acquisition unit is actually a three-dimensional model calculation system located on the train. Like the track model construction unit, it can calculate the three-dimensional model according to the information received by the detection group. In this solution, the track model construction unit will first send the initial three-dimensional model calculated every day to the model information acquisition unit. The model information acquisition unit can obtain an initial three-dimensional model without calculation. Then the information distribution unit will continuously send update information to the model information acquisition unit. The model information acquisition unit will update the three-dimensional model according to the update information and then display it on the display unit.

[0058] Because the track is very long, it is not practical to calculate the 3D model of too large an area at one time. Therefore, this scheme adopts the following method to limit the calculation area of ​​the 3D model.

[0059] refer to Figure 2 , an information distribution unit, which obtains the travel speed and current position of the train in real time, obtains the observation range of the train according to the travel speed and current position of the train, and sends the update information of the changes in the observation range compared with the static three-dimensional model to the model information acquisition unit;

[0060] The height of the observation range relative to the ground is a fixed value. The horizontal distance between any point L on the edge of the observation range and the center line of the train is L1. The distance between the current position of the train and the observation range along the train's forward route is L2. L1 = (V1 / L0) V2; L2 = S v +λS0V1, where V1 is the current train speed, L0 is the length from point L along the train's forward direction to the plane where the train is located, S vis the shortest braking distance at the current speed of the train, S0 is the pre-set compensation distance, λ is the compensation coefficient, 0<λ<1. V2 is the average moving speed of moving objects around the track. V2 is actually an important indicator to measure the width of the observation range. If V2 is set to a large value, the observation range will be very wide, which is not conducive to reducing the amount of calculation. In this scheme, V2 is set to the average moving speed of moving objects around the track, which is actually the general moving rate of the area crossed by the track. If the area crossed by the track is mostly mountainous and forested, the setting of V2 only needs to be related to the rolling rate of stones on the hillside. If the track is intertwined with social vehicles, V2 needs to be set to be related to the driving rate of the vehicle.

[0061] In this way, the observation range is a cone-shaped area that spreads outward along the end of the train. Relatively speaking, if the train speed is fast, the cone area is longer, but correspondingly narrower. If the train speed is low, the cone area is shorter, but wider.

[0062] In this solution, the track model construction unit is used to construct the initial three-dimensional model, and then the update information is sent to the model information acquisition unit, which completes the model update. If the information detected by the detection group is directly used as the basis for updating the information, there will be a lot of redundant information, resulting in a lot of invalid information being transmitted between the three-dimensional model construction device and the early warning device.

[0063] refer to Figure 3 To this end, the present application provides the following technical solutions: the track model construction unit includes: a point cloud data preprocessor, an image data preprocessor, an image fusion module, an information discriminator, and an initial model generator. Among them, the point cloud data preprocessor is used to preprocess the point cloud data; the image data preprocessor is used to preprocess the image data; the image fusion module is used to fuse the point cloud data and the image data to obtain fused data;

[0064] The information discriminator is used to receive the fused data and compare the fused data received by each detection group with the previous fused data. If the fused data has not changed, it does not belong to updated information. If it has changed, it belongs to updated information.

[0065] The information discriminator only determines whether the fused data has changed. In fact, the fused data is the simplified data that can form the model. If the fused data changes, the 3D model will change, so as to determine whether it is an updated information.

[0066] In practice, in order to reduce the workload of the information discriminator, the information discriminator only compares whether the fused data within the observation range is updated information.

[0067] The initial model generator is a unit of the track model building unit used to build an initial three-dimensional model.

[0068] refer to Figure 4 , Embodiment 2: An assisted driving method for a rail train based on three-dimensional modeling, implemented by the assisted driving system for a rail train based on three-dimensional modeling of Embodiment 1;

[0069] The method comprises the following steps:

[0070] Step 1: Multiple lidars and cameras are arranged at intervals along both sides of the track to obtain point cloud data and image data of the areas on both sides of the track in real time.

[0071] The step 1 comprises the following steps:

[0072] Step 11: Arrange multiple laser radars and cameras on both sides of the track. One laser radar and one camera are set as a detection group. The detection ranges of the laser radars and cameras in the same detection group are the same.

[0073] The detection ranges of adjacent detection groups intersect in the track extension direction;

[0074] Step 12: The laser radar and detection group collects point cloud data and image data in real time, and sends the point cloud data and image data to the three-dimensional model building device.

[0075] Step 2: The 3D model building device receives the point cloud data and the image data to build the original 3D model of both sides of the track.

[0076] Step 2 includes the following steps:

[0077] Step 21: The information collection unit downloads point cloud data and image data of the areas on both sides of the track;

[0078] Step 22: The track model building unit generates an original three-dimensional model based on the point cloud data and image data of the areas on both sides of the track.

[0079] Step 3: The early warning device downloads the original three-dimensional model from the three-dimensional model building device;

[0080] Step 4: The 3D model building device receives point cloud data and image data in real time and generates update information;

[0081] The early warning device updates the three-dimensional model in real time based on the updated information and displays the three-dimensional model on both sides of the track.

[0082] Step 4 includes the following steps:

[0083] Step 41: The speed acquisition unit acquires the running speed and the shortest stopping distance of the train, and sends the running speed and the shortest stopping distance to the information distribution unit;

[0084] Step 42: The information distribution unit obtains the train's travel speed and current position in real time, obtains the train's observation range based on the train's travel speed and current position, and sends updated information about changes in the observation range compared to the static three-dimensional model to the model information acquisition unit.

[0085] Step 42 includes the following steps:

[0086] Step 421: The point cloud data preprocessor and the image data preprocessor preprocess the point cloud data and image data received in real time;

[0087] Step 422: The image fusion module fuses the pre-processed point cloud data and the image data to obtain fused data;

[0088] Step 423: The information distinguisher compares the fused data received by each detection group with the previous fused data. If the fused data has not changed, it does not belong to updated information. If it has changed, it belongs to updated information.

[0089] Step 424: The information distribution unit sends the update information within the train observation range to the model information acquisition unit.

[0090] Example 3: The assisted driving system and assisted driving method of the rail train provided in Example 1 and Example 2. The key part is how to construct a three-dimensional model and how to update the three-dimensional model. To this end, this application provides the following three-dimensional model construction method:

[0091] S1: Collect image data, pre-process the image data, collect point cloud data, and pre-process the point cloud data.

[0092] The preprocessing is conventional, mainly including normalization, denoising and other operations.

[0093] S2: Extract feature points from the preprocessed image data using the SIFT algorithm.

[0094] S3: Match different pictures in the image data based on feature points to align the image data with each other.

[0095] S4: Collect the focal length of the camera and restore the three-dimensional coordinates of the feature points through triangulation and other methods to obtain the three-dimensional coordinates of the non-ground points.

[0096] S5: Down-sample the point cloud data, i.e., perform thinning processing.

[0097] S6: Identify and remove discrete points in the point cloud data.

[0098] In point cloud data, discrete points are those points that are far away from surrounding points and do not conform to the overall data distribution pattern. These points may be caused by measurement errors, equipment failures, or environmental factors. The existence of discrete points will interfere with subsequent triangulation modeling, so they need to be removed before triangulation. Whether it is a discrete point can be determined based on the point cloud density.

[0099] S7: The sparse point cloud is interpolated using a linear interpolation method, adjacent points are connected, a triangular mesh is constructed, and a triangulated digital ground model is formed.

[0100] After downsampling and removing discrete points, the point cloud data may become relatively sparse. In order to construct a continuous triangle mesh, the sparse point cloud needs to be interpolated.

[0101] S8: Integrate the triangulated mesh model of the ground points with the three-dimensional coordinate points of the non-ground points.

[0102] After completing the triangulated modeling of ground points and the restoration of the three-dimensional coordinates of non-ground points, these two parts of data need to be integrated to form a complete three-dimensional scene model.

[0103] In this embodiment, steps S1 to S8 are the process of generating fused data in embodiment 2. Determining whether the fused data has changed is the main way to determine whether the information is updated.

[0104] S9: Use 3D modeling software or rendering engine.

[0105] The above description is only some preferred embodiments of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present application is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present application to form a technical solution.

Claims

1. An auxiliary driving system for a rail train based on three-dimensional modeling, characterized in that: include: There are multiple laser radars arranged along both sides of the track, and the maximum detection ranges of two adjacent laser radars located on the same side of the track intersect in the extension direction of the track; A plurality of cameras are arranged along both sides of the track, and the field of view of two adjacent cameras located on the same side of the track intersects in the extension direction of the track; A 3D model building device is connected to the laser radar and camera signals respectively, and builds the original 3D model of both sides of the track based on the point cloud data and image data uploaded by the laser radar and camera; An early warning device, which is installed on the train for assisting driving and is connected to the three-dimensional model building device by signal; Wherein, the warning device updates and displays the three-dimensional models of both sides of the track in real time; The laser radar is used to obtain point cloud data of the areas on both sides of the track, and the camera is used to obtain image data on both sides of the track; The three-dimensional model building device includes: An information collection unit connected to each laser radar and camera signal; A track model building unit is provided, wherein a plurality of track model building units are provided, each track corresponds to a track model building unit, and each track model building unit is used to build a three-dimensional model of the corresponding track; An information distribution unit is connected to the signals of each early warning device. The early warning device is used to upload the track of the train on which it is located to the information distribution unit. The information distribution unit sends the corresponding three-dimensional model to the early warning module according to the track on which the train is located; The early warning module comprises: A speed acquisition unit, used to acquire the running speed and the shortest stopping distance of the train, and send the running speed and the shortest stopping distance to the information distribution unit; A model information acquisition unit, signal-connected to the information distribution unit; The model information acquisition unit acquires all three-dimensional model information of the track before the train enters the corresponding track, and constructs a static three-dimensional model; An information distribution unit obtains the travel speed and current position of the train in real time, obtains the observation range of the train according to the travel speed and current position of the train, and sends update information of changes in the observation range compared with the static three-dimensional model to the model information acquisition unit; The height of the observation range relative to the ground is a fixed value. The horizontal distance between any point L on the edge of the observation range and the center line of the train is L1. The distance between the current position of the train and the observation range along the train's forward route is L2. L1 = (V1 / L0) V2; L2 = S v +λS0V1, where V1 is the current train speed, L0 is the length from point L along the train's forward direction to the plane where the train is located, S v is the shortest braking distance at the current speed of the train, S0 is the preset compensation distance, λ is the compensation coefficient, 0<λ<1, and V2 is the average moving speed of moving objects around the track.

2. The auxiliary driving system for rail vehicles based on three-dimensional modeling according to claim 1, characterized in that: Track model building unit includes: Point cloud data preprocessor, used for preprocessing point cloud data; An image data preprocessor, used for preprocessing image data; An image fusion module is used to fuse point cloud data and image data to obtain fused data; The information discriminator is used to receive the fused data and compare the fused data received by each detection group with the previous fused data. If the fused data has not changed, it does not belong to updated information. If it has changed, it belongs to updated information.

3. A three-dimensional modeling-based rail vehicle auxiliary driving method, characterized in that: The method is implemented by using a three-dimensional modeling-based rail train auxiliary driving system according to any one of claims 1 to 2; the method comprises the following steps: Step 1: Multiple laser radars and cameras are arranged at intervals along both sides of the track to obtain point cloud data and image data of the areas on both sides of the track in real time; Step 2: The 3D model building device receives the point cloud data and the image data to build the original 3D model of both sides of the track; Step 3: The early warning device downloads the original three-dimensional model from the three-dimensional model building device; Step 4: The 3D model building device receives point cloud data and image data in real time and generates update information; The early warning device updates the three-dimensional model in real time based on the updated information and displays the three-dimensional model on both sides of the track.

4. The assisted driving method of a rail train based on three-dimensional modeling according to claim 3 is characterized in that: The step 1 comprises the following steps: Step 11: Arrange multiple laser radars and cameras on both sides of the track. One laser radar and one camera are set as a detection group. The detection ranges of the laser radars and cameras in the same detection group are the same. The detection ranges of adjacent detection groups intersect in the track extension direction; Step 12: The laser radar and detection group collects point cloud data and image data in real time, and sends the point cloud data and image data to the three-dimensional model building device.

5. The assisted driving method of a rail train based on three-dimensional modeling according to claim 4 is characterized in that: Step 2 includes the following steps: Step 21: The information collection unit downloads point cloud data and image data of the areas on both sides of the track; Step 22: The track model building unit generates an original three-dimensional model based on the point cloud data and image data of the areas on both sides of the track.

6. The assisted driving method of a rail train based on three-dimensional modeling according to claim 5, characterized in that: Step 4 includes the following steps: Step 41: The speed acquisition unit acquires the running speed and the shortest stopping distance of the train, and sends the running speed and the shortest stopping distance to the information distribution unit; Step 42: The information distribution unit obtains the train's travel speed and current position in real time, obtains the train's observation range based on the train's travel speed and current position, and sends updated information about changes in the observation range compared to the static three-dimensional model to the model information acquisition unit.

7. The assisted driving method of a rail train based on three-dimensional modeling according to claim 6, characterized in that: Step 42 includes the following steps: Step 421: The point cloud data preprocessor and the image data preprocessor preprocess the point cloud data and image data received in real time; Step 422: The image fusion module fuses the pre-processed point cloud data and the image data to obtain fused data; Step 423: The information distinguisher compares the fused data received by each detection group with the previous fused data. If the fused data has not changed, it does not belong to updated information. If it has changed, it belongs to updated information. Step 424: The information distribution unit sends the update information within the train observation range to the model information acquisition unit.

Images