Methods, devices, and rail vehicles for generating virtual track information

Abstract

This invention provides a method, apparatus, and rail vehicle for generating virtual track information. The rail vehicle is equipped with multiple position positioning sensors. The method includes: before the rail vehicle starts running, determining target reference routes formed by the rail vehicle during its movement along the center of the virtual track based on each position positioning sensor; when the rail vehicle starts running, collecting real-time position information of the rail vehicle based on each position positioning sensor; determining initial pre-aiming travel trajectory data formed under each position positioning sensor based on the real-time position information and the target reference routes corresponding to the real-time position information; and processing each initial pre-aiming travel trajectory data to obtain target virtual track information. This method enables accurate generation of virtual track information, improving the accurate perception capability of the autonomous driving system.

Description

Technical Field

[0001] This invention relates to the field of virtual track technology, and in particular to a method, apparatus, and track vehicle for generating virtual track information. Background Technology

[0002] Autonomous driving technology, a current research hotspot in industry and academia, aims to reduce the driver's workload and significantly improve vehicle safety. To achieve this, autonomous driving systems must ensure that vehicles continuously and stably remain within their lane markings. Therefore, the vehicle's accurate perception of lane markings is crucial for stable driving.

[0003] In traditional solutions, autonomous driving systems primarily rely on data from a single sensor for lane marking and localization. However, single-sensor data can be affected by environmental conditions, obstructions, and other factors, resulting in low accuracy and failing to meet the autonomous driving system's requirement for precise perception of vehicle status.

[0004] Therefore, finding a method to accurately generate virtual track information for rail vehicles has become a research hotspot. Summary of the Invention

[0005] This invention provides a method, apparatus, and rail vehicle for generating virtual track information, which enables accurate generation of virtual track information and improves the precise perception capability of the automatic driving system.

[0006] This invention provides a method for generating virtual track information for a rail vehicle, wherein the rail vehicle is equipped with multiple position positioning sensors. The method includes: before the rail vehicle starts running, determining each target reference route formed by the rail vehicle during its travel along the center of the virtual track based on each of the position positioning sensors; when the rail vehicle starts running, collecting real-time position information of the rail vehicle based on each of the position positioning sensors; determining initial pre-aiming travel trajectory data formed under each of the position positioning sensors based on the real-time position information and the target reference routes corresponding to the real-time position information; and processing each of the initial pre-aiming travel trajectory data to obtain target virtual track information, wherein the target virtual track information is used to characterize the track information on which the rail vehicle is about to travel.

[0007] According to a method for generating virtual track information for a rail vehicle provided by the present invention, the step of processing each of the initial preview travel trajectory data to obtain target virtual track information specifically includes: determining the weight value of each of the initial preview travel trajectory data, wherein the weight value is determined based on the positioning accuracy of each of the position positioning sensors; and performing weighted fusion processing based on the weight value and each of the initial preview travel trajectory data to obtain the target virtual track information.

[0008] According to a method for generating virtual track information for a rail vehicle provided by the present invention, before processing the initial pre-aiming travel trajectory data to obtain target virtual track information, the method further includes: obtaining the Jacobian matrix of the observation equation with respect to the state variables based on the position information output by each of the position positioning sensors and the state variables in the pre-determined error state Kalman filter algorithm; the processing of the initial pre-aiming travel trajectory data to obtain target virtual track information specifically includes: substituting the initial pre-aiming travel trajectory data into the Jacobian matrix of the observation equation with respect to the state variables, performing Kalman gain calculation for each observation to obtain an error state estimate of the state variables, wherein the observations include the position information output by each of the position positioning sensors; the state variables are used to characterize the state variables involved in the operation of the rail vehicle; based on the error state estimate and the state variables, a corrected state variable is obtained; based on the corrected state variable, the target virtual track information is obtained.

[0009] According to a method for generating virtual track information of a rail vehicle provided by the present invention, the step of obtaining target virtual track information based on the corrected state variables specifically includes: determining the target real-time position information of the rail vehicle based on the corrected state variables, wherein the target real-time position information is used to characterize the real-time position information of the rail vehicle with an accuracy higher than an accuracy threshold; and comparing the target real-time position information with standard route data in the target reference route to obtain the target virtual track information.

[0010] According to a method for generating virtual track information for a rail vehicle provided by the present invention, the step of determining initial pre-aiming travel trajectory data formed under each of the location positioning sensors based on the real-time location information and the target reference route corresponding to the real-time location information specifically includes: comparing the real-time location information with standard route data in the target reference route to obtain the initial pre-aiming travel trajectory data formed under each of the location positioning sensors.

[0011] According to a method for generating virtual track information for a rail vehicle provided by the present invention, the rail vehicle is further equipped with a visual sensor; before determining each target reference route formed by the rail vehicle during its travel along the virtual track center based on each of the position positioning sensors, the method further includes: determining a deviation from the lane center reference distance based on the visual sensor, wherein the deviation from the lane center reference distance is used to characterize the distance by which the rail vehicle deviates from the virtual track center during normal operation; the determination of each target reference route formed by the rail vehicle during its travel along the virtual track center based on each of the position positioning sensors specifically includes: determining each reference route formed by the rail vehicle during its travel along the virtual track center based on each of the position positioning sensors; and performing deviation processing on each reference route based on the deviation from the lane center reference distance to obtain each target reference route formed by the rail vehicle during its travel along the virtual track center under the position positioning sensors.

[0012] According to a method for generating virtual track information for a rail vehicle provided by the present invention, the plurality of position positioning sensors include a combined navigation device, a lidar, and a magnetic sensor; the reference route includes a first reference route corresponding to the combined navigation device, a second reference route corresponding to the lidar, and a third reference route corresponding to the magnetic sensor; the target reference route includes a first target reference route corresponding to the first reference route, a second target reference route corresponding to the second reference route, and a third target reference route corresponding to the third reference route; the step of determining each reference route formed by the rail vehicle during its travel along the virtual track center based on each of the position positioning sensors specifically includes: determining the first reference route formed by the rail vehicle during its travel along the virtual track center based on the combined navigation device; determining the second reference route formed by the rail vehicle during its travel along the virtual track center based on the lidar; and so on. The magnetic sensor determines the third reference route formed by the rail vehicle traveling along the virtual track center. The step of performing deviation processing on each reference route based on the deviation distance from the lane center to obtain target reference routes formed by the rail vehicle traveling along the virtual track center under each of the position positioning sensors specifically includes: performing deviation processing on each reference route based on the deviation distance from the lane center to obtain a first target reference route formed by the rail vehicle traveling along the virtual track center under the integrated navigation device; performing deviation processing on each reference route based on the deviation distance from the lane center to obtain a second target reference route formed by the rail vehicle traveling along the virtual track center under the lidar; and performing deviation processing on each reference route based on the deviation distance from the lane center to obtain a third target reference route formed by the rail vehicle traveling along the virtual track center under the magnetic sensor.

[0013] This invention also provides a device for generating virtual track information for a rail vehicle, wherein the rail vehicle is equipped with multiple position positioning sensors; the device includes: a determining module, used to determine, based on each of the position positioning sensors, target reference routes formed by the rail vehicle during its travel along the center of the virtual track before the rail vehicle starts to run; a collecting module, used to collect real-time position information of the rail vehicle based on each of the position positioning sensors when the rail vehicle starts to run; a processing module, used to determine initial pre-aiming travel trajectory data formed under each of the position positioning sensors based on the real-time position information and the target reference routes corresponding to the real-time position information; and a generating module, used to process each of the initial pre-aiming travel trajectory data to obtain target virtual track information, wherein the target virtual track information is used to characterize the track information on which the rail vehicle is about to travel.

[0014] The present invention also provides a rail vehicle, the rail vehicle comprising: a rail vehicle body and a processor, wherein the processor is configured to execute the method for generating virtual track information of the rail vehicle.

[0015] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for generating virtual track information of a rail vehicle as described above.

[0016] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for generating virtual track information of a rail vehicle as described above.

[0017] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the method for generating virtual track information of a rail vehicle as described above.

[0018] This invention provides a method, apparatus, and rail vehicle for generating virtual track information. The rail vehicle is equipped with multiple position positioning sensors. The method includes: before the rail vehicle starts running, determining target reference routes formed by the rail vehicle during its movement along the center of the virtual track based on each position positioning sensor; when the rail vehicle starts running, collecting real-time position information of the rail vehicle based on each position positioning sensor; determining initial pre-aiming travel trajectory data formed under each position positioning sensor based on the real-time position information and the target reference routes corresponding to the real-time position information; and processing each initial pre-aiming travel trajectory data to obtain target virtual track information. This achieves accurate generation of virtual track information, improving the accurate perception capability of the autonomous driving system. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the method for generating virtual track information for rail vehicles provided by the present invention.

[0021] Figure 2This is a flowchart illustrating the process of processing initial pre-aimed driving trajectory data to obtain target virtual track information, as provided by the present invention.

[0022] Figure 3 This is a flowchart illustrating how the present invention determines the reference routes of each target formed during the movement of a rail vehicle along the center of a virtual track, based on positioning sensors at each location.

[0023] Figure 4 This is a schematic diagram of the structure of the device for generating virtual track information for rail vehicles provided by the present invention.

[0024] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0026] The method for generating virtual track information for rail vehicles provided by this invention can accurately identify the target trajectory and lateral offset of the vehicle through multi-sensor data fusion technology, combined with vision, lidar, integrated navigation system, magnetic sensors, etc., and generate virtual track information based on this, thereby supporting the realization of autonomous driving capabilities of the vehicle.

[0027] Figure 1 This is a flowchart illustrating the method for generating virtual track information for rail vehicles provided by the present invention.

[0028] The following will combine Figure 1 The process of generating virtual track information for rail vehicles provided by this invention will be described.

[0029] In an exemplary embodiment of the present invention, the rail vehicle may be equipped with multiple position positioning sensors, such as integrated navigation equipment, lidar, magnetic sensors, and visual sensors. In this embodiment, no specific limitation is made to the position positioning sensors.

[0030] Combination Figure 1 As can be seen, the method for generating virtual track information for rail vehicles may include steps 110 to 140, and each step will be described below.

[0031] In step 110, before the rail vehicle starts running, the target reference routes formed by the rail vehicle during its journey along the virtual track center are determined based on the positioning sensors at each location.

[0032] In one embodiment, before the rail vehicle starts operating, target reference routes formed by the rail vehicle during its travel along the virtual track center can be determined based on each location sensor. These target reference routes can be considered as the routes the rail vehicle travels on the virtual track during normal operation.

[0033] In step 120, when the rail vehicle starts to run, the real-time position information of the rail vehicle is collected in real time based on the positioning sensors at each position.

[0034] In another embodiment, the real-time position information of the rail vehicle can be collected in real time based on each positioning sensor when the rail vehicle starts running. In one example, if the positioning sensor is a combined navigation device, the current real-time position of the vehicle (corresponding real-time position information) can be obtained based on the combined navigation device. If the positioning sensor is a LiDAR, the current real-time position of the vehicle (corresponding real-time position information) can be obtained based on the LiDAR using a SLAM algorithm. If the positioning sensor is a magnetic sensor, the current real-time position of the vehicle (corresponding real-time position information) can be obtained based on the detection of ground magnetic nails by the magnetic sensor.

[0035] In step 130, based on real-time location information and the target reference route corresponding to the real-time location information, the initial aiming driving trajectory data formed under the positioning sensors at each location is determined.

[0036] In another embodiment, the initial aiming trajectory data formed under each position sensor can be determined based on the real-time position information obtained by each position sensor and the target reference route corresponding to the real-time position information (corresponding to the target reference route determined under each position sensor).

[0037] In one example, when the location sensor is a navigation system, the current real-time vehicle location (corresponding to real-time location information) can be obtained based on the navigation system and compared with the target data in the target reference route determined by the navigation system to obtain the initial pre-aiming driving trajectory data formed by the navigation system.

[0038] In another example, when the positioning sensor is a lidar, the current real-time position of the vehicle (corresponding to real-time position information) can be obtained based on the lidar using the SLAM algorithm, and compared with the target data in the target reference route determined by the lidar to obtain the initial pre-aiming driving trajectory data formed by the lidar.

[0039] In another example, when the location sensor is a magnetic sensor, the current real-time position of the vehicle (corresponding to real-time position information) can be obtained by detecting ground magnetic nail information based on the magnetic sensor, and compared with the target data in the target reference route determined by the magnetic sensor to obtain the initial aiming driving trajectory data formed by the magnetic sensor.

[0040] In another embodiment, the information of the lane ahead can be obtained based on the vision sensor device, and the center of the lane formed by the left and right lanes ahead can be used as the target trajectory. The forward aiming driving trajectory based on the vehicle coordinate system can be obtained through perspective transformation, that is, the initial aiming driving trajectory data formed under the vision sensor can be obtained.

[0041] In step 140, the initial pre-aimed travel trajectory data are processed to obtain target virtual track information, which is used to characterize the track information on which the rail vehicle will travel.

[0042] In another embodiment, the initial pre-aimed driving trajectory data can be processed to obtain target virtual track information, and the operation of the rail vehicle can be guided based on the target virtual track information, thereby achieving accurate generation of virtual track information and improving the accurate perception capability of the autonomous driving system.

[0043] In yet another exemplary embodiment of the present invention, the preceding text continues... Figure 1 The above embodiment will be used as an example for illustration. The initial pre-aiming trajectory data is processed to obtain the target virtual track information (corresponding to step 140), which can be achieved in the following way:

[0044] The weight values ​​of each initial pre-aiming driving trajectory data are determined, where the weight values ​​are determined based on the positioning accuracy of each position positioning sensor;

[0045] The target virtual track information is obtained by weighted fusion processing based on the weight values ​​and the initial target trajectory data.

[0046] In one embodiment, after obtaining the initial preview driving trajectory data determined by each position sensor, a weight value for each initial preview driving trajectory data can be determined separately. The weight value can be determined based on the positioning accuracy of each position sensor. It is understood that the higher the positioning accuracy of the position sensor, the higher its corresponding weight value.

[0047] In application, weighted fusion processing can be performed based on weight values ​​and initial pre-aimed driving trajectory data to obtain target virtual track information. In this embodiment, the obtained target virtual track information is determined based on multiple sets of position positioning sensors, and the positioning accuracy of each sensor is also considered, thereby ensuring the rationality and reference value of the obtained target virtual track information.

[0048] Figure 2 This is a flowchart illustrating the process of processing initial pre-aimed driving trajectory data to obtain target virtual track information, as provided by the present invention.

[0049] The following will combine Figure 2 The process of processing the initial pre-aimed trajectory data to obtain the target virtual trajectory information is explained.

[0050] In an exemplary embodiment of the present invention, combined with Figure 2 As can be seen, processing the initial pre-aiming trajectory data to obtain the target virtual trajectory information may include steps 210 to 240, which will be described in detail below.

[0051] In step 210, based on the position information output by each positioning sensor and the state variables in the pre-determined error state Kalman filter algorithm, the Jacobian matrix of the observation equation with respect to the state variables is obtained.

[0052] In one embodiment, the Jacobian matrix of the observation equation versus the state variables can be obtained based on the position information output by each positioning sensor and the state variables in the pre-determined error state Kalman filter algorithm. The state variables can be used to characterize the state variables involved in the operation of the rail vehicle. These state variables may include translation p in the three-dimensional space of the world coordinate system; velocity v in the three-dimensional space of the world coordinate system; rotation R in the three-dimensional space of the world coordinate system; and accelerometer bias b. a ; angular velocity meter zero bias b g ; g in the three-dimensional space of the world coordinate system; e, the lateral offset between the vehicle center and the virtual trajectory.

[0053] In step 220, the initial pre-aiming trajectory data is substituted into the Jacobian matrix of the state variables in the observation equation, and the Kalman gain is calculated for each observation to obtain the error state estimate of the state variables. The observations include the position information output by each positioning sensor; the state variables are used to characterize the state variables involved in the operation of the rail vehicle.

[0054] In step 230, the corrected state variables are obtained based on the error state estimation and the state variables.

[0055] In step 240, the target virtual orbit information is obtained based on the corrected state variables.

[0056] In another embodiment, the initial preview trajectory data can be substituted into the Jacobian matrix of the state variables in the observation equation. Kalman gain is calculated for each observation to obtain an error state estimate of the state variables. The observations may include position information output by each positioning sensor. Further, a corrected state variable is obtained based on the error state estimate and the state variables, and the target virtual track information is obtained based on the corrected state variable. In this embodiment, the Error State Kalman Filter (ESKF) algorithm is used to estimate the state of the vehicle system. By simultaneously estimating a set of state variables, tight coupling of multiple observation modules is achieved, thereby accurately obtaining the target virtual track information based on each initial preview trajectory data.

[0057] In yet another exemplary embodiment of the present invention, the preceding text continues... Figure 2 The above embodiment will be used as an example for illustration. Based on the corrected state variables, the target virtual orbit information (corresponding to step 240) can be obtained in the following way:

[0058] Based on the corrected state variables, the target real-time position information of the rail vehicle is determined, wherein the target real-time position information is used to characterize the real-time position information of rail vehicles with an accuracy higher than the accuracy threshold.

[0059] The target's real-time location information is compared with the standard route data in the target reference route to obtain the target's virtual track information.

[0060] In one embodiment, the target real-time position information of the rail vehicle can be determined based on the corrected state variables determined by the Error State Kalman Filter (ESKF) algorithm. In one example, the corrected state variables can be used as the target real-time position information of the rail vehicle. It is understood that the target real-time position information can be considered to be real-time position information with higher accuracy than real-time position information collected based on position positioning sensors.

[0061] Furthermore, the real-time target location information is compared with the standard route data in the target reference route to obtain the target virtual track information. The target reference route can be formed using any type of positioning sensor. In this embodiment, the obtained real-time target location information is more accurate than the real-time location information collected by the positioning sensor, thus allowing for the acquisition of more accurate target virtual track information based on the real-time target location information and the standard route data in the target reference route.

[0062] In yet another exemplary embodiment of the present invention, the preceding text continues... Figure 1The above embodiment will be used as an example for illustration. Based on real-time location information and the target reference route corresponding to the real-time location information, the initial pre-aiming driving trajectory data formed under the positioning sensors at each location (comparison step 130) can be determined in the following way:

[0063] By comparing the real-time location information with the standard route data in the target reference route, the initial pre-aiming driving trajectory data formed under the positioning sensors at each location is obtained.

[0064] In one embodiment, real-time location information can be compared with standard route data in the target reference route to obtain forward-looking driving trajectory data (corresponding to initial forward-looking driving trajectory data) formed under the positioning sensors at each location.

[0065] Figure 3 This is a flowchart illustrating how the present invention determines the reference routes of each target formed during the movement of a rail vehicle along the center of a virtual track, based on positioning sensors at each location.

[0066] The following will combine Figure 3 The process of determining the reference routes of each target formed during the movement of the rail vehicle along the virtual track center, based on the positioning sensors at each location, is explained.

[0067] In an exemplary embodiment of the present invention, combined with Figure 3 It is known that the rail vehicle is also equipped with a visual sensor. Combined with... Figure 3 As can be seen, determining the reference routes of each target formed during the movement of the rail vehicle along the virtual track center based on the positioning sensors at each location can include steps 310 to 330, and each step will be described below.

[0068] In step 310, a reference distance from the center of the lane is determined based on a visual sensor. This reference distance is used to characterize the distance by which the rail vehicle deviates from the virtual track center during normal operation.

[0069] In one embodiment, the deviation from the lane center reference distance can be determined based on a visual sensor, wherein the deviation from the lane center reference distance can be used to characterize the distance by which the rail vehicle deviates from the virtual track center during normal operation of the rail vehicle.

[0070] In one embodiment, the visual sensor device can acquire lane information ahead at a frequency of 30Hz, detect left and right lane lines using a lane line detection algorithm, and then obtain lane line coordinates based on the vehicle's coordinate system through perspective transformation. This information, along with the set of left lane line points, right lane line points, and vehicle distance from the lane center, is stored in text format. The vehicle distance from the lane center can be a reference distance from the lane center. It should be noted that during normal operation of rail vehicles, such as in station entry or turnaround scenarios, it is often necessary for the rail vehicle to deviate from the track center. Therefore, the reference distance from the lane center can be determined based on the visual sensor, and the deviation of each reference route determined by the positioning sensors at each location can be processed based on this reference distance to obtain each target reference route.

[0071] In step 320, each reference route formed by the rail vehicle during its journey along the virtual track center is determined based on the positioning sensors at each location.

[0072] In step 330, the deviation of each reference route is processed based on the deviation distance from the lane center reference distance to obtain each target reference route formed by the rail vehicle traveling along the virtual track center under the positioning sensors at each position.

[0073] In another embodiment, reference routes formed by the rail vehicle as it travels along the virtual track center can be determined based on each location sensor. It is understood that a reference route can be considered the route formed by the rail vehicle traveling along the track center. Furthermore, since the rail vehicle often needs to deviate from the track center during normal operation, such as in station entry or transfer scenarios, a reference distance from the track center can be determined based on a visual sensor. This deviation distance can then be used to process the progress deviation of each reference route determined by the location sensors to obtain the target reference route.

[0074] In application, deviation processing can be performed on each reference route based on the distance from the lane center reference, thereby obtaining the target reference routes formed by the rail vehicle traveling along the virtual track center under the positioning sensors at each location. This lays the foundation for accurately and reasonably determining the initial pre-aiming trajectory data.

[0075] In yet another exemplary embodiment of the present invention, the embodiments described above will continue to be used as examples. In this embodiment, the plurality of location sensors may include a combined navigation device, a lidar, and a magnetic sensor; the reference route may include a first reference route corresponding to the combined navigation device, a second reference route corresponding to the lidar, and a third reference route corresponding to the magnetic sensor; the target reference route may include a first target reference route corresponding to the first reference route, a second target reference route corresponding to the second reference route, and a third target reference route corresponding to the third reference route.

[0076] The determination of reference routes formed by the rail vehicle as it travels along the virtual track center, based on positioning sensors at each location, can be achieved in the following way:

[0077] Based on the integrated navigation equipment, the first reference route formed by the rail vehicle during its journey along the virtual track center is determined;

[0078] Based on lidar, a second reference route is determined as the rail vehicle travels along the virtual track center;

[0079] Based on magnetic sensors, a third reference route is determined as the rail vehicle travels along the virtual track center.

[0080] Furthermore, by performing deviation processing on each reference route based on the distance from the lane center reference, the target reference routes formed by the rail vehicle traveling along the virtual track center under the positioning sensors at each location can be obtained in the following way:

[0081] Based on the deviation from the lane center reference distance, each reference route is processed to obtain the first target reference route formed when the rail vehicle travels along the virtual track center under the integrated navigation equipment.

[0082] Based on the deviation from the lane center reference distance, each reference route is processed to obtain the second target reference route formed during the process of the rail vehicle traveling along the virtual track center under lidar.

[0083] Based on the deviation from the lane center reference distance, the deviation of each reference route is processed to obtain the third target reference route formed during the process of the rail vehicle traveling along the virtual track center under the magnetic sensor.

[0084] In one embodiment, a first reference route formed by a rail vehicle traveling along the center of a virtual track can be determined based on a combined navigation device. During application, the combined navigation device can output the vehicle's position at a frequency of 100Hz and store the [longitude, latitude, vehicle orientation] information in text format, thereby determining the first reference route formed by the rail vehicle traveling along the center of the virtual track. The combined navigation device can be mounted on the top of the vehicle, enabling centimeter-level positioning.

[0085] In another embodiment, a second reference route formed by the rail vehicle as it travels along the center of a virtual track can be determined based on lidar. In application, the lidar device can output a point cloud around the vehicle at a frequency of 10Hz, and an electronic map of the travel route can be determined based on this point cloud. For example, a point cloud map file, called an "electronic map," can be constructed and stored using a lidar SLAM algorithm. This map is stored in PCD format and in text format [longitude, latitude, vehicle orientation], thereby determining the second reference route formed by the rail vehicle as it travels along the center of the virtual track. The lidar can be mounted on the top front of the vehicle, and by running a lidar SLAM algorithm, the vehicle's position, attitude, and other states can be measured.

[0086] In another embodiment, a third reference route formed by a rail vehicle traveling along the center of a virtual track can be determined based on a magnetic sensor. In application, the magnetic sensor device can detect information from ground magnetic nails at a frequency of 100Hz and store this information in text format [magnetic nail serial number, magnetic nail's location on the magnetic sensor], thereby determining the third reference route formed by the rail vehicle traveling along the center of the virtual track. The magnetic sensor can be installed on the bottom of the vehicle, and by detecting magnetic nails pre-installed on the ground, it can measure the vehicle's position and attitude.

[0087] In another embodiment, deviation processing can be applied to each reference route based on the deviation distance from the lane center reference distance to obtain a first target reference route formed during the movement of the rail vehicle along the virtual track center under the integrated navigation device. In yet another example, deviation processing can be applied to each reference route based on the deviation distance from the lane center reference distance to obtain a second target reference route formed during the movement of the rail vehicle along the virtual track center under the lidar. In yet another example, deviation processing can be applied to each reference route based on the deviation distance from the lane center reference distance to obtain a third target reference route formed during the movement of the rail vehicle along the virtual track center under the magnetic sensor. In this embodiment, deviation processing is applied to each reference route (including the first, second, and third reference routes) based on the deviation distance from the lane center reference distance, thereby obtaining each target reference route (including the first, second, and third target reference routes) formed during the movement of the rail vehicle along the virtual track center under each positioning sensor, thus laying the foundation for accurately and reasonably determining each initial pre-aiming driving trajectory data.

[0088] As described above, the method for generating virtual track information for rail vehicles provided by this invention addresses the problems of insufficient accuracy of single sensors and high risk of sensor failure in traditional systems by employing multi-sensor data fusion technology. By comprehensively utilizing data from multiple sensors such as vision and lidar, it achieves accurate identification of road information and vehicle lateral offset, and generates virtual track information. This improves the autonomous driving system's ability to accurately perceive vehicle status, reduces the risk of sensor failure, and further enhances the system's robustness and safety.

[0089] The apparatus for generating virtual track information of rail vehicles provided by the present invention is described below. The apparatus for generating virtual track information of rail vehicles described below can be referred to in correspondence with the method for generating virtual track information of rail vehicles described above.

[0090] Figure 4 This is a schematic diagram of the structure of the device for generating virtual track information for rail vehicles provided by the present invention.

[0091] The following will combine Figure 4 The structure of the device for generating virtual track information for rail vehicles is described.

[0092] In an exemplary embodiment of the present invention, the rail vehicle may be equipped with multiple position positioning sensors; combined with Figure 4 As can be seen, the device may include a determining module 410, a data acquisition module 420, a processing module 430, and a generating module 440. Each module will be described in detail below.

[0093] The determining module 410 can be configured to determine, based on each of the position positioning sensors, each target reference route formed by the rail vehicle during its travel along the virtual track center before the rail vehicle starts to run.

[0094] The acquisition module 420 can be configured to acquire real-time position information of the rail vehicle based on each of the position positioning sensors when the rail vehicle starts to run.

[0095] The processing module 430 can be configured to determine the initial aiming driving trajectory data formed under each of the location positioning sensors based on the real-time location information and the target reference route corresponding to the real-time location information.

[0096] The generation module 440 can be configured to process the initial pre-aiming travel trajectory data to obtain target virtual track information, wherein the target virtual track information is used to characterize the track information on which the rail vehicle is about to travel.

[0097] In an exemplary embodiment of the present invention, the generation module 440 can process the initial pre-aiming trajectory data to obtain target virtual track information in the following manner:

[0098] The weight values ​​of each of the initial pre-aiming driving trajectory data are determined respectively, wherein the weight values ​​are determined based on the positioning accuracy of each of the position positioning sensors;

[0099] The target virtual track information is obtained by performing weighted fusion processing based on the weight values ​​and the initial pre-aiming driving trajectory data.

[0100] In an exemplary embodiment of the present invention, the generation module 440 may further be configured to:

[0101] Based on the position information output by each of the position positioning sensors and the state variables in the pre-determined error state Kalman filter algorithm, the Jacobian matrix of the observation equation with respect to the state variables is obtained.

[0102] The generation module 440 can process the initial pre-aiming trajectory data to obtain the target virtual trajectory information in the following manner:

[0103] The initial pre-aiming trajectory data is substituted into the Jacobian matrix of the state variables in the observation equation, and the Kalman gain is calculated for each observation to obtain the error state estimate of the state variables. The observations include the position information output by each of the position positioning sensors. The state variables are used to characterize the state variables involved in the operation of the rail vehicle.

[0104] Based on the error state estimation and the state quantity, the corrected state quantity is obtained;

[0105] Based on the corrected state variables, the target virtual orbit information is obtained.

[0106] In an exemplary embodiment of the present invention, the generation module 440 can obtain the target virtual orbit information based on the corrected state variables in the following manner:

[0107] Based on the corrected state variables, the target real-time position information of the rail vehicle is determined, wherein the target real-time position information is used to characterize the real-time position information of the rail vehicle with an accuracy higher than an accuracy threshold.

[0108] The target's real-time location information is compared with the standard route data in the target reference route to obtain the target's virtual track information.

[0109] In an exemplary embodiment of the present invention, the processing module 430 may determine the initial pre-aiming driving trajectory data formed under each of the location positioning sensors based on the real-time location information and the target reference route corresponding to the real-time location information in the following manner:

[0110] The real-time location information is compared with the standard route data in the target reference route to obtain the initial pre-aiming driving trajectory data formed under each of the location positioning sensors.

[0111] In an exemplary embodiment of the present invention, the rail vehicle is further provided with a vision sensor; the processing module 430 may also be configured to:

[0112] The reference distance from the lane center is determined based on the visual sensor, wherein the reference distance from the lane center is used to characterize the distance from the virtual track center of the rail vehicle during normal operation;

[0113] The processing module 430 can determine the target reference routes formed by the rail vehicle during its travel along the virtual track center based on each of the aforementioned position positioning sensors in the following manner:

[0114] Based on each of the aforementioned position positioning sensors, the reference routes formed by the rail vehicle during its travel along the virtual track center are determined.

[0115] Based on the deviation from the lane center reference distance, each of the reference routes is processed to obtain each target reference route formed by the rail vehicle traveling along the virtual track center under each of the position positioning sensors.

[0116] In an exemplary embodiment of the present invention, the plurality of position positioning sensors include a combined navigation device, a lidar, and a magnetic sensor; the reference route includes a first reference route corresponding to the combined navigation device, a second reference route corresponding to the lidar, and a third reference route corresponding to the magnetic sensor; the target reference route includes a first target reference route corresponding to the first reference route, a second target reference route corresponding to the second reference route, and a third target reference route corresponding to the third reference route.

[0117] The processing module 430 can determine the reference routes formed by the rail vehicle during its travel along the virtual track center based on each of the aforementioned position sensors in the following manner:

[0118] Based on the integrated navigation device, the first reference route formed by the rail vehicle during its journey along the virtual track center is determined;

[0119] Based on the lidar, the second reference route formed by the rail vehicle during its journey along the virtual track center is determined;

[0120] Based on the magnetic sensor, the third reference route formed by the rail vehicle during its journey along the virtual track center is determined;

[0121] The processing module 430 can perform deviation processing on each of the reference routes based on the deviation distance from the lane center reference distance in the following manner to obtain each target reference route formed by the rail vehicle traveling along the virtual track center under each of the position positioning sensors:

[0122] Based on the deviation from the lane center reference distance, each of the reference routes is deviated to obtain the first target reference route formed by the rail vehicle traveling along the virtual track center under the integrated navigation device.

[0123] Based on the deviation from the lane center reference distance, each of the reference routes is deviated to obtain the second target reference route formed by the rail vehicle traveling along the virtual track center under the lidar.

[0124] Based on the deviation from the lane center reference distance, each of the reference routes is processed to obtain a third target reference route formed by the rail vehicle traveling along the virtual track center under the magnetic sensor.

[0125] Based on the same inventive concept, the present invention also provides a rail vehicle, which will be described below in conjunction with the following embodiments.

[0126] In an exemplary embodiment of the present invention, the rail vehicle may include a rail vehicle body and a processor, wherein the processor is configured to execute the method for generating virtual track information of the rail vehicle as described in any of the preceding items. This enables the accurate generation of virtual track information, improving the precise perception capability of the autonomous driving system.

[0127] Figure 5 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530, and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540. The processor 510 can call logical instructions in the memory 530 to execute a method for generating virtual track information for a rail vehicle. The rail vehicle is equipped with multiple position sensors. The method includes: before the rail vehicle starts running, determining each target reference route formed by the rail vehicle during its travel along the center of the virtual track based on each of the position sensors; when the rail vehicle starts running, collecting real-time position information of the rail vehicle based on each of the position sensors; determining initial pre-aiming travel trajectory data formed under each of the position sensors based on the real-time position information and the target reference routes corresponding to the real-time position information; and processing each of the initial pre-aiming travel trajectory data to obtain target virtual track information, wherein the target virtual track information is used to characterize the track information on which the rail vehicle is about to travel.

[0128] Furthermore, the logical instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0129] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the method for generating virtual track information of a rail vehicle provided by the above methods. The rail vehicle is equipped with multiple position positioning sensors. The method includes: before the rail vehicle starts running, determining each target reference route formed by the rail vehicle during its travel along the center of the virtual track based on each of the position positioning sensors; when the rail vehicle starts running, collecting each real-time position information of the rail vehicle based on each of the position positioning sensors; determining initial pre-aiming travel trajectory data formed under each of the position positioning sensors based on the real-time position information and the target reference routes corresponding to the real-time position information; and processing each of the initial pre-aiming travel trajectory data to obtain target virtual track information, wherein the target virtual track information is used to characterize the track information on which the rail vehicle is about to travel.

[0130] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a method for generating virtual track information for a rail vehicle provided by the methods described above. The rail vehicle is equipped with multiple position sensors. The method includes: before the rail vehicle starts running, determining target reference routes formed by the rail vehicle during its movement along the center of the virtual track based on each of the position sensors; when the rail vehicle starts running, collecting real-time position information of the rail vehicle based on each of the position sensors; determining initial pre-aiming travel trajectory data formed under each of the position sensors based on the real-time position information and the target reference routes corresponding to the real-time position information; and processing the initial pre-aiming travel trajectory data to obtain target virtual track information, wherein the target virtual track information is used to characterize the track information on which the rail vehicle is about to travel.

[0131] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0132] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0133] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of generating virtual track information of a railway vehicle, characterized by, The rail vehicle is equipped with multiple position positioning sensors; the method includes: Before the rail vehicle starts running, each target reference route formed by the rail vehicle during its travel along the virtual track center is determined based on each of the position positioning sensors, wherein the rail vehicle is also equipped with a vision sensor; before determining each target reference route formed by the rail vehicle during its travel along the virtual track center based on each of the position positioning sensors, the method further includes: The reference distance from the lane center is determined based on the visual sensor, wherein the reference distance from the lane center is used to characterize the distance from the virtual track center of the rail vehicle during normal operation; The step of determining the target reference routes formed by the rail vehicle during its travel along the virtual track center based on each of the location sensors specifically includes: Based on each of the aforementioned position positioning sensors, the reference routes formed by the rail vehicle during its travel along the virtual track center are determined. Based on the deviation from the lane center reference distance, each of the reference routes is processed to obtain each target reference route formed by the rail vehicle traveling along the virtual track center under each of the position positioning sensors; When the rail vehicle starts to run, the real-time position information of the rail vehicle is collected in real time based on each of the position positioning sensors; Based on the real-time location information and the target reference route corresponding to the real-time location information, the initial pre-aiming driving trajectory data formed under each of the location positioning sensors is determined; The initial pre-aiming trajectory data are processed to obtain target virtual track information, wherein the target virtual track information is used to characterize the track information on which the rail vehicle is about to travel.

2. The method of claim 1, wherein, The process of processing the initial pre-aiming trajectory data to obtain the target virtual trajectory information specifically includes: The weight values ​​of each of the initial pre-aiming driving trajectory data are determined respectively, wherein the weight values ​​are determined based on the positioning accuracy of each of the position positioning sensors; The target virtual track information is obtained by performing weighted fusion processing based on the weight values ​​and the initial pre-aiming driving trajectory data.

3. The method of claim 1, wherein, Before processing the initial pre-aiming trajectory data to obtain the target virtual trajectory information, the method further includes: Based on the position information output by each of the position positioning sensors and the state variables in the pre-determined error state Kalman filter algorithm, the Jacobian matrix of the observation equation with respect to the state variables is obtained. The process of processing the initial pre-aiming trajectory data to obtain the target virtual trajectory information specifically includes: The initial pre-aiming trajectory data is substituted into the Jacobian matrix of the state variables in the observation equation, and the Kalman gain is calculated for each observation to obtain the error state estimate of the state variables. The observations include the position information output by each of the position positioning sensors. The state variables are used to characterize the state variables involved in the operation of the rail vehicle. Based on the error state estimation and the state quantity, the corrected state quantity is obtained; Based on the corrected state variables, the target virtual orbit information is obtained.

4. The method of claim 3, wherein, The process of obtaining the target virtual orbit information based on the corrected state variables specifically includes: Based on the corrected state variables, the target real-time position information of the rail vehicle is determined, wherein the target real-time position information is used to characterize the real-time position information of the rail vehicle with an accuracy higher than an accuracy threshold. The target's real-time location information is compared with the standard route data in the target reference route to obtain the target's virtual track information.

5. The method of claim 1, wherein, The step of determining the initial pre-aiming driving trajectory data formed under each of the location sensors based on the real-time location information and the target reference route corresponding to the real-time location information specifically includes: The real-time location information is compared with the standard route data in the target reference route to obtain the initial pre-aiming driving trajectory data formed under each of the location positioning sensors.

6. The method of claim 1, wherein, The plurality of position sensors include a combined navigation device, a lidar, and a magnetic sensor; the reference route includes a first reference route corresponding to the combined navigation device, a second reference route corresponding to the lidar, and a third reference route corresponding to the magnetic sensor; The target reference route includes a first target reference route corresponding to the first reference route, a second target reference route corresponding to the second reference route, and a third target reference route corresponding to the third reference route; The step of determining the reference routes formed by the rail vehicle during its journey along the virtual track center, based on each of the location sensors, specifically includes: Based on the integrated navigation device, the first reference route formed by the rail vehicle during its journey along the virtual track center is determined; Based on the lidar, the second reference route formed by the rail vehicle during its journey along the virtual track center is determined; Based on the magnetic sensor, the third reference route formed by the rail vehicle during its journey along the virtual track center is determined; The process of performing deviation processing on each of the reference routes based on the deviation distance from the lane center to obtain each target reference route formed by the rail vehicle traveling along the virtual track center under the position positioning sensors specifically includes: Based on the deviation from the lane center reference distance, each of the reference routes is deviated to obtain the first target reference route formed by the rail vehicle traveling along the virtual track center under the integrated navigation device. Based on the deviation from the lane center reference distance, each of the reference routes is deviated to obtain a second target reference route formed by the rail vehicle traveling along the virtual track center under the lidar. Based on the deviation from the lane center reference distance, each of the reference routes is processed to obtain a third target reference route formed by the rail vehicle traveling along the virtual track center under the magnetic sensor.

7. A device for generating virtual track information for rail vehicles, characterized in that, The rail vehicle is equipped with multiple position sensors; the device is used to implement the method for generating virtual track information of the rail vehicle according to any one of claims 1 to 6, and the device includes: The determination module is used to determine, based on each of the position positioning sensors, each target reference route formed by the rail vehicle during its travel along the virtual track center before the rail vehicle starts to run. The data acquisition module is used to acquire real-time position information of the rail vehicle based on the position positioning sensors when the rail vehicle starts to run. The processing module is used to determine the initial aiming driving trajectory data formed under each of the location positioning sensors based on the real-time location information and the target reference route corresponding to the real-time location information. The generation module is used to process the initial pre-aiming trajectory data to obtain target virtual track information, wherein the target virtual track information is used to characterize the track information on which the rail vehicle is about to travel.

8. A rail vehicle, characterized by The rail vehicles include: The rail vehicle itself, and A processor, wherein the processor is configured to execute the method for generating virtual track information for a rail vehicle as described in any one of claims 1 to 6.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for generating virtual track information for rail vehicles as described in any one of claims 1 to 6.

Images