A speed measurement positioning method and device based on SLAM modeling and a medium

Abstract

The application relates to a speed measurement positioning method and device based on SLAM modeling and a medium, the method is realized based on SLAM and beacon information fusion, the method comprises the following steps: finding possible beacon objects in a laser radar point cloud, and receiving beacon information from a vehicle-mounted signal system; if the beacon information is read within the expected distance, it is confirmed that the object recognized by the laser radar is a beacon, and the positioning error is calculated according to the train reckoning position information and the actual train position information when the beacon is initially detected; and the positioning error is removed when the train position is updated subsequently. Compared with the prior art, the application has the advantages of greatly improving the speed measurement positioning accuracy of the vehicle and the like.

Description

Technical Field

[0001] This invention relates to rail transit signaling systems, and in particular to a speed measurement and positioning method, device, and medium based on SLAM (Simultaneous Localization and Mapping) modeling. Background Technology

[0002] With the development and maturation of LiDAR technology, cost reduction, and performance improvement, LiDAR has become an essential configuration for rail transit trains. LiDAR can detect obstacles in the track area ahead, replacing the driver's lookout function or reducing the driver's workload. Compared to autonomous driving in cars, rail transit requires a longer detection distance. Because it travels on a fixed line, the track area can be determined in advance using mapping. When the train reaches a known location, it can accurately obtain the track area ahead and then determine whether there are LiDAR reflection points within the track area, thereby identifying any obstacles.

[0003] The method of obtaining the track area based on positioning requires extremely high train positioning accuracy, especially at track curves. A deviation of 1-2 meters can cause the track area to overlap with tunnel walls, platforms, etc., resulting in false alarms. Therefore, it is necessary to study train speed measurement and positioning technology to improve the accuracy of train speed measurement and positioning. Speed ​​measurement and positioning technology is a fundamental function in the field of rail transit signaling. Traditional speed measurement and positioning sensors include coded odometers, Doppler radar, accelerometers, or IMUs, used to measure the instantaneous speed of trains. The absolute position of the train is obtained using track circuits, satellite receivers, beacons, and other equipment, and then the position during operation is calculated based on the speed information. Due to insufficient accuracy of speed sensors, inherent drift, or relative slippage between wheels and rails, the actual speed measurement result deviates from the train's true speed. From a safety perspective, the onboard signaling system tends to overestimate the speed value, resulting in the speed and positioning values ​​generally being larger than the true values. Directly using the positioning provided by the onboard signaling system will lead to frequent false alarms.

[0004] A search revealed a Chinese patent publication number CN118597226A that discloses a high-precision positioning method for rail vehicles based on multimodal fusion. This method integrates image features, millimeter-wave radar data, and lidar point cloud data, and uses point cloud matching to correct the cumulative error. Although this scheme can improve positioning accuracy, the data acquisition of each sensor needs to be precisely synchronized during the multi-sensor data fusion process, resulting in high system complexity.

[0005] With the addition of a LiDAR sensor, SLAM technology based on LiDAR point cloud processing can be used to achieve real-time train mapping and positioning, outputting train speed and displacement, and combining the initial positioning to obtain the train's position at any given time. Chinese Patent Publication No. CN112455502A discloses a LiDAR-based train positioning method, which creates a 3D point cloud map and matches the structured features extracted from the current LiDAR scan data with the feature map to calculate the train's current position. However, SLAM technology introduces errors in pose transformation estimation for each frame during mapping, and these errors accumulate over long periods. Common methods to eliminate these errors rely on loop closure detection, but rail transit routes are unidirectional straight lines and curves, making loop closure impossible. Therefore, other methods are needed to eliminate accumulated errors, such as reading beacon information.

[0006] Chinese patent publication CN119879903A discloses a fusion positioning method based on beacon correction. This method predicts the train's current pose using point cloud data monitored by lidar and train status information detected by IMU, then corrects the predicted pose using beacon odometer values ​​detected by beacons, and synchronizes the sensor data based on the time difference between the timestamp of the received beacon and the timestamp of the lidar-monitored point cloud data. While this scheme can correct positioning errors, it relies on a precise timestamp synchronization mechanism. Because the beacon information is sent by the onboard signaling system, its communication cycle with the active obstacle detection system is asynchronous, making it impossible to obtain accurate corresponding times. At normal train speeds, a difference of one detection cycle can cause an error of 2 to 3 meters, making it unsuitable for direct use.

[0007] Furthermore, in areas with weak texture, such as long tunnels, environmental features degrade severely, causing the speed and accuracy of SLAM technology to decrease or even fail through point cloud registration calculations, resulting in positioning drift.

[0008] Therefore, a new train speed measurement and positioning technology is needed to effectively integrate lidar SLAM positioning and beacon absolute positioning without requiring precise time synchronization, eliminate SLAM cumulative errors, and solve the positioning problem in weak texture environments, thereby improving the accuracy of train speed measurement and positioning and reducing false alarms in obstacle detection. Summary of the Invention

[0009] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a speed measurement and positioning method, device and medium based on SLAM modeling, which greatly improves the speed measurement and positioning accuracy of vehicles.

[0010] The objective of this invention can be achieved through the following technical solutions: According to a first aspect of the present invention, a velocity measurement and positioning method based on SLAM modeling is provided, the method being implemented based on the fusion of SLAM and beacon information, the method comprising: Search for potential beacon objects in the lidar point cloud and receive beacon information from the vehicle signal system; If beacon information is read within the expected distance, the object identified by the lidar is confirmed to be a beacon, and the positioning error is calculated based on the train's estimated position information when the beacon was initially detected and the train's actual position information. The positioning error will be removed when the train position is updated subsequently.

[0011] As a preferred technical solution, the method specifically includes a positioning calculation process and a beacon detection process.

[0012] As a preferred technical solution, the positioning calculation process specifically includes: Step S101: Upon power-on, acquire the positioning position L of the vehicle signal system. c And this point is used as the 0 point of the SLAM mapping, L0=L c ; Step S102: In the next cycle, the point cloud of the current cycle is acquired, preprocessed, and then matched. The pose transformation between the two frames is calculated, and the three-dimensional motion is projected as a one-dimensional displacement D1 in the depth direction. The position of the current cycle is set to L1=L0+D1. Step S103: In the subsequent i-th cycle, continue to collect the point cloud data of the current cycle and calculate D. i And set the position of the corresponding cycle to L i =L i-1 + D i .

[0013] As a preferred technical solution, in step S103, if the pose transformation calculation times out or fails to converge, the displacement D of the current period is calculated by multiplying the speed transmitted by the vehicle signal system in the previous cycle by the cycle duration. i D i = V c *ΔT, where V c ΔT represents the speed of the vehicle signaling system, and ΔT represents the period duration.

[0014] As a preferred technical solution, in step S103, if the train speed is lower than a set low threshold, the displacement D for the current cycle is calculated by multiplying the speed sent by the previous onboard signal system by the cycle duration. i .

[0015] As a preferred technical solution, the beacon detection process includes: Step S201: In each cycle, filter the collected point cloud to see if there are any point cloud targets similar to beacons. If so, calculate the beacon's position L based on the current positioning. B =Li + L, and add a displacement accumulation parameter L for representing the moving distance of the train after recognizing the beacon a , and save the corresponding information L B 、L i 、L and L a into the beacon point cloud to-be-verified list, where L i is the train position in this period, and L is the distance from the beacon to the lidar calculated from the lidar point cloud; Step S202, in the next period, check whether a beacon message from the on-vehicle signal system is received. If so, parse out the absolute position L of the beacon Ba , and search for a similar beacon record in the to-be-verified list. If |L Ba- L B | < Threshold, then consider L B as the corresponding beacon, where Threshold is the allowable detection threshold; Meanwhile, calculate the difference E Ba - L B of L Loc , and correct this difference to the train positioning in the current period, and delete the record corresponding to L B from the to-be-verified list; Step S203, record each remaining point cloud target in the to-be-verified list, update the value of L a , add the displacement calculated in the current period to L a , and judge whether L a is greater than L + Offset, where Offset is the offset distance from the radar to the beacon antenna. If so, the point cloud target is a false target, and clear the record of this point cloud target; otherwise, continue to retain this point cloud target and wait for the next period to process.

[0016] As a preferred technical solution, in step S201, if there are multiple point cloud targets, save the point cloud targets into the to-be-verified list respectively.

[0017] As a preferred technical solution, in step S202, if the same lidar is detected in multiple periods, L B in the multi-period records all satisfy the threshold judgment, and multiple positioning errors E Loc can be calculated, then the average value calculation method of the positioning error is used to calculate the final positioning error value.

[0018] As a preferred technical solution, the value of Threshold in step S202 is less than the minimum distance between adjacent beacons.

[0019] As a preferred technical solution, the Threshold value in step S202 is greater than the set minimum threshold.

[0020] According to a second aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement the method described thereon.

[0021] According to a third aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method described thereon.

[0022] Compared with the prior art, the present invention has the following advantages: 1) This invention calculates inter-frame displacement based on point cloud feature matching using SLAM technology, which significantly improves speed measurement accuracy compared to traditional speed sensors. By receiving beacon positions and calculating beacon positioning errors, it eliminates the cumulative positioning errors calculated by SLAM technology. The positioning accuracy on straight and curved segments reaches an average error of 0.3 meters and 0.4 meters, respectively, compared to 0.8 meters and 1.5 meters for traditional methods, representing an improvement in positioning accuracy of 62.5% to 73.3%. 2) This invention can simultaneously handle situations where multiple beacons are within the detection range in the same period, as well as situations where a single beacon appears in multiple detection periods, thus increasing its versatility; 3) By integrating with the speed of the vehicle signal system, this invention expands the application range of SLAM technology in weak texture regions, and can provide effective positioning data even if SLAM displacement calculation is unusable. 4) This invention does not require precise time measurement. The vehicle signal system and the active obstacle detection system do not need to perform precise time synchronization, nor do they need to record the precise time of communication between the two parties, which reduces the difficulty of system implementation. 5) This invention effectively eliminates the accumulated error caused by long-term operation of SLAM technology by receiving external absolute position information and calculating positioning error for real-time correction. After 30 minutes of continuous operation, the accumulated error decreased from 5.2 meters to 0.6 meters, and the long-term positioning stability was improved by 88.5%. 6) This invention improves the reliability of beacon identification and the accuracy of the system by using a cumulative displacement parameter tracking mechanism to automatically remove false candidate targets when the cumulative displacement exceeds the sum of the relative distance and the preset offset and no match is found; 7) By fusing external speed information, this invention automatically switches to using vehicle speed to calculate displacement in low-speed scenes (speed below 1km / h) and weak-texture areas (SLAM speed and external speed difference greater than a preset threshold), thus expanding the applicability of SLAM technology. In weak-texture environments, the positioning error is reduced from drift of 3.5 meters to 0.5 meters. 8) This invention supports calculating multiple errors and taking the average value when the same cloud target is identified in multiple cycles. By reducing the error of a single measurement through multiple measurements, the positioning accuracy is further improved. In the scenario of low-speed approach to the beacon, the error of a single measurement is ±1.5 meters, and the error after averaging is reduced to ±0.3 meters, which improves the accuracy by 5 times. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the beacon detection process of the present invention; Figure 2 This is a schematic diagram illustrating the method for determining the relative distance between the train and the beacon according to the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present 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 the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0025] This invention proposes a novel method to enhance train positioning accuracy by fusing SLAM modeling technology with beacon information. Specifically, the method involves searching for potential beacon objects in the lidar point cloud, then receiving beacon information from the onboard signaling system. If the beacon information is detected within the expected distance, the object identified by the lidar is confirmed as a beacon. The positioning error is calculated based on the train's estimated position information at the initial beacon detection and the actual train position. This error value is then removed during subsequent train position updates, thereby reducing the error and improving the system's positioning accuracy.

[0026] A lidar sensor emits laser pulses that reflect back after hitting an object. The sensor measures the pulse's flight time to determine the distance from the object to the lidar. Two-dimensional scanning acquires a three-dimensional point cloud within the train's forward field of view, containing environmental features and distances. After acquisition, the point cloud is preprocessed to eliminate motion distortion and invalid points. Then, geometric feature points such as corner points and planes are extracted. Based on these feature points, adjacent frame point clouds are matched to calculate the relative motion of the lidar (train). Using the distance traveled in the depth direction as the train's direction of travel, the train's three-dimensional motion can be projected into one-dimensional motion—the distance traveled on the rails. Dividing this distance by the time interval between adjacent frames yields the train's speed. Accumulated speed errors are introduced when calculating the train's position, requiring periodic clearing. This is achieved by calculating the error between the accumulated and absolute positioning positions when encountering beacons and removing this error in subsequent updates.

[0027] In point clouds, beacons are typically located at a certain distance above the sleepers and have higher reflectivity. To identify beacons in the point cloud, the cloud is filtered, retaining only points close to the ground track area. Then, clustering is performed, and a filtering threshold for the cluster targets is determined based on the beacon size, thus identifying the beacons. Based on the beacon clustering results, the center of each cluster is used as the beacon's location. The track distance L from the train to the beacon is calculated based on the radar's installation location and height. Finally, the estimated beacon position L is calculated based on the train's position in the current cycle. beacon The given position of the beacon on the rail will not change, therefore the calculated position L based on the beacon... beacon And the true location of the beacon L BEACON The positioning error E can then be calculated. Loc = L BEACON - L beacon After passing the beacon, subtract E from the train's current calculated position. Loc This value allows for a more precise location.

[0028] The specific implementation process of this invention is as follows: When the system powers on, it first needs to receive the positioning position L from the vehicle signal system. c And this point is used as the 0 point for SLAM mapping. The initial position of the system is L0=L c Acquire and save the point cloud data for the current period as the point cloud for the previous frame. Set the beacon verification list to an empty list to complete initialization.

[0029] In the first cycle, the point cloud for the current cycle is acquired. After preprocessing, point cloud matching is performed, and the pose transformation between two frames is calculated. The 3D motion is projected as a 1D displacement D1 in the depth direction, and the position for this cycle is L1 = L0 + D1. If the pose transformation calculation times out or fails to converge, the displacement D1 for this cycle is calculated by multiplying the speed transmitted by the onboard signaling system in the previous cycle by the cycle length. When the train is traveling at a low speed or is stationary, the speed of the onboard signaling system is also used to avoid long-term drift in positioning.

[0030] In the subsequent i-th cycle, the point cloud of the current cycle is collected again, and D is calculated. i L i L i =L i-1 + D i When the train is running at a low speed or is stopped, the speed of the onboard signaling system is also used to avoid long-term positioning drift.

[0031] In practical applications, it was found that when the train speed is below 1 km / h, the speed calculated by SLAM is less accurate than the speed provided by the train's onboard positioning system. Therefore, a filter condition was set: when the speed V sent by the onboard signal system...c SLAM speed is not used when the train speed is below a threshold; it is only used when the train speed exceeds the threshold. Furthermore, in areas with weak texture, the SLAM speed calculated through point cloud registration is compared with the speed V of the onboard signaling system. c As the difference increases, the speed information from the vehicle signal system is also used to solve the problem of SLAM being unable to locate accurately. By setting a specific threshold, when the difference between the SLAM speed and the vehicle signal system speed exceeds the threshold, the speed of the vehicle information system is switched.

[0032] beacon detection process, such as Figure 1 As shown: Starting from the first cycle, in each cycle, the collected point cloud is filtered to check for point cloud targets similar to beacons. By setting thresholds such as the furthest and nearest detection distances and beacon size, the probability of false detections can be reduced. For example, beacons can be detected only within a range of 20 to 60 meters, where the number of beacon points in a single frame is relatively large, resulting in higher detection accuracy. After a beacon is detected, its position L is calculated based on the current location. B =L i + L, where L i The position of the train in this cycle is given by L, which is the distance from the beacon to the lidar calculated from the lidar point cloud. A displacement accumulation parameter L is added. a Current period L a Setting it to 0 indicates the cumulative distance the train has traveled since the beacon was detected; the displacement calculated in each cycle is accumulated to L. a In the middle. This information (L) B L i L, L a Save the points to the beacon point cloud to be verified list. If there are multiple point cloud targets (corresponding to multiple beacons) within a period, save each point cloud target to the to-be verified list. The beacon point cloud to-be verified list is a 1-dimensional multi-vector list, similar to Beacon-info[]={((L B L i L, L a )1, (L B L i L, L a )2, (L B L i L, L a )3, ...}, the total number is Beacon-Count. If a beacon is detected in multiple periods, an entry is added to the list of beacons to be verified (L B L k L, L a ) record, where k represents the period in which the beacon was detected (k>i). Since this record contains the same beacon, LB The values ​​should be similar, while the other three parameters vary depending on the detection cycle, and all parameters can be detected and passed during the detection process.

[0033] In the next cycle, check if a beacon message has been received from the vehicle signaling system. If a beacon message has been received, then resolve the absolute position L of the beacon. Ba And search for similar beacon records in the list of records to be verified. for (i=1 to Beacon-Count) if abs( (L Ba - Beacon-info[i]. L B Threashold Then a matching beacon is found.

[0034] If there exists a record that satisfies |L Ba- L B If |< Threshold, where Threshold is the allowed detection threshold, then L is considered to be... B The corresponding beacon, i.e. the beacon previously detected by the lidar, travels a distance L... a It was subsequently detected by the vehicle-mounted beacon antenna. L was calculated. Ba -L B The difference E Loc And correct the difference to the current cycle positioning L. n In the middle, that is, the period L n =L n-1 + D n + E Loc At the same time, remove this L from the list of items to be verified. B The corresponding record.

[0035] The Threshold value should be less than the minimum distance between adjacent beacons; otherwise, when there are multiple beacons in the list to be verified, it will be impossible to distinguish between two consecutive beacons. Threshold also reflects the allowable positioning error E for train positioning. Loc Size. If the threshold is too small, the correspondence between the lidar detection beacon and the vehicle signal system's read beacon may be lost.

[0036] Check all records in the Beacon-info[] list. If multiple records satisfy |L Ba- L B If |< Threshold, then multiple positioning error values ​​E can be calculated. Loc This situation occurs because the same beacon is detected and recorded multiple times across multiple periods. In this case, it is advisable to adjust E... LocAveraging the value multiple times is equivalent to taking multiple measurements, which can reduce the error of the beacon measurement value L and thus improve positioning accuracy.

[0037] Assuming a detection distance of 20-60 meters, a train speed of 20 m / s (72 km / h), and a detection cycle of 100 ms, there should be at least (60-20) / 20 / 0.1 = 20 records. Considering that beacon clustering detection has a certain probability of missing some records, the actual number will be slightly less. The number of records will increase when the train speed is less than 20 m / s.

[0038] We need to consider some extreme scenarios, such as a train approaching the beacon at extremely low speeds. In this case, the number of detection logs would increase significantly, and obviously, too many logs would be a burden. We could consider comparing the L of each log entry. B The value is used to confirm whether they are records for the same beacon. If L B If the deviation is less than 1 meter, a maximum of 100 records will be saved. Records exceeding this limit will not be recorded to prevent the Beacon-info[] list from growing indefinitely.

[0039] For each remaining point cloud target record in the verification list, update the L in the record. a The value of L is added to the displacement calculated in the current period. a In, that is, L a = L a +D i Check L a Is it greater than L + Offset, where Offset is the offset distance from the radar to the beacon antenna? (Reference) Figure 2 L a Accumulate in each cycle, when L a If the value is greater than L + Offset, it means the train has passed the possible beacon location, and the train's beacon antenna has moved beyond the beacon's position. In this case, if no beacon message is received, it indicates the target is a false target, and the target record should be cleared. If L... a If the value is less than L + Offset, it means the train has not yet reached the beacon position and the target should be retained for processing in the next cycle.

[0040] The above is an introduction to the method embodiments. The following embodiments using electronic devices and storage media will further illustrate the solution of the present invention.

[0041] This invention also provides an electronic device including a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) or loaded from a storage unit into a random access memory (RAM). The RAM may also store various programs and data required for device operation. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0042] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0043] The processing unit performs the various methods and processes described above, such as the methods of the present invention. For example, in some embodiments, the methods of the present invention may be implemented as computer software programs tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the CPU, one or more steps of the methods of the present invention described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute the methods of the present invention by any other suitable means (e.g., by means of firmware).

[0044] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0045] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0046] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0047] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A speed measurement and positioning method based on SLAM modeling, characterized in that, This method is based on the fusion of SLAM and beacon information, and the method includes: Search for potential beacon objects in the lidar point cloud and receive beacon information from the vehicle signal system; If beacon information is read within the expected distance, the object identified by the lidar is confirmed to be a beacon, and the positioning error is calculated based on the train's estimated position information when the beacon was initially detected and the train's actual position information. The positioning error will be removed when the train position is updated subsequently.

2. The speed measurement and positioning method based on SLAM modeling according to claim 1, characterized in that, The method specifically includes a positioning calculation process and a beacon detection process.

3. The speed measurement and positioning method based on SLAM modeling according to claim 2, characterized in that, The positioning calculation process specifically includes: Step S101: Upon power-on, acquire the positioning position L of the vehicle signal system. c And this point is used as the 0 point of the SLAM mapping, L0=L c ; Step S102: In the next cycle, the point cloud of the current cycle is acquired, preprocessed, and then matched. The pose transformation between the two frames is calculated, and the three-dimensional motion is projected as a one-dimensional displacement D1 in the depth direction. The position of the current cycle is set to L1=L0+D1. Step S103: In the subsequent i-th cycle, continue to collect the point cloud data of the current cycle and calculate D. i And set the position of the corresponding cycle to L i =L i-1 + D i .

4. The speed measurement and positioning method based on SLAM modeling according to claim 3, characterized in that, In step S103, if the pose transformation calculation times out or fails to converge, the displacement D for the current period is calculated by multiplying the speed transmitted by the vehicle signal system in the previous cycle by the cycle duration. i D i = V c ×ΔT, where V c ΔT represents the speed of the vehicle signaling system, and ΔT represents the period duration.

5. The speed measurement and positioning method based on SLAM modeling according to claim 3, characterized in that, In step S103, if the train speed is lower than a set low threshold, the displacement D for the current cycle is calculated by multiplying the speed sent by the previous onboard signal system by the cycle duration. i .

6. The speed measurement and positioning method based on SLAM modeling according to claim 2, characterized in that, The beacon detection process includes: Step S201: In each cycle, filter the collected point cloud to see if there are any point cloud targets similar to beacons. If so, calculate the beacon's position L based on the current positioning. B =L i + L, and simultaneously add a cumulative displacement parameter L to represent the distance the train has moved after the beacon has been detected. a and the corresponding information L B L i L and L a Saved to the beacon point cloud pending verification list, where L i L represents the train position for that period, and L is the distance from the beacon to the lidar calculated from the lidar point cloud. Step S202, in the next cycle, check whether a beacon message from the in-vehicle signal system is received. If so, parse out the absolute position L of the beacon Ba , and search for a similar beacon record in the list to be verified. If there is |L Ba- L B |<Threshold, then consider L B to be the corresponding beacon, where Threshold is the allowable detection threshold; Simultaneously calculate L Ba -L B The difference E Loc The difference is then corrected in the train positioning for the current cycle, and L is removed from the list of trains to be verified. B The corresponding record; Step S203: Record each remaining point cloud target in the verification list and update L. a The value of L is added to the displacement calculated in the current period. a In the middle, determine L a If the distance is greater than L+Offset, where Offset is the offset distance from the radar to the beacon antenna, and if yes, the point cloud target is a false target, and the point cloud target record is cleared; otherwise, the point cloud target is retained and awaits processing in the next cycle.

7. The speed measurement and positioning method based on SLAM modeling according to claim 6, characterized in that, In step S201, if there are multiple point cloud targets, the point cloud targets are saved to the list to be verified respectively.

8. The speed measurement and positioning method based on SLAM modeling according to claim 6, characterized in that, In step S202, if the same lidar is detected in multiple cycles, then the L in the multi-cycle records... B All meet the threshold criteria and can calculate multiple positioning errors E. Loc The final positioning error value is calculated by averaging the positioning error.

9. The speed measurement and positioning method based on SLAM modeling according to claim 6, characterized in that, In step S202, the Threshold value is less than the minimum distance between adjacent beacons.

10. The speed measurement and positioning method based on SLAM modeling according to claim 6, characterized in that, The Threshold value in step S202 is greater than the set minimum threshold.

11. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 10.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 10.

Images