Track clearance detection method and system

Abstract

The present invention provides a method and system for detecting track limits, the method comprising: loading a target point cloud map into a trackside device, and matching the target point cloud map with point cloud data acquired by a laser radar of the trackside device to obtain a first precise position and posture of the trackside device in the target point cloud map; determining the positional relationship of the track relative to the trackside device based on the first precise position and the position of the track; and determining the limit area of ​​the track based on the positional relationship and the track width. The system executes the method. The present invention achieves precise self-positioning of the trackside device by loading a high-precision target point cloud map into the trackside device, and detects track limits based on the relative relationship between the trackside device and the track position.

Description

Technical Field

[0001] The present invention relates to the field of rail transit technology, and in particular to a track clearance detection method and system. Background Art

[0002] For the track limit intrusion detection function, the algorithm principles of the vehicle-mounted and trackside detection systems are basically the same. At present, it can be realized based on lidar and the application of certain perception processing algorithms. However, the difficulty of this function lies in how to accurately judge whether the obstacle has invaded the track limit. It is difficult to accurately determine the position of the track and its limit by relying solely on lidar or camera. The farther the distance, the more difficult it is to judge.

[0003] Especially for the trackside detection system, how to accurately find the track and its limits has not been studied. The common processing method at present is to determine the track position through image recognition or LiDAR point cloud recognition based on deep learning technology.

[0004] For trackside detection systems equipped with cameras and lidar, the track position can be determined through image recognition or lidar point cloud recognition based on deep learning technology, but a large amount of data needs to be collected to train the deep learning model, and the cost of data collection and annotation is very high. For scenes that have not been trained, the deep learning model is prone to misidentification and missed recognition, and the reliability of the algorithm is limited.

[0005] If the track limits are inaccurate, the system's frequent false alarms of obstacles will affect the driver's driving experience, and missed reports of obstacles may lead to serious safety accidents. In addition, for subway scenarios, most trackside detection systems are installed in tunnels. If the lights are not turned on in the tunnel, the image recognition model cannot work properly. Compared with images, the resolution of lidar point clouds is too low, and the resolution drops rapidly as the distance increases.

[0006] In addition, the features of the track in the LiDAR point cloud are not obvious, and the detection distance of the track in the LiDAR point cloud by the deep learning model is not too far, usually not more than 80 meters. If the sensing distance is too close, the efficiency of the trackside detection system will be greatly reduced.

[0007] In addition, the safety and reliability of deep learning-based detection algorithms have not been fully proven and cannot meet or satisfy the requirements of rail transit safety integrity. Summary of the invention

[0008] The track limit detection method and system provided by the present invention are used to solve at least one problem existing in the prior art. By loading a high-precision target point cloud map into the trackside equipment, accurate self-positioning of the trackside equipment is achieved, and based on the relative relationship between the trackside equipment and the track position, the track limit detection is achieved.

[0009] The present invention provides a track clearance detection method, comprising:

[0010] Loading the target point cloud map into the trackside equipment, and matching the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment to obtain a first precise position and posture of the trackside equipment in the target point cloud map;

[0011] Determining a positional relationship of the track relative to the trackside equipment based on the first precise position and the track position;

[0012] Determining a limit area of ​​the track according to the positional relationship and the track width;

[0013] The step of matching the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment includes:

[0014] According to the initial position of the trackside equipment in the target point cloud map and the point cloud matching algorithm, the target point cloud map and the laser radar point cloud data of the trackside equipment are matched.

[0015] According to a track clearance detection method provided by the present invention, determining the positional relationship of the track relative to the trackside equipment according to the first precise posture and the position of the track includes:

[0016] The position of the track is projected into the point cloud data according to the first precise posture to obtain the positional relationship.

[0017] According to a track clearance detection method provided by the present invention, the position of the track is obtained in the following manner:

[0018] Determine the position of the track point in the target point cloud map according to the position of the center point of the bottom of the train;

[0019] The position of the track is determined according to the position of the track trajectory point in the target point cloud map.

[0020] According to a track clearance detection method provided by the present invention, the point cloud matching algorithm at least includes:

[0021] Iterative closest point ICP algorithm, general iterative closest point GICP algorithm and standard iterative closest point NICP algorithm.

[0022] According to a track clearance detection method provided by the present invention, the target point cloud map is obtained in the following manner:

[0023] Acquire the laser radar point cloud data, vehicle speed data, and angular velocity data of the train based on the onboard equipment;

[0024] Based on a simultaneous localization and mapping (SLAM) algorithm, the laser radar point cloud data, the vehicle speed data and the angular velocity data are fused and processed to determine a second precise position and posture of the train relative to a world coordinate system;

[0025] According to the second precise posture, the laser radar point cloud data is transformed to obtain the target point cloud map.

[0026] A track clearance detection method provided by the present invention also includes:

[0027] According to the limit area of ​​the track, it is judged whether the obstacle is an intrusion obstacle. The present invention also provides a track limit detection system, including: a posture determination module, a relationship determination module and a limit determination module;

[0028] The posture determination module is used to load the target point cloud map into the trackside equipment, and match the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment to obtain a first accurate posture of the trackside equipment in the target point cloud map;

[0029] The relationship determination module is used to determine the positional relationship of the position of the track relative to the trackside equipment based on the first precise posture and the position of the track;

[0030] The limit determination module is used to determine the limit area of ​​the track according to the position relationship and the track width;

[0031] The step of matching the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment includes:

[0032] According to the initial position of the trackside equipment in the target point cloud map and the point cloud matching algorithm, the target point cloud map and the laser radar point cloud data of the trackside equipment are matched.

[0033] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any of the above-mentioned track limit detection methods are implemented.

[0034] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned track clearance detection methods are implemented.

[0035] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the steps of any of the above-mentioned track clearance detection methods are implemented.

[0036] The track clearance detection method and system provided by the present invention realize accurate self-positioning of the trackside equipment by loading a high-precision target point cloud map into the trackside equipment, and realize the detection of the track clearance based on the relative relationship between the trackside equipment and the track position. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 It is one of the flow charts of the track clearance detection method provided by the present invention;

[0039] Figure 2 This is the second flow chart of the track clearance detection method provided by the present invention;

[0040] Figure 3 It is a schematic diagram of determining the first accurate position and posture of the trackside equipment in the target point cloud map based on the ICP algorithm provided by the present invention;

[0041] Figure 4 It is a schematic diagram of a process of loading a target point cloud map into a trackside device provided by the present invention;

[0042] Figure 5 It is a schematic diagram of the process of locating the position of the track provided by the present invention;

[0043] Figure 6 It is a structural schematic diagram of the track clearance detection system provided by the present invention;

[0044] Figure 7 It is a schematic diagram of the physical structure of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Figure 1 It is one of the flow charts of the track clearance detection method provided by the present invention, such as Figure 1 As shown, the method includes:

[0047] S1. Load the target point cloud map into the trackside equipment, and match the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment to obtain the first accurate position and posture of the trackside equipment in the target point cloud map;

[0048] S2. Determine the positional relationship of the track relative to the trackside equipment based on the first precise position and the track position;

[0049] S3. Determine a limit area of ​​the track according to the first positional relationship and the track width.

[0050] It should be noted that the execution subject of the above method may be a computer device.

[0051] Optionally, first, the acquired high-precision target point cloud map is loaded into the trackside equipment, and the real-time scanning point cloud (i.e., point cloud data) output by the laser radar carried by the trackside equipment is matched with the high-precision target point cloud map to find the positional relationship between the real-time scanning point cloud and the target point cloud map. The precise posture of the trackside equipment in the target point cloud map can be obtained, i.e., the first precise posture.

[0052] By loading the target point cloud map into the trackside equipment and matching the point cloud data of the trackside equipment's LiDAR with the high-precision target point cloud map, the precise position of the trackside equipment in the target point cloud map can be calculated. However, point cloud matching can only be achieved with point clouds within a small range. Therefore, before matching the trackside equipment's LiDAR point cloud data with the high-precision target point cloud map, the initial position of the trackside equipment in the point cloud map needs to be set. The initial position does not need to be too accurate, and an error of about 3 meters from the actual position can be allowed.

[0053] It should be noted that the initial position can be set specifically in the following manner. First, find the target point cloud map of the corresponding section according to the approximate installation position of the trackside equipment (for example, along the up / down direction, XX meters from XX station), and then open the target point cloud map of the section in the three-dimensional point cloud observation tool. According to the shape of the trackside equipment and the surrounding environment, find the approximate position of the trackside equipment in the target point cloud map, obtain the coordinates and approximate posture of the trackside equipment from the point cloud map, and use the coordinates and approximate posture of the trackside equipment as the initial position.

[0054] Based on the initial position and the point cloud matching algorithm, the laser radar point cloud data of the trackside equipment is matched with the high-precision target point cloud map to obtain the first accurate position and posture of the trackside equipment in the target point cloud map, thereby achieving accurate positioning of the trackside equipment.

[0055] Trackside equipment (such as trackside detection equipment) is a fixed detection device installed on the ground beside the track. It can detect the track limits in specific areas on the track line to confirm whether there are foreign objects or obstacles invading the track limits and prevent the occurrence of safety accidents.

[0056] After detecting an intruding foreign object, the trackside equipment sends the detection results to the on-board obstacle detection system approaching via wireless communication, allowing the vehicle to take braking measures in advance, significantly improving the detection distance of the on-board obstacle detection system.

[0057] Secondly, based on the first accurate position and posture of the trackside equipment in the high-precision target point cloud map, the position relationship of the track relative to the trackside equipment can be obtained.

[0058] Finally, the track limit area can be determined based on the position of the track relative to the trackside equipment and by extending the track position to both sides by a certain width (track width). With this track limit area, it is possible to accurately determine whether an obstacle is an intrusive obstacle.

[0059] Track limits are the outlines of buildings, equipment and rolling stock that cannot be crossed on the line to ensure transportation safety. Both onboard and trackside obstacle detection systems need to detect obstacles in the track limit area in front of the vehicle. When an intruding obstacle is found, the driver is promptly reminded or the vehicle is directly controlled to brake.

[0060] The track clearance detection method provided by the present invention realizes accurate self-positioning of the trackside equipment by loading a high-precision target point cloud map into the trackside equipment, and realizes the detection of the track clearance based on the relative relationship between the trackside equipment and the track position.

[0061] Furthermore, in one embodiment, step S2 may specifically include:

[0062] S21. Projecting the position of the track into the point cloud data according to the first precise posture to obtain a positional relationship.

[0063] Furthermore, in one embodiment, the point cloud matching algorithm at least includes:

[0064] Iterative closest point ICP algorithm, general iterative closest point GICP algorithm and standard iterative closest point NICP algorithm. Optionally, the location of the track trajectory point is also the location of the track. One of the main purposes of establishing a high-precision target point cloud map is to find the location of the track. Therefore, after the target point cloud map is established, the location of the source track is marked in the point cloud map through the target. The marking method is that during the vehicle mapping process, after each frame of point cloud data is spliced ​​and mapped, the center point of the bottom of the vehicle is marked as a trajectory point. When the mapping is completed, the track trajectory point is also marked, that is, the coordinates of the track trajectory and the coordinates of the target point cloud map are unified, both of which are map coordinate systems. Therefore, the position relationship of the track trajectory point relative to the target point cloud map is fixed and known. To know the position relationship of the track relative to the trackside equipment, it is necessary to know the precise position of the trackside equipment relative to the target point cloud map.

[0065] See also Figure 2 Based on the initial position of the trackside equipment in the target point cloud map, the laser radar point cloud data of the trackside equipment and the target point cloud map are matched through the point cloud matching algorithm, and the precise position and posture of the trackside equipment in the target point cloud map can be calculated, that is, the first precise position and posture. Specifically:

[0066] The point cloud matching algorithm may be any one of the iterative closest point (ICP) algorithm, the general iterative closest point (GICP) algorithm and the standard iterative closest point (NICP) algorithm. By using any of the above point cloud matching algorithms, the laser radar point cloud data of the trackside equipment and the target point cloud map are matched. The following is an explanation using the ICP algorithm as an example:

[0067] See also Figure 3 , assuming that A is the point cloud data obtained by the laser radar of the trackside equipment, and B is the target point cloud map:

[0068] Step 1: For each point in A, find a point in B with the closest spatial distance as the corresponding point to form new point clouds α and β. α is a subset of A, and β is a subset of B. The number of points in α and β is exactly the same, and they correspond one to one.

[0069] Step 2: Calculate the optimal rigid body transformation between α and β so that the average spatial distance between the corresponding points of α and β is minimized, and obtain the translation t and rotation R;

[0070] Step 3: Translate A by t and rotate it by R to obtain point cloud A′;

[0071] Step 4: If the decrease in the average spatial distance between the corresponding points of α and β relative to the previous iteration is less than a certain threshold, or the upper limit of the number of iterations is reached, then exit the iteration; otherwise, return to step 1, replace A with A′, and continue iterating.

[0072] Finally, the translation and rotation in all iterations are superimposed to obtain the total translation and total rotation. The total translation and total rotation are the relative position relationship between the point cloud data (radar coordinate system) and the target point cloud map (map coordinate system), that is, the precise position and posture of the trackside equipment in the point cloud map, that is, the first precise posture. The first precise posture of the trackside equipment relative to the target point cloud map is determined based on the result of matching the target point cloud map and the point cloud data obtained by the laser radar of the trackside equipment. By projecting the track trajectory point (that is, the position of the track) onto the scanned point cloud according to the precise posture, the position relationship of the track trajectory relative to the trackside equipment can be determined. Projection method: Based on the first precise posture of the trackside equipment relative to the target point cloud map, the track trajectory is transformed from the map coordinate system to the radar coordinate system corresponding to the scanned point cloud (that is, point cloud data), which completely overlaps with the point cloud data.

[0073] The track clearance detection method provided by the present invention determines the positional relationship of the track relative to the trackside equipment based on the position of the track and the precise position and posture of the trackside equipment relative to the target point cloud map, laying a foundation for subsequent track clearance detection based on the positional relationship.

[0074] Further, in one embodiment, the position of the track is obtained by:

[0075] According to the position of the center point of the bottom of the train, determine the position of the track point in the target point cloud map;

[0076] The position of the track is determined according to the position of the track trajectory point in the target point cloud map.

[0077] Optionally, one of the main purposes of establishing a high-precision target point cloud map is to find the location of the track. Therefore, after establishing the high-precision target point cloud map, the location of the track is marked in the target point cloud map, and the three-dimensional coordinates of the track are exported after marking to form a track trajectory, so that the location of the track can be obtained.

[0078] Among them, the three-dimensional coordinates of the track trajectory and the coordinates of the target point cloud map are unified (map coordinate system), but in order to facilitate the subsequent determination of the track area, the track trajectory and the target point cloud map are saved separately.

[0079] The marking method is that during the process of building the target point cloud map, each time a frame of point cloud data is stitched and mapped, the center point of the bottom of the train is marked as a track point. When the construction of the target point cloud map is completed, the track point is also marked.

[0080] The position of the marked track track point is then used as the position of the track.

[0081] The track limit detection method provided by the present invention obtains the position of the track based on the constructed high-precision target point cloud map, thereby realizing self-positioning of the track position. At the same time, since the process does not rely on deep learning technology and the algorithm implementation method is theoretically explainable, it can meet the requirements of rail transit safety integrity.

[0082] Furthermore, in one embodiment, the target point cloud map is obtained by:

[0083] Acquire the laser radar point cloud data, vehicle speed data, and angular velocity data of the train based on the onboard equipment;

[0084] Based on the simultaneous localization and mapping SLAM algorithm, the laser radar point cloud data, vehicle speed data and angular velocity data are fused and processed to determine the second most accurate position of the train relative to the world coordinate system;

[0085] According to the second precise pose, the lidar point cloud data is transformed to obtain a target point cloud map.

[0086] Optionally, the track limit detection method proposed in the present invention first uses on-board equipment to establish a high-precision electronic map of the track line (i.e., a target point cloud map), and then loads the target point cloud map into the trackside equipment to achieve accurate self-positioning of the trackside equipment, track line positioning, and vehicle limit area detection.

[0087] Among them, high-precision target point cloud maps can be produced by collecting data through vehicle-mounted laser radar, inertial measurement unit (IMU) and speed sensor (such as millimeter wave radar, etc.), and then completed through offline tool software. The quality of map construction is not affected by light brightness, and it is also applicable to tunnel scenes. Specifically:

[0088] See also Figure 4 First, the on-board equipment collects the original data of the track line, including the laser radar point cloud data, train angular velocity data, and train speed data of the train. Then, based on these data, a high-precision target point cloud map is produced. Finally, the high-precision target point cloud map is loaded into the trackside equipment to realize the self-positioning of the trackside equipment and the track limit detection.

[0089] The raw data can be collected through the following process:

[0090] It is necessary to collect the train's lidar point cloud data, train speed data, and train angular velocity data through on-board equipment to create a high-precision target point cloud map.

[0091] Train speed: Speed ​​information can be obtained through vehicle axle counter or millimeter wave radar speed measurement algorithm.

[0092] Train angular velocity: The IMU sensor mainly provides the angular velocity of the train to calculate the vehicle's posture changes.

[0093] LiDAR point cloud data of the train: The on-board radar point cloud is collected by the LiDAR installed at the front end of the train.

[0094] The construction of high-precision target point cloud maps can be achieved through the simultaneous positioning and mapping (SLAM) algorithm, such as Figure 5 As shown, after the map is built, the track trajectory is marked in the target point cloud map to obtain the position of the track.

[0095] Then, based on the simultaneous positioning and mapping SLAM algorithm, the lidar point cloud data, vehicle speed data and IMU angular velocity data are fused and processed to determine the second most accurate position of the train relative to the world coordinate system;

[0096] And according to the second precise pose, the laser radar point cloud data is transformed to obtain the target point cloud map. Specifically:

[0097] There are two main purposes of producing high-precision target point cloud maps. One is to accurately describe the direction and range of the track by establishing a three-dimensional point cloud map model of the track scene. The other is to store the three-dimensional feature information of the scene in the form of a point cloud so that the trackside equipment can calculate the precise position and posture of the trackside equipment in the target point cloud map through point cloud matching.

[0098] The establishment of the target point cloud map can be achieved through the laser SLAM algorithm. By fusing multiple sensor data such as the train speed, train angular velocity, and the train's lidar point cloud, the real-time and accurate posture calculation of the vehicle relative to the world coordinate system (i.e., the second precise posture) is realized. The train's lidar point cloud is then transformed using the calculated second precise posture, and then the target point cloud map is obtained after downsampling and storage.

[0099] The track limit detection method provided by the present invention is not affected by light conditions in the detection of track limit areas. As long as a high-precision target point cloud map is established, accurate limits of track areas with almost infinite distances can be achieved. Since no deep learning technology is involved, the principle can be explained and the requirements of rail transit safety integrity can be met. At the same time, the present invention has low implementation costs and only requires vehicle speed, IMU, and lidar. There is no need for large amounts of data collection, labeling, and other costs. There is basically no need for subsequent optimization and maintenance, and there is no need for the support of a high-performance GPU.

[0100] Furthermore, in one embodiment, after step S3, the following steps may be specifically included:

[0101] S4. Determine whether the obstacle is an intrusive obstacle based on the limit area of ​​the track.

[0102] Optionally, after knowing the position of the track, it is only necessary to expand a certain width (i.e., track width) on both sides of the track to determine the limit area of ​​the track. With the track limit area, it is possible to accurately determine whether the obstacle found in the scan point cloud is an intrusion obstacle.

[0103] The track limit detection method provided by the present invention realizes accurate self-positioning of the trackside equipment by loading a high-precision target point cloud map into the trackside equipment, and realizes the detection of track limits based on the relative relationship between the trackside equipment and the track position, so that it is possible to accurately determine whether an obstacle is an intrusive obstacle, thereby ensuring the safe operation of the train.

[0104] The track clearance detection system provided by the present invention is described below. The track clearance detection system described below and the track clearance detection method described above can be referred to each other.

[0105] Figure 6 Schematic diagram of the structure of the track clearance detection system provided by the present invention. Figure 6 As shown, it includes: a posture determination module 610, a relationship determination module 611 and a limit determination module 612;

[0106] A posture determination module 610 is used to load the target point cloud map into the trackside equipment, and match the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment to obtain a first accurate posture of the trackside equipment in the target point cloud map;

[0107] A relationship determination module 611 is used to determine the position relationship of the track relative to the trackside equipment according to the first precise position and the position of the track;

[0108] A limit determination module 612, for determining a limit area of ​​the track according to the position relationship and the track width;

[0109] Among them, the target point cloud map and the point cloud data obtained by the laser radar of the trackside equipment are matched, including:

[0110] According to the initial position of the trackside equipment in the target point cloud map and the point cloud matching algorithm, the target point cloud map and the lidar point cloud data of the trackside equipment are matched.

[0111] The track clearance detection system provided by the present invention realizes accurate self-positioning of the trackside equipment by loading a high-precision target point cloud map into the trackside equipment, and realizes the detection of the track clearance based on the relative relationship between the trackside equipment and the track position.

[0112] Figure 7 is a schematic diagram of the physical structure of an electronic device provided by the present invention, such as Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 711, a memory 712 and a bus 713, wherein the processor 710, the communication interface 711 and the memory 712 communicate with each other through the bus 713. The processor 710 may call the logic instructions in the memory 712 to execute the following method:

[0113] Load the target point cloud map into the trackside equipment, and match the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment to obtain the first accurate position and posture of the trackside equipment in the target point cloud map;

[0114] Determine the positional relationship of the track relative to the trackside equipment based on the first precise position and the track position;

[0115] Determine the limit area of ​​the track based on the position relationship and track width;

[0116] Among them, the target point cloud map and the point cloud data obtained by the laser radar of the trackside equipment are matched, including:

[0117] According to the initial position of the trackside equipment in the target point cloud map and the point cloud matching algorithm, the target point cloud map and the lidar point cloud data of the trackside equipment are matched.

[0118] In addition, the logic instructions in the above-mentioned memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer power screen (which can be a personal computer, a server, or a network power screen, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.

[0119] Furthermore, the present invention discloses a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the track clearance detection method provided by the above-mentioned method embodiments, for example, including:

[0120] Load the target point cloud map into the trackside equipment, and match the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment to obtain the first accurate position and posture of the trackside equipment in the target point cloud map;

[0121] Determine the positional relationship of the track relative to the trackside equipment based on the first precise position and the track position;

[0122] Determine the limit area of ​​the track based on the position relationship and track width;

[0123] Among them, the target point cloud map and the point cloud data obtained by the laser radar of the trackside equipment are matched, including:

[0124] According to the initial position of the trackside equipment in the target point cloud map and the point cloud matching algorithm, the target point cloud map and the lidar point cloud data of the trackside equipment are matched.

[0125] On the other hand, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to perform the track clearance detection method provided in the above embodiments, for example, including:

[0126] Load the target point cloud map into the trackside equipment, and match the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment to obtain the first accurate position and posture of the trackside equipment in the target point cloud map;

[0127] Determine the positional relationship of the track relative to the trackside equipment based on the first precise position and the track position;

[0128] Determine the limit area of ​​the track based on the position relationship and track width;

[0129] Among them, the target point cloud map and the point cloud data obtained by the laser radar of the trackside equipment are matched, including:

[0130] According to the initial position of the trackside equipment in the target point cloud map and the point cloud matching algorithm, the target point cloud map and the lidar point cloud data of the trackside equipment are matched.

[0131] The system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0132] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer power screen (which can be a personal computer, a server, or a network power screen, etc.) to execute the methods described in each embodiment 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, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A track clearance detection method, characterized in that: include: Loading the target point cloud map into the trackside equipment, and matching the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment to obtain a first precise position and posture of the trackside equipment in the target point cloud map; Determining a positional relationship of the track relative to the trackside equipment based on the first precise position and the track position; Determining a limit area of ​​the track according to the positional relationship and the track width; The step of matching the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment includes: Matching the target point cloud map with the laser radar point cloud data of the trackside equipment according to the initial position of the trackside equipment in the target point cloud map and a point cloud matching algorithm; The target point cloud map is obtained in the following way: Acquire the laser radar point cloud data, vehicle speed data, and angular velocity data of the train based on the onboard equipment; Based on a simultaneous localization and mapping (SLAM) algorithm, the laser radar point cloud data, the vehicle speed data and the angular velocity data are fused and processed to determine a second precise position and posture of the train relative to a world coordinate system; According to the second precise posture, the laser radar point cloud data is transformed to obtain the target point cloud map.

2. The track clearance detection method according to claim 1, characterized in that: Determining the positional relationship of the track relative to the trackside equipment according to the first precise posture and the track position includes: The position of the track is projected into the point cloud data according to the first precise posture to obtain the positional relationship.

3. The track clearance detection method according to claim 2, characterized in that: The position of the track is obtained as follows: Determine the position of the track point in the target point cloud map according to the position of the center point of the bottom of the train; The position of the track is determined according to the position of the track trajectory point in the target point cloud map.

4. The track clearance detection method according to claim 1, characterized in that: The point cloud matching algorithm at least includes: Iterative closest point ICP algorithm, general iterative closest point GICP algorithm and standard iterative closest point NICP algorithm.

5. The track clearance detection method according to any one of claims 1 to 4, characterized in that: Also includes: According to the limit area of ​​the track, it is determined whether the obstacle is an intrusive obstacle.

6. A track clearance detection system, characterized in that: include: A posture determination module, a relationship determination module, and a limit determination module; The posture determination module is used to load the target point cloud map into the trackside equipment, and match the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment to obtain a first accurate posture of the trackside equipment in the target point cloud map; The relationship determination module is used to determine the positional relationship of the position of the track relative to the trackside equipment based on the first precise posture and the position of the track; The limit determination module is used to determine the limit area of ​​the track according to the position relationship and the track width; The step of matching the target point cloud map with the point cloud data acquired by the laser radar of the trackside equipment includes: Matching the target point cloud map with the laser radar point cloud data of the trackside equipment according to the initial position of the trackside equipment in the target point cloud map and a point cloud matching algorithm; The target point cloud map is obtained in the following way: Acquire the laser radar point cloud data, vehicle speed data, and angular velocity data of the train based on the onboard equipment; Based on a simultaneous localization and mapping (SLAM) algorithm, the laser radar point cloud data, the vehicle speed data and the angular velocity data are fused and processed to determine a second precise position and posture of the train relative to a world coordinate system; According to the second precise posture, the laser radar point cloud data is transformed to obtain the target point cloud map.

7. An electronic device comprising a processor and a memory storing a computer program, characterized in that: When the processor executes the computer program, the steps of the track clearance detection method according to any one of claims 1 to 5 are implemented.

8. A processor-readable storage medium, characterized in that: The processor-readable storage medium stores a computer program, and the computer program is used to enable the processor to execute the steps of the track clearance detection method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the track clearance detection method according to any one of claims 1 to 5 are implemented.

Images