Coal pile identification method, device, equipment and storage medium

Abstract

The present application discloses a method, device, equipment and storage medium for identifying coal piles, which relates to the field of automatic identification technology, including: obtaining the original point cloud data collected by a 3D laser radar, filtering the original point cloud data, and determining the corresponding original point cloud set based on the filtered original point cloud data; forming a first target point cloud set with point cloud data whose height is higher than a preset height threshold in the original point cloud set, and sorting the point cloud data in the first target point cloud set to obtain a second target point cloud set; identifying whether there is target point cloud data in the second target point cloud set that meets the preset coal pile volume conditions based on a preset clustering method, and determining the corresponding coal pile point according to the identification result. In this way, the present application can accurately identify the location of the coal pile in a closed cabin using a 3D laser radar without human intervention, and can work continuously in the harsh environment of the closed cabin with high reliability.

Description

Technical Field

[0001] The present invention relates to the field of automatic identification technology, and in particular to a coal pile identification method, device, equipment and storage medium. Background Art

[0002] In China, the coal supply of many coal-fired power plants usually follows this process: first, the coal is transported to the dock by coal carriers, then the grab ship unloader gradually grabs the coal in the ship's hold and places it on the conveyor belt, and then the conveyor belt transports the coal to the warehouse for storage in preparation for power generation. However, due to the limited grabbing range of the ship unloader, it can usually only handle the coal in the center of the ship's hold, and cannot reach the coal stuck to the ship's wall. Therefore, it is usually necessary to use a cabin cleaning machine to push the coal near the cabin wall to the center of the cabin so that the ship unloader can grab it again to ensure that the coal in the cabin is thoroughly cleaned.

[0003] At present, the operation of the cargo hold cleaning machine relies on manual labor. Workers need to work continuously for several hours under tight time constraints to clean the coal in a cargo hold. At present, the sensors that can be used for coal pile identification are mainly divided into three categories: cameras, millimeter-wave radars, and lidars. When cameras are used as sensors for cargo hold cleaning machines, they are greatly affected by environmental factors and the recognition accuracy fluctuates greatly. When millimeter-wave radars are used as sensors for cargo hold cleaning machines, the detection accuracy is limited and the judgment of the object outline is relatively vague.

[0004] In summary, how to select suitable sensors and accurately identify the location of coal piles without human intervention is a technical problem that needs to be solved urgently. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a coal pile identification method, device, equipment and storage medium, which can select a suitable sensor and accurately identify the location of the coal pile without human intervention. The specific scheme is as follows:

[0006] In a first aspect, the present application provides a coal pile identification method, which is applied to a tank cleaning machine, comprising:

[0007] Acquire original point cloud data collected by the 3D laser radar, filter the original point cloud data, and determine a corresponding original point cloud set based on the filtered original point cloud data;

[0008] The point cloud data in the original point cloud set whose height is higher than a preset height threshold form a first target point cloud set, and the point cloud data in the first target point cloud set are sorted to obtain a second target point cloud set;

[0009] Based on a preset clustering method, it is identified whether there is target point cloud data that meets the preset coal pile volume condition in the second target point cloud set, and the corresponding coal pile point is determined according to the identification result.

[0010] Optionally, before obtaining the original point cloud data collected by the 3D laser radar, the method further includes:

[0011] Controlling the tank cleaning machine to be in a stationary state, and determining a corresponding area to be identified according to the current position of the tank cleaning machine;

[0012] The area to be identified is scanned by the 3D laser radar to obtain a priori map corresponding to the area to be identified, and a map coordinate system corresponding to the priori map is constructed.

[0013] Optionally, after acquiring the original point cloud data collected by the 3D laser radar, the method further includes:

[0014] Determining the body point cloud data corresponding to the tank cleaning machine from the original point cloud data according to the body parameter information of the tank cleaning machine;

[0015] Accordingly, the filtering process of the original point cloud data includes:

[0016] The original point cloud data is filtered based on the vehicle body point cloud data to obtain the filtered original point cloud data.

[0017] Optionally, determining a corresponding original point cloud set based on the filtered original point cloud data includes:

[0018] Inputting the original point cloud data, the prior map, and the current motion state data of the tank cleaning machine into a preset 3D laser positioning program, and determining the current 6D global position of the tank cleaning machine in the map coordinate system through the preset 3D laser positioning program;

[0019] The filtered original point cloud data is converted to the map coordinate system based on each posture corresponding to the current 6D global posture of the tank cleaning machine, and the corresponding original point cloud set is determined according to the converted filtered original point cloud data.

[0020] Optionally, sorting the point cloud data in the first target point cloud set to obtain a second target point cloud set includes:

[0021] Determine the straight-line distance between each point cloud data in the first target point cloud set and the tank cleaning machine, and sort the point cloud data in the first target point cloud set in ascending order based on the straight-line distance to obtain the sorted point cloud data;

[0022] A sequence number corresponding to the sorted point cloud data is set, and a second target point cloud set is determined based on each of the sorted point cloud data and the sequence number corresponding to each of the sorted point cloud data.

[0023] Optionally, the step of identifying whether there is target point cloud data meeting a preset coal pile volume condition in the second target point cloud set based on a preset clustering method, and determining a corresponding coal pile point according to the identification result, includes:

[0024] Based on the sequence numbers corresponding to the sorted point cloud data in the second target point cloud set and a preset clustering method, one by one, identifying whether there is target point cloud data in the second target point cloud set that meets the preset coal pile volume condition;

[0025] If it is identified that the second target point cloud set contains target point cloud data that meets the preset coal pile volume condition, the identification is stopped and the corresponding coal pile point is determined according to the target point cloud data.

[0026] Optionally, the coal pile identification method further includes:

[0027] If it is determined that no collision will occur when the coal pile point is pushed and raked by the cabin cleaning machine, it is determined that the coal pile point meets the preset safe pushing and raking conditions, and the position and height corresponding to the coal pile point are determined, so that the cabin cleaning machine can push and rake the coal pile point based on the position and height corresponding to the coal pile point.

[0028] In a second aspect, the present application provides a coal pile identification device, which is applied to a cabin cleaning machine, comprising:

[0029] A point cloud data filtering module, used to obtain the original point cloud data collected by the 3D laser radar, filter the original point cloud data, and determine the corresponding original point cloud set based on the filtered original point cloud data;

[0030] a point cloud data sorting module, configured to form a first target point cloud set from point cloud data in the original point cloud set whose height is higher than a preset height threshold, and to sort the point cloud data in the first target point cloud set to obtain a second target point cloud set;

[0031] The coal pile identification module is used to identify whether there is target point cloud data that meets the preset coal pile volume conditions in the second target point cloud set based on a preset clustering method, and determine the corresponding coal pile point according to the identification result.

[0032] In a third aspect, the present application provides an electronic device, including:

[0033] Memory, used to store computer programs;

[0034] A processor is used to execute the computer program to implement the aforementioned coal pile identification method.

[0035] In a fourth aspect, the present application provides a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the aforementioned coal pile identification method is implemented.

[0036] In the present application, the tank cleaning machine first obtains the original point cloud data collected by the 3D laser radar, filters the original point cloud data, and determines the corresponding original point cloud set based on the filtered original point cloud data; then, the point cloud data in the original point cloud set whose height is higher than the preset height threshold is formed into a first target point cloud set, and the point cloud data in the first target point cloud set is sorted to obtain a second target point cloud set; finally, based on a preset clustering method, it is identified whether there is target point cloud data that meets the preset coal pile volume condition in the second target point cloud set, and the corresponding coal pile point is determined according to the identification result. As can be seen from the above, the present application collects original point cloud data through a 3D laser radar, determines the original point cloud set after filtering the original point cloud data, and selects the point cloud data that constitutes the first target point cloud set from the original point cloud set based on the preset height threshold, then re-sorts the point cloud data of the first target point cloud set to obtain the second target point cloud set, and finally identifies whether there is target point cloud data that meets the preset coal pile volume condition in the second target point cloud set to obtain a coal pile identification result. In this application, 3D laser radar is used as a sensor for coal pile identification. The 3D laser radar has a long effective detection distance and a detection accuracy within the centimeter level. In this way, this application automates the transformation of the cabin cleaning machine so that the cabin cleaning machine can operate autonomously in a closed cabin, follow the preset operating procedures, and gradually push the coal on the edge of the cabin to the center area to facilitate the subsequent grabbing work of the ship unloader. In this way, this application can accurately identify the specific outline of the obstacle without the need for human participation and without being affected by environmental factors, so that the location of the coal pile in the closed cabin can be accurately identified, and it can work continuously in the harsh environment of the closed cabin with high reliability. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0038] Figure 1 A flow chart of a coal pile identification method provided for this application;

[0039] Figure 2A specific flow chart of a coal pile identification method provided for this application;

[0040] Figure 3 A specific flow chart of a coal pile identification method provided for this application;

[0041] Figure 4 A schematic diagram of the structure of a coal pile identification device provided in this application;

[0042] Figure 5 A structural diagram of an electronic device provided for this application. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0044] At present, the operation of the cabin cleaning machine relies on manual labor, and the staff needs to work continuously for several hours under time pressure to clean the coal in a cabin. And the sensors that can be used for coal pile identification are mainly divided into three categories: cameras, millimeter-wave radars and lidars. When cameras are used as sensors for cabin cleaning machines, they are greatly affected by environmental factors and the recognition accuracy fluctuates greatly; when millimeter-wave radars are used as sensors for cabin cleaning machines, the detection accuracy is limited and the judgment of the object contour is relatively vague. To this end, the present application provides a coal pile identification solution that can select suitable sensors and accurately identify the location of the coal pile without manual participation.

[0045] See also Figure 1 As shown, an embodiment of the present invention discloses a coal pile identification method, which is applied to a tank cleaning machine and may include:

[0046] Step S11, obtaining the original point cloud data collected by the 3D laser radar, filtering the original point cloud data, and determining the corresponding original point cloud set based on the filtered original point cloud data.

[0047] In this embodiment, the original point cloud data within the target range can be collected by installing a 3D laser radar above the cab of the tank cleaning machine. Specifically, the 3D laser radar device can emit a laser beam. When the laser beam encounters an object, part of the light wave will be reflected back and captured by the receiver of the 3D laser radar. The receiver of the 3D laser radar converts the received light signal into an electrical signal to subsequently generate three-dimensional point cloud data within the target range. At the same time, the configuration parameters of the 3D laser radar, such as scanning range, resolution, scanning frequency, etc., can be adjusted according to the application requirements and working environment of the 3D laser radar. The coal pile recognition program of the tank cleaning machine can receive the laser raw point cloud data from the 3D laser radar in real time. The raw point cloud data is located in the laser sensor coordinate system.

[0048] It is understandable that before the above-mentioned acquisition of the original point cloud data collected by the 3D laser radar, it may also include: controlling the tank cleaning machine to be in a stationary state, and determining the corresponding area to be identified according to the current position of the tank cleaning machine; scanning the area to be identified by the 3D laser radar to obtain a priori map corresponding to the area to be identified, and constructing a map coordinate system corresponding to the priori map. In a specific embodiment, after the tank cleaning machine is placed in the cabin, the configuration parameters of the 3D laser radar located above the cab of the tank cleaning machine can be adjusted according to the current environmental conditions in the cabin. After that, the 3D laser radar scans the environment in the cabin by emitting a laser beam and receiving the reflected signal. When the laser beam encounters an obstacle, it will be reflected back and received and processed by the 3D laser radar. By measuring the time difference between the emission and reception of the laser beam, the 3D laser radar can calculate the distance between the obstacle and the 3D laser radar. At the same time, the 3D laser radar can record the deflection angle of the laser beam to determine the specific position and shape of the obstacle. The 3D laser radar processes the collected light signals and converts them into point cloud data, which is a series of points representing spatial positions. After filtering, denoising and registration, the point cloud data can generate a more accurate and complete 3D spatial model. Then, based on the 3D spatial model, a priori map of the cabin environment can be constructed. The priori map details the structure of the cabin, the location of obstacles, channels and openings, etc. Finally, the corresponding map coordinate system can be set on the priori map to convert the spatial position in the cabin into coordinate values. The generated priori map can be applied to the navigation, positioning, path planning and obstacle avoidance operations of the cabin cleaning machine. The accuracy and reliability of the priori map can be verified through actual operation and testing, and the priori map can be corrected and optimized as necessary based on the test results.

[0049] In this embodiment, in order to prevent the point cloud data falling on the body of the tank cleaning machine from being misjudged as coal pile points, after obtaining the original point cloud data collected by the 3D laser radar, it can also include: determining the body point cloud data corresponding to the tank cleaning machine from the original point cloud data according to the body parameter information of the tank cleaning machine; correspondingly, filtering the original point cloud data can include: filtering the original point cloud data based on the body point cloud data to obtain the filtered original point cloud data. Specifically, the coal pile recognition program first needs to obtain the body parameter information of the tank cleaning machine, such as body size information, body shape characteristics, and special marking points; wherein the special marking points include but are not limited to: wheel position, headlight position, body line intersection point, etc. Then the coal pile recognition program can set a reasonable body point cloud data range according to the body size information, and filter out the point cloud data that exceeds the body point cloud data range, so as to preliminarily screen out the body point cloud data. Afterwards, the body point cloud data can be further confirmed by using methods such as geometric shape matching and feature extraction according to the body shape characteristics, such as lines, curved surfaces, etc. Finally, the accuracy of the body point cloud data can be further confirmed based on special landmarks such as wheel positions and headlight positions. At the same time, the body point cloud data finally extracted can be smoothed to reduce noise and unevenness in the body point cloud data. After the body point cloud data corresponding to the cabin cleaning machine is determined, the corresponding body point cloud data is removed from the original point cloud data. After that, the filtered original point cloud data can be checked using visualization tools to ensure that the body point cloud data has been completely removed. If it is found that the remaining body point cloud data or other point cloud data has been deleted by mistake, the body point cloud data needs to be filtered again. In a specific embodiment, the coal pile recognition program can generate a point cloud set A based on the original point cloud data after filtering the body point cloud data in the laser sensor coordinate system.

[0050] It can be understood that in order to determine the original point cloud data after filtering the vehicle body point cloud data in the map coordinate system, the determining the corresponding original point cloud set based on the filtered original point cloud data can include: inputting the original point cloud data, the prior map, and the current motion state data of the tank cleaning machine into a preset 3D laser positioning program, and determining the current 6D global position of the tank cleaning machine in the map coordinate system through the preset 3D laser positioning program; based on the postures corresponding to the current 6D global position of the tank cleaning machine, converting the filtered original point cloud data to the map coordinate system, and determining the corresponding original point cloud set based on the converted filtered original point cloud data. Specifically, the preset 3D laser positioning program can determine the 6D global position of the tank cleaning machine in combination with the original point cloud data of the cabin, the prior map of the cabin and the current motion state data of the tank cleaning machine, and then convert the original point cloud data after filtering the body point cloud data in the laser sensor coordinate system to the map coordinate system corresponding to the prior map of the cabin based on the various poses corresponding to the current 6D global position of the tank cleaning machine, and then the preset 3D laser positioning program sends the original point cloud data after filtering the body point cloud data in the map coordinate system to the coal pile recognition program of the tank cleaning machine in real time, so that the coal pile recognition program generates the corresponding original point cloud set according to the original point cloud data after filtering the body point cloud data in the map coordinate system. In a specific implementation, the coal pile recognition program will receive the 6D global position of the tank cleaning machine in the map coordinate system sent by the preset 3D laser positioning program in real time, and use the various poses to convert the point cloud set A in the laser sensor coordinate system to the map coordinate system to obtain the point cloud set B.

[0051] Step S12: The point cloud data in the original point cloud set whose height is higher than a preset height threshold are combined into a first target point cloud set, and the point cloud data in the first target point cloud set are sorted to obtain a second target point cloud set.

[0052] In this embodiment, the coal pile identification program first filters out point cloud data with heights higher than a preset height threshold from the original point cloud set, and forms a first target point cloud set. Then, in order to screen the first target point cloud set again, the point cloud data in the first target point cloud set are sorted to obtain a second target point cloud set, which may include: determining the straight-line distance between each point cloud data in the first target point cloud set and the tank cleaning machine, and sorting each point cloud data in the first target point cloud set in an ascending order based on the straight-line distance to obtain the sorted point cloud data; setting a serial number corresponding to the sorted point cloud data, and determining the second target point cloud set based on each of the sorted point cloud data and the serial number corresponding to each of the sorted point cloud data. Specifically, the coal pile identification program can calculate the straight-line distance between each point cloud data in the first target point cloud set and the tank cleaning machine based on the Euclidean distance formula or the projection distance method. Then, the point cloud data in the first target point cloud set can be sorted in order from small to large based on the straight-line distance. When sorting, a suitable sorting algorithm can be selected, such as quick sort, merge sort and other algorithms, to improve sorting efficiency and stability. After that, it is necessary to set a serial number for the sorted point cloud data. The serial number can start from 0 and increase in sequence. That is, the point with serial number 0 is the point cloud data with the closest straight-line distance to the tank cleaning machine, and the point with the largest serial number is the point cloud data with the farthest straight-line distance from the tank cleaning machine. Finally, the second target point cloud set is generated according to the sorted point cloud data and the serial number corresponding to the sorted point cloud data. In a specific embodiment, the coal pile identification program selects point cloud data with a height higher than a preset height threshold from point cloud set B and forms point cloud set C. Then, according to the distance of the point cloud data in the point cloud set C relative to the straight-line distance of the tank cleaning machine, the points in the point cloud set C are reordered from near to far to obtain the point cloud set D.

[0053] Step S13: Identify whether there is target point cloud data that meets the preset coal pile volume condition in the second target point cloud set based on a preset clustering method, and determine the corresponding coal pile point according to the identification result.

[0054] In this embodiment, it is first necessary to select an algorithm suitable for point cloud data clustering, such as the DBSCAN algorithm (DBSCAN, i.e., Density-Based Spatial Clustering of Applications with Noise) or the K-means algorithm (K-means, i.e., K-means clustering algorithm). Then, it is necessary to set the parameters of the clustering algorithm, such as the clustering radius, the minimum number of samples corresponding to the DBSCAN algorithm, or the number of clusters K corresponding to the K-means algorithm. Then, the point cloud data in the second target point cloud set is input into the selected clustering algorithm for clustering analysis. The clustering algorithm can divide the point cloud data into multiple clusters, and determine the cluster center of the coal pile, the number of clusters, and the boundaries of each cluster based on the output of the clustering algorithm. Specifically, in order to prevent the coal pile identification program from identifying isolated and discrete point cloud data, such as point cloud data printed on coal dust, as coal pile points, the volume threshold corresponding to the preset coal pile volume condition can be set first, and then the distance from the point cloud data in the second target point cloud set to each cluster center and the distance to each cluster boundary can be calculated. If the point cloud data in a certain second target point cloud set is located inside the cluster, the point cloud data in the second target point cloud set that meets the preset coal pile volume condition can be determined as the target point cloud data, and it is determined that the target point cloud data is located in the coal pile.

[0055] As can be seen from the above, in this embodiment, the tank cleaning machine first obtains the original point cloud data collected by the 3D laser radar, and filters the original point cloud data, and determines the corresponding original point cloud set based on the filtered original point cloud data; then, the point cloud data in the original point cloud set with a height higher than a preset height threshold is formed into a first target point cloud set, and the point cloud data in the first target point cloud set is sorted to obtain a second target point cloud set; finally, based on a preset clustering method, it is identified whether there is target point cloud data that meets the preset coal pile volume condition in the second target point cloud set, and the corresponding coal pile point is determined according to the identification result. As can be seen from the above, in this embodiment, the original point cloud data is collected by the 3D laser radar, the original point cloud data is filtered to determine the original point cloud set, and the point cloud data constituting the first target point cloud set is screened out from the original point cloud set based on the preset height threshold, and then the point cloud data of the first target point cloud set is re-sorted to obtain the second target point cloud set, and finally, it is identified whether there is target point cloud data that meets the preset coal pile volume condition in the second target point cloud set to obtain the coal pile identification result. In this embodiment, a 3D laser radar is used as a sensor for coal pile identification. The 3D laser radar has a long effective detection distance and a detection accuracy within the centimeter level. In this way, this embodiment transforms the cabin cleaning machine through automation, so that the cabin cleaning machine can operate autonomously in a closed cabin, follow the preset operating procedures, and gradually push the coal on the edge of the cabin to the central area to facilitate the subsequent grabbing work of the ship unloader. In this way, this embodiment can accurately identify the specific outline of the obstacle without the need for human intervention and without being affected by environmental factors, so that the position of the coal pile in the closed cabin can be accurately identified, and it can work continuously in the harsh environment of the closed cabin, with strong reliability.

[0056] Based on the previous embodiment, it can be known that the present application can identify the coal pile point based on the point cloud data scanned by the 3D laser radar. Next, the present embodiment will elaborate on the determination of whether the identified coal pile point is a safe coal pile point that can be pushed and raked. Figure 2 As shown, the embodiment of the present invention further discloses a coal pile identification method, which is applied to a tank cleaning machine and may include:

[0057] Step S21, based on the serial numbers corresponding to the sorted point cloud data in the second target point cloud set and a preset clustering method, identify one by one whether there is target point cloud data that meets the preset coal pile volume condition in the second target point cloud set.

[0058] In this embodiment, the point cloud data in the second target point cloud set are sorted in ascending order according to the straight-line distance to the tank cleaning machine and corresponding serial numbers are set. In a specific implementation, based on the serial numbers corresponding to the point cloud data in the second target point cloud set, it is identified one by one whether there is target point cloud data in the second target point cloud set that meets the preset coal pile volume condition, and the position of the coal pile point closest to the tank cleaning machine can be determined based on the identification result.

[0059] Step S22: if it is identified that the second target point cloud set contains target point cloud data that meets the preset coal pile volume condition, then the identification is stopped and the corresponding coal pile point is determined according to the target point cloud data.

[0060] It is understandable that when identifying the point cloud data in the second target point cloud set, if the target point cloud data that meets the preset coal pile volume condition is identified in the second target point cloud set, the preset clustering method should be stopped to process the remaining point cloud data in the second target point cloud set, and the current identification result and the sequence number of the target point cloud data that meets the preset coal pile volume condition should be recorded. The coal pile identification program can determine the corresponding coal pile point based on the target point cloud data, and determine the corresponding coal pile shape, coal pile size and coal pile position.

[0061] Step S23: If it is determined that no collision will occur when the coal pile point is pushed and raked by the cabin cleaning machine, it is determined that the coal pile point meets the preset safe pushing and raking conditions, and the position and height corresponding to the coal pile point are determined, so that the cabin cleaning machine can push and rake the coal pile point based on the position and height corresponding to the coal pile point.

[0062] In this embodiment, it can be determined whether the body of the tank cleaning machine will collide with the cabin bulkhead when the tank cleaning machine pushes and raks the coal pile point based on the body parameter information of the tank cleaning machine and the prior map of the cabin. If a collision will occur, the coal pile point is determined to be an unsafe coal pile point that is close to the cabin bulkhead, because pushing and raking the coal pile point will cause damage to the tank cleaning machine and the cabin; if no collision will occur, the coal pile point is determined to be a coal pile point that can be pushed and raked safely. If the coal pile point is determined to be a coal pile point that can be pushed and raked safely, the position and height of the coal pile point can be determined based on the map coordinate system corresponding to the prior map of the cabin, so that the subsequent tank cleaning machine pushes and raks the coal pile point based on the position and height corresponding to the coal pile point.

[0063] In a specific implementation, for the point cloud data in the point cloud set D, it can be judged from sequence number 0 whether the preset coal pile volume condition and the preset safety pushing and raking condition are satisfied at the same time. If, when judging the point cloud data with sequence number i in the point cloud set D, it is determined that the point cloud data with sequence number i satisfies the preset coal pile volume condition and the preset safety pushing and raking condition at the same time, then the position of the point cloud data with sequence number i is the position of the coal pile point finally found, and the loop ends; if, when judging the point cloud data with sequence number i in the point cloud set D, the point cloud data with sequence number i does not satisfy the preset coal pile volume condition and the preset safety pushing and raking condition at the same time, then it is judged whether the point cloud data with sequence number i+1 satisfies the preset coal pile volume condition and the preset safety pushing and raking condition at the same time. If, until the last point cloud data in the point cloud set D, there is no point cloud data that satisfies the preset coal pile volume condition and the preset safety pushing and raking condition at the same time, then it means that there is no coal pile point that can be pushed and raked in the environment around the cabin cleaning machine. After the above judgment process, the coal pile point finally obtained is the nearest coal pile point position in the point cloud set D that has a certain volume and can be pushed and raked by the cabin cleaning machine.

[0064] See also Figure 3 As shown, in a specific implementation manner, the process of the coal pile identification method may specifically be:

[0065] First, the coal pile recognition program receives the original laser point cloud data from the 3D laser radar and removes the laser point cloud data on the vehicle body to obtain point cloud set A. Then the coal pile recognition program receives the 3D laser positioning program based on IMU data (IMU, i.e. Inertial Measurement Unit) in real time. The 6D global pose of the tank cleaning machine in the map coordinate system is determined by the prior map of the cabin environment and the 3D laser original point cloud data, and the original laser point cloud data is converted to the map coordinate system using the pose to obtain a point cloud set B; then the laser point cloud data above the minimum height threshold of the coal pile is cut out to obtain a point cloud set C; the laser points in the point cloud set C are sorted from near to far according to the straight-line distance of the laser points relative to the tank cleaning machine to obtain a point cloud set D; according to the arranged order, a clustering method is used to determine in turn whether each point cloud in the point cloud set D is a point in a coal pile with a certain volume, and the point found is the coal pile point identified on the running path of the tank cleaning machine; it is determined whether the tank cleaning machine will collide with the cabin wall when pushing and raking the coal pile point, and if there is no collision, the coal pile point that can be pushed and raked is identified; the position and height of the coal pile point that can be pushed and raked in the map coordinate system are determined.

[0066] As can be seen from the above, this embodiment can identify the point cloud data in the second target point cloud set one by one according to the serial number to determine whether it is the target point cloud data that meets the preset coal pile volume condition. If the target point cloud data is identified in the second target point cloud set, it is determined whether the target point cloud data meets the preset safe pushing and raking condition. In this way, if the point cloud data in the second target point cloud set meets both the preset coal pile volume condition and the preset safe pushing and raking condition, the position of the nearest coal pile point that can be pushed and raked by the cabin cleaning machine is identified. In this way, the coal pile identification program not only identifies and determines whether there is a coal pile that can be pushed and raked near the cabin cleaning machine, but also determines the position and height of the nearest coal pile point identified in the map coordinate system, so that the cabin cleaning machine can be navigated to the position of the coal pile to push and rake the coal pile.

[0067] Accordingly, see Figure 4 As shown, the embodiment of the present application also provides a coal pile identification device, which is applied to a cabin cleaning machine and may include:

[0068] The point cloud data filtering module 11 is used to obtain the original point cloud data collected by the 3D laser radar, filter the original point cloud data, and determine the corresponding original point cloud set based on the filtered original point cloud data;

[0069] A point cloud data sorting module 12 is used to form a first target point cloud set from point cloud data with heights higher than a preset height threshold in the original point cloud set, and to sort the point cloud data in the first target point cloud set to obtain a second target point cloud set;

[0070] The coal pile identification module 13 is used to identify whether there is target point cloud data that meets the preset coal pile volume conditions in the second target point cloud set based on a preset clustering method, and determine the corresponding coal pile point according to the identification result.

[0071] In the present application, the tank cleaning machine first obtains the original point cloud data collected by the 3D laser radar, filters the original point cloud data, and determines the corresponding original point cloud set based on the filtered original point cloud data; then, the point cloud data in the original point cloud set whose height is higher than the preset height threshold is formed into a first target point cloud set, and the point cloud data in the first target point cloud set is sorted to obtain a second target point cloud set; finally, based on a preset clustering method, it is identified whether there is target point cloud data that meets the preset coal pile volume condition in the second target point cloud set, and the corresponding coal pile point is determined according to the identification result. As can be seen from the above, the present application collects original point cloud data through a 3D laser radar, determines the original point cloud set after filtering the original point cloud data, and selects the point cloud data that constitutes the first target point cloud set from the original point cloud set based on the preset height threshold, then re-sorts the point cloud data of the first target point cloud set to obtain the second target point cloud set, and finally identifies whether there is target point cloud data that meets the preset coal pile volume condition in the second target point cloud set to obtain a coal pile identification result. In this application, 3D laser radar is used as a sensor for coal pile identification. The 3D laser radar has a long effective detection distance and a detection accuracy within the centimeter level. In this way, this application automates the transformation of the cabin cleaning machine so that the cabin cleaning machine can operate autonomously in a closed cabin, follow the preset operating procedures, and gradually push the coal on the edge of the cabin to the center area to facilitate the subsequent grabbing work of the ship unloader. In this way, this application can accurately identify the specific outline of the obstacle without the need for human participation and without being affected by environmental factors, so that the location of the coal pile in the closed cabin can be accurately identified, and it can work continuously in the harsh environment of the closed cabin with high reliability.

[0072] In some specific embodiments, the coal pile identification device may further include:

[0073] An area determination module, used to control the tank cleaning machine to be in a stationary state, and determine a corresponding area to be identified according to the current position of the tank cleaning machine;

[0074] The map coordinate system construction module is used to scan the area to be identified by using the 3D laser radar to obtain a priori map corresponding to the area to be identified, and to construct a map coordinate system corresponding to the priori map.

[0075] In some specific embodiments, the coal pile identification device may further include:

[0076] A vehicle body point cloud data determination module, used to determine the vehicle body point cloud data corresponding to the tank cleaning machine from the original point cloud data according to the vehicle body parameter information of the tank cleaning machine;

[0077] Correspondingly, the point cloud data filtering module 11 includes:

[0078] The point cloud data filtering unit is used to filter the original point cloud data based on the vehicle body point cloud data to obtain the filtered original point cloud data.

[0079] In some specific implementations, the point cloud data filtering module 11 may include:

[0080] A posture determination unit, used for inputting the original point cloud data, the prior map, and the current motion state data of the tank cleaning machine into a preset 3D laser positioning program, and determining the current 6D global posture of the tank cleaning machine in the map coordinate system through the preset 3D laser positioning program;

[0081] The original point cloud set determining unit is used to convert the filtered original point cloud data into the map coordinate system based on each posture corresponding to the current 6D global posture of the tank cleaning machine, and determine the corresponding original point cloud set according to the converted filtered original point cloud data.

[0082] In some specific implementations, the point cloud data sorting module 12 may include:

[0083] a point cloud data sorting unit, configured to determine a straight-line distance between each point cloud data in the first target point cloud set and the tank cleaning machine, and sort each point cloud data in the first target point cloud set in an ascending order of the straight-line distance to obtain the sorted point cloud data;

[0084] The second target point cloud set determining unit is used to set a serial number corresponding to the sorted point cloud data, and determine a second target point cloud set based on each of the sorted point cloud data and the serial number corresponding to each of the sorted point cloud data.

[0085] In some specific embodiments, the coal pile identification module 13 may include:

[0086] a target point cloud data identification unit, configured to identify, based on the sequence numbers corresponding to the sorted point cloud data in the second target point cloud set and a preset clustering method, whether there is target point cloud data in the second target point cloud set that meets a preset coal pile volume condition;

[0087] The coal pile point determination unit is used to stop further identification and determine the corresponding coal pile point according to the target point cloud data if it is identified that the second target point cloud set contains target point cloud data that meets the preset coal pile volume condition.

[0088] In some specific embodiments, the coal pile identification device may further include:

[0089] The coal pile point pushing and raking module is used to determine whether the coal pile point meets the preset safe pushing and raking conditions if it is determined that no collision will occur when the coal pile point is pushed and raked by the cabin cleaning machine, and determine the position and height corresponding to the coal pile point, so that the cabin cleaning machine can push and rake the coal pile point based on the position and height corresponding to the coal pile point.

[0090] Furthermore, the present application also discloses an electronic device. Figure 5 This is a structural diagram of an electronic device 20 shown according to an exemplary embodiment, and the content in the figure cannot be regarded as any limitation on the scope of use of this application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 is used to store a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the coal pile identification method disclosed in any of the aforementioned embodiments. In addition, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0091] In this embodiment, the power supply 23 is used to provide working voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and the external device, and the communication protocol it follows is any communication protocol that can be applied to the technical solution of the present application, and is not specifically limited here; the input and output interface 25 is used to obtain external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs and is not specifically limited here.

[0092] In addition, the memory 22, as a carrier for storing resources, can be a read-only memory, a random access memory, a disk or an optical disk, etc. The resources stored thereon can include an operating system 221, a computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0093] The operating system 221 is used to manage and control the hardware devices and computer program 222 on the electronic device 20, and can be Windows Server, Netware, Unix, Linux, etc. In addition to the computer program that can be used to complete the coal pile identification method performed by the electronic device 20 disclosed in any of the aforementioned embodiments, the computer program 222 can further include computer programs that can be used to complete other specific tasks.

[0094] Furthermore, the present application also discloses a computer-readable storage medium for storing a computer program; wherein the computer program, when executed by a processor, implements the aforementioned disclosed coal pile identification method. The specific steps of the method can refer to the corresponding contents disclosed in the aforementioned embodiments, and will not be repeated here.

[0095] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0096] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0097] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0098] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0099] The technical solution provided by the present application is introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for general technicians in this field, according to the idea of ​​the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for identifying a coal pile, characterized in that: Applicable to tank cleaning machines, including: Acquire original point cloud data collected by the 3D laser radar, filter the original point cloud data, and determine a corresponding original point cloud set based on the filtered original point cloud data; The point cloud data in the original point cloud set whose height is higher than a preset height threshold form a first target point cloud set, and the point cloud data in the first target point cloud set are sorted to obtain a second target point cloud set; Based on a preset clustering method, it is identified whether there is target point cloud data that meets the preset coal pile volume condition in the second target point cloud set, and the corresponding coal pile point is determined according to the identification result.

2. The coal pile identification method according to claim 1, characterized in that: Before obtaining the original point cloud data collected by the 3D laser radar, the method further includes: Controlling the tank cleaning machine to be in a stationary state, and determining a corresponding area to be identified according to the current position of the tank cleaning machine; The area to be identified is scanned by the 3D laser radar to obtain a priori map corresponding to the area to be identified, and a map coordinate system corresponding to the priori map is constructed.

3. The coal pile identification method according to claim 2, characterized in that: After obtaining the original point cloud data collected by the 3D laser radar, the method further includes: Determining the body point cloud data corresponding to the tank cleaning machine from the original point cloud data according to the body parameter information of the tank cleaning machine; Accordingly, the filtering process of the original point cloud data includes: The original point cloud data is filtered based on the vehicle body point cloud data to obtain the filtered original point cloud data.

4. The coal pile identification method according to claim 3, characterized in that: The determining a corresponding original point cloud set based on the filtered original point cloud data comprises: Inputting the original point cloud data, the prior map, and the current motion state data of the tank cleaning machine into a preset 3D laser positioning program, and determining the current 6D global position of the tank cleaning machine in the map coordinate system through the preset 3D laser positioning program; The filtered original point cloud data is converted to the map coordinate system based on each posture corresponding to the current 6D global posture of the tank cleaning machine, and the corresponding original point cloud set is determined according to the converted filtered original point cloud data.

5. The coal pile identification method according to claim 1, characterized in that: The step of sorting the point cloud data in the first target point cloud set to obtain a second target point cloud set includes: Determine the straight-line distance between each point cloud data in the first target point cloud set and the tank cleaning machine, and sort the point cloud data in the first target point cloud set in ascending order based on the straight-line distance to obtain the sorted point cloud data; A sequence number corresponding to the sorted point cloud data is set, and a second target point cloud set is determined based on each of the sorted point cloud data and the sequence number corresponding to each of the sorted point cloud data.

6. The coal pile identification method according to claim 5, characterized in that: The step of identifying whether there is target point cloud data meeting a preset coal pile volume condition in the second target point cloud set based on a preset clustering method, and determining a corresponding coal pile point according to the identification result, includes: Based on the sequence numbers corresponding to the sorted point cloud data in the second target point cloud set and a preset clustering method, one by one, identifying whether there is target point cloud data in the second target point cloud set that meets the preset coal pile volume condition; If it is identified that the second target point cloud set contains target point cloud data that meets the preset coal pile volume condition, the identification is stopped and the corresponding coal pile point is determined according to the target point cloud data.

7. The coal pile identification method according to any one of claims 1 to 6, characterized in that: Also includes: If it is determined that no collision will occur when the coal pile point is pushed and raked by the cabin cleaning machine, it is determined that the coal pile point meets the preset safe pushing and raking conditions, and the position and height corresponding to the coal pile point are determined, so that the cabin cleaning machine can push and rake the coal pile point based on the position and height corresponding to the coal pile point.

8. A coal pile identification device, characterized in that: Applicable to tank cleaning machines, including: A point cloud data filtering module, used to obtain the original point cloud data collected by the 3D laser radar, filter the original point cloud data, and determine the corresponding original point cloud set based on the filtered original point cloud data; a point cloud data sorting module, configured to form a first target point cloud set from point cloud data in the original point cloud set whose height is higher than a preset height threshold, and to sort the point cloud data in the first target point cloud set to obtain a second target point cloud set; The coal pile identification module is used to identify whether there is target point cloud data that meets the preset coal pile volume conditions in the second target point cloud set based on a preset clustering method, and determine the corresponding coal pile point according to the identification result.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory; wherein the memory is used to store a computer program, and the computer program is loaded and executed by the processor to implement the coal pile identification method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store a computer program, which, when executed by a processor, implements the coal pile identification method as described in any one of claims 1 to 7.

Images