Anti-collision early warning method, device and equipment for grab ship unloader, storage medium and computer program

Abstract

The invention relates to the technical field of grab ship unloaders, in particular to an anti-collision early warning method, device and equipment for a grab ship unloader, a storage medium and a computer program.The grab ship unloader comprises a grab bucket and a cabin, and the method comprises the steps that cabin point cloud data is obtained; constructing a coordinate system according to the cabin point cloud data; determining an initial coordinate of the grab bucket according to the coordinate system; under the condition that the initial coordinate of the grab bucket is located in a target area, dividing the target area into a plurality of sub-areas according to a preset step length; obtaining point cloud data of the plurality of sub-regions; determining real-time coordinates of the grab bucket according to the point cloud data of two adjacent sub-regions in the plurality of sub-regions; calculating the real-time distance between the grab bucket and the cabin according to the real-time coordinates of the grab bucket; and under the condition that the real-time distance between the grab bucket and the cabin is smaller than or equal to a preset distance threshold value, early warning prompt operation is executed. The method can prevent the grab bucket from colliding with the cabin in the operation process.

Description

Technical Field

[0001] The present disclosure relates to the technical field of grab ship unloaders, and particularly to a method, device, equipment, storage medium and computer program for anti-collision warning of a grab ship unloader. Background Art

[0002] The safety guarantee system of a coal port is crucial for the normal operation of the port. It not only relates to the safety of the port, but also directly affects the operation efficiency and environmental protection of the port. Coal ports rely on various large-scale mechanical equipment for the loading, unloading, stacking and transfer of coal. These equipment are usually costly, and once damaged due to accidents, it will not only affect the normal operation of the port, but also cause huge economic losses.

[0003] The traditional solution mainly collects real-time data through radar equipment, and delimits a protection area under the lower part of the girder in the sea side direction or delimits a protection area in the left, right and lower side directions of the receiving plate with the installation position as a reference, so as to realize the real-time monitoring and protection of the ship unloader. This method can detect whether there will be a collision between the girder, the receiving plate and the cabin, but other positions, especially the grab mechanism, cannot be detected.

[0004] Since the shape of the grab is usually relatively complex, with multiple moving parts and irregular geometric shapes. This complexity makes it difficult to accurately identify and track the position and attitude of the grab through traditional radar equipment. At the same time, the grab needs to perform complex movements during the ship unloading operation, including up and down movement, rotation, grasping and releasing, etc. These complex movements make the position and attitude of the grab difficult to predict and monitor in real time. Therefore, how to detect the real-time distance between the grab and the cabin in real time has become an urgent problem to be solved. Summary of the Invention

[0005] The present disclosure provides a method, device, equipment, storage medium and computer program for anti-collision warning of a grab ship unloader, which can more accurately detect the grab position in real time and prevent the grab from colliding with the cabin during the operation.

[0006] In a first aspect, the present disclosure provides a method for anti-collision warning of a grab ship unloader, where the grab ship unloader includes a grab and a cabin, and the method includes:

[0007] Obtain the cabin point cloud data;

[0008] Construct a coordinate system according to the cabin point cloud data, where the origin of the coordinate system is the center point of the base of the grab ship unloader, and the directions of the coordinate system include the trolley traveling direction of the grab ship unloader and the winch traveling direction of the grab ship unloader;

[0009] Determine the initial coordinates of the grab according to the coordinate system;

[0010] When the initial coordinates of the grab are within the target area, divide the target area into multiple sub-areas according to a preset step size;

[0011] Obtain the point cloud data of multiple sub-areas;

[0012] Determine the real-time coordinates of the grab according to the point cloud data of two adjacent sub-areas among multiple sub-areas;

[0013] Calculate the real-time distance between the grab and the cabin according to the real-time coordinates of the grab;

[0014] When the real-time distance between the grab and the cabin is less than or equal to a preset distance threshold, perform a warning prompt operation.

[0015] In some embodiments, determining the real-time coordinates of the grab according to the point cloud data of two adjacent sub-areas among multiple sub-areas includes:

[0016] Determine the quantity difference between the point cloud data of two adjacent sub-areas among multiple sub-areas according to the point cloud data of multiple sub-areas;

[0017] Determine the real-time coordinates of the grab according to the quantity difference between the point cloud data of two adjacent sub-areas among multiple sub-areas.

[0018] In some embodiments, determining the quantity difference between the point cloud data of two adjacent sub-areas among multiple sub-areas according to the point cloud data of multiple sub-areas includes:

[0019] Perform point cloud filtering processing on the point cloud data of multiple sub-areas to obtain filtered point cloud data;

[0020] Perform key point extraction processing on the filtered point cloud data to obtain key point cloud data;

[0021] Perform point cloud registration processing on the key point cloud data to obtain target point cloud data;

[0022] Determine the quantity difference between the point cloud data of two adjacent sub-areas among multiple sub-areas according to the target point cloud data.

[0023] In some embodiments, determining the real-time coordinates of the grab according to the quantity difference between the point cloud data of two adjacent sub-areas among multiple sub-areas includes:

[0024] When the quantity difference between the point cloud data of two adjacent sub-areas among multiple sub-areas is greater than or equal to a preset quantity threshold, determine the target sub-area corresponding to the grab from the two adjacent sub-areas, and the quantity of the point cloud data in the target sub-area is greater than the quantity of the point cloud data in the non-target sub-area;

[0025] Determine the real-time coordinates of the grab according to the point cloud data in the target sub-area.

[0026] In some embodiments, calculating the real-time distance between the grab and the cabin according to the real-time coordinates of the grab includes:

[0027] Determining the boundary point coordinates corresponding to the cabin according to the cabin point cloud data;

[0028] Fitting the boundary line corresponding to the cabin according to the boundary point coordinates;

[0029] Determining the real-time distance between the grab and the cabin according to the distance between the real-time coordinates of the grab and the boundary line.

[0030] In some embodiments, calculating the real-time distance between the grab and the cabin according to the real-time coordinates of the grab includes:

[0031] Predicting the real-time coordinates of the grab according to the pre-trained grab trajectory prediction model to obtain the predicted real-time coordinates;

[0032] Calculating the real-time distance between the grab and the cabin according to the predicted real-time coordinates.

[0033] In some embodiments, obtaining the cabin point cloud data includes:

[0034] Scanning the area corresponding to the cabin by using at least two radars to obtain the scanned point cloud data, the horizontal field of view angle of the radar is 70.4°, and the vertical field of view angle of the radar is 77.2°;

[0035] Performing point cloud stitching processing on the scanned point cloud data to obtain the cabin point cloud data.

[0036] In a second aspect, the present disclosure provides an anti-collision warning device for a grab ship unloader, including:

[0037] An acquisition unit for acquiring the cabin point cloud data;

[0038] A processing unit for constructing a coordinate system according to the cabin point cloud data, the coordinate origin of the coordinate system is the center point of the base of the grab ship unloader, and the directions of the coordinate system include the traveling direction of the trolley of the grab ship unloader and the traveling direction of the crane of the grab ship unloader;

[0039] The processing unit is further configured to determine the initial coordinates of the grab according to the coordinate system;

[0040] A division unit for dividing the target area into a plurality of sub-areas at a preset step length when the initial coordinates of the grab are within the target area;

[0041] The acquisition unit is further configured to acquire the point cloud data of the plurality of sub-areas;

[0042] The processing unit is further configured to determine the real-time coordinates of the grab according to the point cloud data of two adjacent sub-areas among the plurality of sub-areas;

[0043] The processing unit is further configured to calculate the real-time distance between the grab and the cabin according to the real-time coordinates of the grab;

[0044] The warning unit is configured to perform a warning prompt operation when the real-time distance between the grab and the cabin is less than or equal to a preset distance threshold.

[0045] In a third aspect, the present disclosure provides an electronic device, including a memory, a processor, and a computer program stored on the memory, and the processor executes the computer program to implement the steps of the method described in the above aspect.

[0046] In a fourth aspect, the present disclosure provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described in the above aspect are implemented.

[0047] In a fifth aspect, the present disclosure provides a computer program product, including a computer program / instructions, and when the computer program is executed by a processor, the steps of the method described in the above aspect are implemented.

[0048] A grab ship unloader anti-collision warning method, device, equipment, storage medium and computer program provided by the present disclosure can accurately describe the shape and spatial position of the cabin by obtaining the cabin point cloud data, providing the latest data support for subsequent coordinate system construction and grab position determination; taking the center point of the base of the grab ship unloader as the coordinate origin, and the directions of the coordinate system include the traveling direction of the trolley and the traveling direction of the grab ship unloader, clarifying the movement direction of the grab, establishing a unified coordinate system, facilitating subsequent grab position calculation and collision risk assessment; determining the initial coordinates of the grab according to the coordinate system enables real-time monitoring from the start of grab operation, and timely discovery of potential collision risks; dividing the target area into multiple sub-areas can more finely monitor the movement of the grab, improving the accuracy of position detection; the preset step size can be adjusted according to actual needs to adapt to cabins of different sizes and shapes; by dividing sub-areas, complex calculation tasks can be decomposed into multiple smaller tasks, optimizing the use of computing resources; through the point cloud data of multiple sub-areas, the position of the grab can be determined more accurately, improving the accuracy of collision risk assessment; by comparing the point cloud data of adjacent sub-areas, the position of the grab can be determined in real time, ensuring real-time monitoring of the grab position. By calculating the real-time distance between the grab and the cabin, the change in the distance between the grab and the cabin can be monitored in real time. When the real-time distance between the grab and the cabin is less than or equal to the preset distance threshold, the system can timely send out a warning signal to remind the operator to pay attention, realizing real-time detection of the grab position and collision prevention, and ensuring the safety and efficiency of the ship unloader operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The present disclosure will be described in more detail below based on embodiments and with reference to the drawings:

[0050] Figure 1 It is a schematic flowchart of a collision prevention and warning method for a grab ship unloader provided by an embodiment of the present disclosure;

[0051] Figure 2 It is a process framework diagram of a collision prevention and warning method for a grab ship unloader provided by an embodiment of the present disclosure;

[0052] Figure 3 It is a schematic diagram of the installation method of a modeling radar provided by an embodiment of the present disclosure;

[0053] Figure 4 It is a top view of a ship's hatch provided by an embodiment of the present disclosure;

[0054] Figure 5 It is a schematic diagram of a coordinate system provided by an embodiment of the present disclosure;

[0055] Figure 6 It is a schematic structural diagram of a collision prevention and warning device for a grab ship unloader provided by an embodiment of the present disclosure;

[0056] Figure 7 It is a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure.

[0057] In the drawings, the same components are denoted by the same reference numerals, and the drawings are not drawn to actual scale. Detailed implementation manners

[0058] In order to enable those skilled in the art to better understand the technical solutions of the present disclosure, and to fully understand how the present disclosure uses technical means to solve technical problems and achieve the corresponding technical effects, and to implement accordingly, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The embodiments of the present disclosure and each feature in the embodiments can be combined with each other without conflict, and the formed technical solutions are all within the protection scope of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present disclosure.

[0059] It should be noted that the terms "first", "second", etc. in the description, claims and the above-mentioned drawings of the present disclosure are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described here can be implemented in an order different from those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0060] It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that here.

[0061] Example 1

[0062] Figure 1 is a schematic flowchart of a collision prevention and warning method for a grab ship unloader provided by an embodiment of the present disclosure. As Figure 1 shown, the method includes the following steps S101 to S108.

[0063] S101. Obtain the cabin point cloud data.

[0064] The point cloud data is a set of a large number of points obtained by a three-dimensional scanning device (such as a lidar device, etc.). Each point contains coordinate information (x, y, z) in space. These points are densely distributed on the surface of the object and can accurately describe the shape and spatial position of the object.

[0065] In some embodiments, obtaining the cabin point cloud data includes: scanning the area corresponding to the cabin with at least two radars to obtain scanned point cloud data, the horizontal field of view angle of the radar is 70.4°, and the vertical field of view angle of the radar is 77.2°; performing point cloud stitching processing on the scanned point cloud data to obtain the cabin point cloud data.

[0066] It can be understood that the radar emits laser pulses and receives the reflected signals, measures the time difference of the reflected signals to calculate the distance, and thus obtains the point cloud data.

[0067] Among them, the horizontal field of view angle refers to the scanning range of the radar in the horizontal direction, indicating the horizontal angle range that the radar can cover. In this embodiment, the horizontal field of view angle is 70.4°. The vertical field of view angle refers to the scanning range of the radar in the vertical direction, indicating the vertical angle range that the radar can cover. In this embodiment, the vertical field of view angle is 77.2°.

[0068] Point cloud stitching refers to aligning and fusing the point cloud data obtained by multiple radar scans to form a complete point cloud data set. Since the field of view angle of a single radar is limited, multiple radars usually need to scan from different angles, and then the point cloud data obtained from these scans is stitched together to obtain the complete point cloud data of the target area.

[0069] It can be understood that by using multiple radars to work together, the entire cabin can be covered, and the point cloud data within the cabin area can be obtained. Then, the point cloud data obtained by multiple radar scans is stitched together to form a complete cabin point cloud data set.

[0070] High-precision point cloud data is obtained through radar scanning, and each point contains accurate spatial coordinate information, providing a reliable data basis for subsequent coordinate system construction and grab position recognition.

[0071] S102. Construct a coordinate system based on the cabin point cloud data.

[0072] The coordinate origin of the coordinate system is the center point of the base of the grab ship unloader, and the directions of the coordinate system include the traveling direction of the trolley of the grab ship unloader and the traveling direction of the grab of the grab ship unloader.

[0073] The traveling direction of the trolley can be understood as the direction in which the entire grab ship unloader moves, usually the moving direction of the ship unloader.

[0074] The traveling direction of the grab can be understood as the moving direction of the boom of the ship unloader, or it can also be understood as the moving direction of the grab of the grab ship unloader relative to the entire ship unloader.

[0075] In this embodiment, for the constructed coordinate system, the center point of the base of the grab ship unloader is used as the origin, the traveling direction of the ship unloader trolley is used as the x direction, the direction of the boom (traveling direction of the grab) of the ship unloader is used as the y direction, and the upward direction (vertical direction) of the ship unloader is used as the z direction.

[0076] By establishing a unified coordinate system with the center point of the base of the grab ship unloader as the coordinate origin, the positional relationship between the grab and the cabin can be described more accurately, improving the positioning accuracy. At the same time, all point cloud data and subsequent calculations can be unified under the same reference system, facilitating subsequent grab position recognition and collision risk assessment.

[0077] S103. Determine the initial coordinates of the grab according to the coordinate system.

[0078] The initial coordinates of the grab can be understood as the starting position coordinates of the grab in the coordinate system. For example, the position when the grab starts working, or it can also be other positions set or specified by the user, which is not limited here.

[0079] By aligning the position of the grab with the coordinate system, the initial position of the grab in the coordinate system is determined, that is, the position data of the grab is converted to the coordinate system with the center of the ship unloader base as the origin.

[0080] As an example rather than a limitation, using GPS positioning technology and encoder positioning technology, the position data of the grab at the start of operation is obtained, and the position data of the grab is converted from the original coordinate system to the coordinate system with the center of the ship unloader base as the origin. For example, the position data of the grab is converted using a common coordinate transformation matrix algorithm.

[0081] By determining the initial coordinates of the grab according to the coordinate system, an accurate starting point can be provided for subsequent grab position tracking and collision risk assessment. This method not only improves the accuracy of the system, but also optimizes the operation process, ensuring the safety and efficiency of the ship unloader operation.

[0082] S104. When the initial coordinates of the grab are within the target area, divide the target area into multiple sub-areas according to a preset step size.

[0083] The target area refers to the area where the grab needs to operate during the ship unloader operation.

[0084] The preset step size refers to the preset division step size used to divide the target area into multiple sub-areas. As an example rather than a limitation, the preset step size can be set according to actual requirements and accuracy requirements. The preset step size can be fixed or dynamically adjusted according to the area size. For example, the preset step size can be 0.2m.

[0085] A sub-area refers to a smaller area obtained by dividing the target area according to the preset step size.

[0086] It can be understood that when the initial coordinates of the grab are within the target area, in order to more accurately monitor the position and movement of the grab, the target area needs to be divided into multiple sub-areas. The size and shape of each sub-area can be adjusted according to actual requirements. For example, the target area can be divided into multiple cubic or cuboid sub-areas, and the side length of each sub-area is the preset step size.

[0087] By dividing the target area into multiple sub - areas, the position of the grab can be determined more precisely and the movement of the grab can be monitored more carefully. After dividing the sub - areas, each sub - area can be monitored in real - time to promptly detect changes in the position of the grab, which helps to achieve real - time tracking of the grab and assessment of collision risks. At the same time, dividing the sub - areas can decompose complex computing tasks into multiple smaller tasks, facilitating parallel processing and optimizing the use of computing resources. By performing calculations within smaller sub - areas, the amount of calculation can be reduced and the response speed of the system can be increased.

[0088] S105. Obtain the point cloud data of multiple sub - areas.

[0089] Combined with the above, through radar scanning, obtain the point cloud data within each sub - area.

[0090] As an example rather than a limitation, starting from a preset height, sequentially obtain the point cloud data of multiple sub - areas.

[0091] By obtaining the high - precision point cloud data within each sub - area in real - time through radar scanning, real - time monitoring of the grab's position can be achieved, promptly obtaining the real - time position of the grab, and improving the judgment accuracy of the grab's position.

[0092] S106. Determine the real - time coordinates of the grab according to the point cloud data of two adjacent sub - areas among multiple sub - areas.

[0093] As an example rather than a limitation, by comparing the point cloud data of two adjacent sub - areas, the position change of the grab between these sub - areas can be determined. Further, the movement trajectory of the grab between these sub - areas can be determined, thereby calculating the real - time coordinates of the grab.

[0094] By comparing the point cloud data of adjacent sub - areas, the position change of the grab can be monitored in real - time to ensure the position accuracy of the grab during operation. By performing point cloud data processing between adjacent sub - areas, the amount of calculation can be reduced, the use of computing resources can be optimized, the coordinate information of the grab can be updated in real - time, the response speed of the system can be increased, and the timeliness of real - time monitoring of the grab's position and assessment of collision risks can be ensured. This method is applicable to ship cabins of different sizes and shapes and can adapt to complex operating environments and working conditions.

[0095] S107. Calculate the real - time distance between the grab and the ship cabin according to the real - time coordinates of the grab.

[0096] The real - time coordinates of the grab refer to the precise position of the grab at consecutive time points, usually represented by the (x, y, z) values in a coordinate system.

[0097] The real - time distance between the grab and the ship cabin refers to the distance between the grab and the ship cabin at a specific time point, and this distance is updated in real - time as the position of the grab changes.

[0098] By way of example and not limitation, after knowing the coordinates of the grab bucket and the cabin, the real-time distance between the grab bucket and the cabin can be calculated by using the common distance formula.

[0099] By calculating the real-time distance between the grab bucket and the cabin, it is possible to evaluate whether the grab bucket is approaching the cabin, thereby determining whether there is a collision risk. According to the change of the real-time distance, the movement trajectory or speed of the grab bucket can be dynamically adjusted to avoid collision.

[0100] S108. When the real-time distance between the grab bucket and the cabin is less than or equal to a preset distance threshold, perform a warning prompt operation.

[0101] The preset distance threshold can be understood as a safety threshold for determining the safe operating distance between the grab bucket and the cabin. When the distance between the grab bucket and the cabin is less than or equal to this value, it is considered that there is a collision risk and a warning prompt operation needs to be performed.

[0102] The warning prompt operation refers to a series of operations taken when the system detects that the real-time distance between the grab bucket and the cabin is less than or equal to the preset distance threshold to remind the operator to pay attention and take measures to avoid collision.

[0103] By way of example and not limitation, the warning prompt operation may include the following:

[0104] Emit an alarm signal. For example, emit an alarm signal through an audible and visual alarm, a display screen or other communication devices to remind the operator of the potential collision risk.

[0105] Adjust the position of the grab bucket. According to the specific situation, the movement trajectory or speed of the grab bucket can be automatically adjusted to avoid collision with the cabin, or the grab bucket can be instructed to stop moving until the risk is eliminated.

[0106] Through real-time monitoring and warning prompts, collisions between the grab bucket and the cabin can be detected and avoided in a timely manner, significantly improving the safety of operations, reducing the occurrence of such accidents. At the same time, the warning prompt operation can prompt the operator or the automated system to adjust the movement of the grab bucket in a timely manner, optimize the operation process, and improve the operation efficiency.

[0107] By obtaining the point cloud data of the cabin, the shape and spatial position of the cabin can be accurately described, providing the latest data support for subsequent coordinate system construction and grab position determination; taking the center point of the base of the grab ship unloader as the coordinate origin, the directions of the coordinate system include the traveling direction of the trolley and the traveling direction of the gantry of the grab ship unloader, clarifying the movement direction of the grab, establishing a unified coordinate system, facilitating subsequent grab position calculation and collision risk assessment; determining the initial coordinates of the grab according to the coordinate system enables real-time monitoring from the start of the grab operation, and timely detection of potential collision risks; dividing the target area into multiple sub-areas can more finely monitor the movement of the grab and improve the accuracy of position detection; the preset step size can be adjusted according to actual needs to adapt to cabins of different sizes and shapes; by dividing into sub-areas, complex calculation tasks can be decomposed into multiple smaller tasks, optimizing the use of computing resources; through the point cloud data of multiple sub-areas, the position of the grab can be determined more accurately, improving the accuracy of collision risk assessment; by comparing the point cloud data of adjacent sub-areas, the position of the grab can be determined in real time to ensure real-time monitoring of the grab position. By calculating the real-time distance between the grab and the cabin, the change in the distance between the grab and the cabin can be monitored in real time. When the real-time distance between the grab and the cabin is less than or equal to the preset distance threshold, the system can issue a warning signal in a timely manner to remind the operator to pay attention, realizing real-time detection of the grab position and collision prevention, and ensuring the safety and efficiency of the ship unloader operation.

[0108] Example 2

[0109] In some embodiments, determining the real-time coordinates of the grab according to the point cloud data of two adjacent sub-areas among multiple sub-areas includes: determining the quantity difference of the point cloud data of two adjacent sub-areas among multiple sub-areas according to the point cloud data of multiple sub-areas; determining the real-time coordinates of the grab according to the quantity difference of the point cloud data of two adjacent sub-areas among multiple sub-areas.

[0110] The quantity difference of the point cloud data of two adjacent sub-areas among multiple sub-areas refers to calculating the difference in the number of point cloud data points between adjacent sub-areas after dividing the target area into sub-areas. This difference can reflect the movement of the grab between these sub-areas.

[0111] By way of example and not limitation, count the point cloud data of each sub-area and calculate the number of point cloud data points in each sub-area. For example, there are 1000 points in sub-area A and 1200 points in sub-area B. Calculate the difference in the number of point cloud data points between adjacent sub-areas as: 1200 - 1000 = 200.

[0112] It can be understood that the point cloud data is continuous, and the movement of the grab will cause the distribution of the point cloud data between different sub-areas to change. By comparing the difference in the number of point cloud data in adjacent sub-areas, these changes can be captured, thereby inferring the movement direction and position of the grab. Specifically, if the number of point cloud data in a sub-area increases significantly, while the number of point cloud data in the adjacent sub-area decreases significantly, it can be inferred that the grab has moved from the sub-area with a decrease to the sub-area with an increase.

[0113] It can also be understood that the quantity difference is very sensitive to the movement of the grab. Even if the grab moves a small distance, the number of point cloud data in the adjacent sub-area will change. By calculating the quantity difference, these changes can be accurately captured to determine the real-time coordinates of the grab.

[0114] As an example and not a limitation, the real-time coordinates of the grab bucket are calculated based on the moving direction and quantity difference of the grab bucket, combined with the motion model and timestamp of the grab bucket. For example, if the grab bucket moves from sub-area A to sub-area B, the specific position of the grab bucket in sub-area B can be calculated using an interpolation method.

[0115] The calculation method of judging the real-time position of the grab bucket by the difference in quantity is simple and efficient, and can realize real-time monitoring and positioning to ensure the real-time update of the grab bucket position.

[0116] In some embodiments, determining the quantitative difference of point cloud data of two adjacent sub-areas among the multiple sub-areas based on the point cloud data of the multiple sub-areas includes: performing point cloud filtering processing on the point cloud data of the multiple sub-areas to obtain filtered point cloud data; performing key point extraction processing on the filtered point cloud data to obtain key point cloud data; performing point cloud registration processing on the key point cloud data to obtain target point cloud data; and determining the quantitative difference of point cloud data of two adjacent sub-areas among the multiple sub-areas based on the target point cloud data.

[0117] Point cloud filtering refers to filtering the point cloud data of each sub-area to remove noise points and isolated points and improve data quality. Commonly used filtering methods include at least the following:

[0118] Bilateral filtering smoothes point cloud data while preserving edge information.

[0119] Gaussian filtering: Smooth the point cloud data using a Gaussian kernel function.

[0120] Conditional filtering: Filter the point cloud data according to preset conditions (such as point density, curvature, etc.).

[0121] Pass-through filtering: By setting a threshold, points that exceed the threshold range are removed.

[0122] Random Sample Consensus (RANSAC): Remove outliers through random sampling and consistency checking.

[0123] VoxelGrid Filtering: Divide the point cloud data into voxel grids and retain one point in each voxel grid to reduce the amount of point cloud data.

[0124] Point cloud filtering can remove noise points and isolated points, making the point cloud data more accurate and reliable and reducing the computational load.

[0125] Key point extraction refers to extracting key points from the filtered point cloud data. Key points usually have high information content and can better represent the characteristics of the object. Commonly used key point extraction algorithms include interest point detection algorithms based on intensity and scale space, 3D key point extraction algorithms based on Harris corner detection, key point extraction algorithms based on normal homogeneity, 3D key point extraction algorithms based on Scale-Invariant Feature Transform, and so on.

[0126] By extracting representative key points from the point cloud data, the accuracy of feature matching can be improved, thereby improving the accuracy of point cloud registration and the efficiency of subsequent processing.

[0127] Point cloud registration refers to registering the extracted key point cloud data to align it with the target point cloud data. Commonly used point cloud registration algorithms include: ICP (Iterative Closest Point) algorithm, point cloud registration algorithm based on normal distribution transformation, feature-based registration method, and so on.

[0128] Point cloud registration can align the point cloud data of different sub-regions to the same coordinate system to ensure the consistency and accuracy of the data.

[0129] After the above point cloud data preprocessing steps, the target point cloud data can be obtained, and then the difference in the number of point cloud data between two adjacent sub-regions in multiple sub-regions is calculated using the target point cloud data.

[0130] Point cloud filtering removes noise points and isolated points, improves the quality of the point cloud data, and reduces the computational load of subsequent processing; key point extraction extracts representative key points from the point cloud data, reduces the amount of data, and improves the accuracy of feature matching; point cloud registration aligns the point cloud data obtained from different perspectives or at different times to the same coordinate system, ensures the consistency and accuracy of the data, improves the registration accuracy, and provides reliable data support for subsequent grab position recognition and collision risk assessment.

[0131] Example 3

[0132] In some embodiments, determining the real-time coordinates of the grab according to the difference in the number of point cloud data of two adjacent sub-regions among multiple sub-regions includes: when the difference in the number of point cloud data of two adjacent sub-regions among multiple sub-regions is greater than or equal to a preset number threshold, determining a target sub-region corresponding to the grab from the two adjacent sub-regions, where the number of point cloud data in the target sub-region is greater than the number of point cloud data in the non-target sub-region; determining the real-time coordinates of the grab according to the point cloud data in the target sub-region.

[0133] The preset number threshold is a preset threshold used to judge whether the difference in the number of point cloud data is significant. When the difference in quantity is greater than or equal to this threshold, it is considered that the grab has moved between adjacent sub-regions.

[0134] The target sub-region refers to the sub-region where the number of point cloud data increases when the difference in quantity is greater than or equal to the preset threshold. This indicates that the grab has moved to this sub-region, and it can also be understood as the sub-region where the grab exists.

[0135] Combined with the above, assume that there are 1000 points in sub-region A and 1200 points in sub-region B, and the difference in the number between the two is: 1200 - 1000 = 200. According to actual requirements and experience, set a preset number threshold to judge whether the difference in quantity is greater than or equal to the preset number threshold. If the difference in quantity is greater than or equal to the preset number threshold, then it is considered that sub-region B is the target sub-region, indicating that the grab has moved to sub-region B.

[0136] As an example but not a limitation, calculate the real-time coordinates of the grab according to the point cloud data in the target sub-region. This can be achieved through the following methods:

[0137] Geometric center method: Calculate the geometric center of the point cloud data in the target sub-region as the real-time coordinates of the grab.

[0138] Feature point method: Extract key points in the target sub-region and calculate the real-time coordinates of the grab through the coordinates of the feature points.

[0139] Interpolation method: Calculate the real-time coordinates of the grab according to the point cloud data in the target sub-region through interpolation.

[0140] By calculating the difference in the number of point cloud data, the movement of the grab between adjacent sub-regions can be accurately determined, thereby achieving high-precision real-time positioning and timely discovering potential collision risks.

[0141] Example 4

[0142] In some embodiments, calculating the real-time distance between the grab and the cabin according to the real-time coordinates of the grab includes: determining the boundary point coordinates corresponding to the cabin according to the cabin point cloud data; fitting the boundary line corresponding to the cabin according to the boundary point coordinates; and determining the real-time distance between the grab and the cabin according to the distance between the real-time coordinates of the grab and the boundary line.

[0143] The boundary point coordinates refer to the coordinates of the points on the boundary of the cabin in the cabin point cloud data.

[0144] The boundary line refers to the boundary line of the cabin obtained by fitting the boundary point coordinates, which is used to represent the shape and position of the cabin.

[0145] As an example but not a limitation, a point cloud processing algorithm (such as the RANSAC algorithm) is used to extract boundary points from the point cloud data. The RANSAC algorithm can effectively extract boundary points from the point cloud data through random sampling and consistency checking.

[0146] A fitting algorithm (such as the least squares method) is used to fit the extracted boundary point coordinates to obtain the boundary line of the cabin. For example, the least squares method can be used to fit the boundary line of the cabin to obtain the mathematical expression of the boundary line. The fitted boundary line is represented as a mathematical equation, such as a straight line equation, a curve equation, etc.

[0147] It can be understood that the number of boundary lines can be set according to the actual situation and is not limited here.

[0148] As an example but not a limitation, the boundary line of the cabin can be represented as a straight line equation y = mx + b, where m is the slope and b is the intercept.

[0149] Then, the distance formula from a point to a line is used to calculate the distance between the real-time coordinates of the grab and the cabin boundary line. As the grab moves, the coordinates of the grab are updated in real time, and the real-time distance between the grab and the cabin is recalculated.

[0150] By using the real-time coordinates of the grab and the exact position of the cabin boundary line, the distance between the grab and the cabin can be accurately calculated, which helps to improve the calculation accuracy. Compared with other methods, such as rough estimation based on sensors, the method using coordinates and boundary lines can significantly reduce errors and provide more reliable distance information.

[0151] In some embodiments, calculating the real-time distance between the grab and the cabin according to the real-time coordinates of the grab includes: predicting the real-time coordinates of the grab according to a pre-trained grab trajectory prediction model to obtain predicted real-time coordinates; and calculating the real-time distance between the grab and the cabin according to the predicted real-time coordinates.

[0152] The pre-trained grab trajectory prediction model refers to a model trained through historical data corresponding to the grab movement trajectory, which can predict the real-time coordinates of the grab in the future.

[0153] The predicted real-time coordinates refer to the coordinates of the future position of the grab calculated through a prediction model based on the historical movement trajectory and current state of the grab.

[0154] As an example rather than a limitation, collect the historical movement data of the grab, including information such as position coordinates, speed, acceleration, etc. Preprocess the data, such as removing noise, normalizing, etc., to improve the training effect of the model. A suitable prediction model can be selected according to the actual situation, such as Kalman filtering, LSTM (Long Short-Term Memory Network), etc. Use the historical data to train the model and optimize the model parameters to enable it to accurately predict the future position of the grab. According to the current state of the grab (such as the current position, speed, etc.), use the trained model for real-time prediction to obtain the predicted real-time coordinates.

[0155] By predicting the future position of the grab, the distance between the grab and the ship's hold can be calculated in advance, potential collision risks can be detected in a timely manner, early warnings can be issued, and more time can be spared to take protective measures, such as adjusting the movement trajectory or speed of the grab, to avoid the occurrence of collision accidents and improve the safety of operations.

[0156] Example 5

[0157] Based on the above embodiments, this embodiment provides an application example.

[0158] Figure 2 It is a flow framework diagram of a method for preventing collision and warning of a grab ship unloader provided by an embodiment of the present disclosure.

[0159] As Figure 2 shown, first, in the first step, the recognition of the current hatch edge is realized. Through 2 modeling radars (field of view angle 77.2°×70.4°) installed under the platform near the driver's cab, full coverage of the current hatch is achieved.

[0160] Figure 3 It is a schematic diagram of the installation method of a modeling radar provided by an embodiment of the present disclosure.

[0161] As Figure 3 shown, set the radar to continuous scanning mode, comprehensively scan the ship's hold area, collect point cloud data in real time, splice the point cloud data scanned by the 2 modeling radars to form complete point cloud data of the ship's hold. Use the coordinate transformation matrix algorithm to convert the data in the radar coordinate system to the coordinate system with the center of the unloader base as the coordinate origin for subsequent processing.

[0162] Figure 4 It is a top view of a ship's hatch provided by an embodiment of the present disclosure.

[0163] Combined with Figure 4, extract the plane where the hatch is located from the preprocessed point cloud data. Combining the geometric features of the hatch (such as rectangle, square), use the Ransac algorithm to perform linear fitting on the final boundary point cloud set in the track coordinates to obtain 4 contour lines Left1, Right2, Front3, and Back4, and calculate the coordinates of the four corner points of the hatch.

[0164] Figure 5 It is a schematic diagram of a coordinate system provided by an embodiment of the present disclosure.

[0165] As Figure 5 shown, establish a coordinate system with the center of the ship unloader base as the coordinate origin, the origin coordinates are (x0, y0, z0), the x - direction is the traveling direction of the ship unloader's trolley, the y - direction is the direction of the ship unloader's boom (trolley traveling), and the z - direction is the upward direction of the ship unloader. According to the GPS positioning technology and encoder positioning technology of the ship unloader, determine the x - axis coordinate and y - axis coordinate of the grab during the process of the grab moving towards the ship's hatch.

[0166] When the grab moves directly above the current working hatch, that is, when the x - coordinate value and y - coordinate value of the grab are both within the hatch range, start point cloud grab recognition. In the horizontal area of the hatch, search downward in sequence from a specified height with a step size of 0.2 m, and calculate the number of point clouds in each height range; when the difference in the number of point clouds between two adjacent height ranges is greater than the threshold T, that is, N i+1 -N i >T, it is considered that the transition from the rope to the grab range occurs, and it is determined that the radar recognizes the grab position; after searching for the grab position, take the median values of the point clouds in the X and Y directions as the center position (x m , y m , z m ) of the grab, thereby determining the upper center position of the grab.

[0167] The following introduces how to process point cloud data.

[0168] 1. Point cloud filtering: The originally collected point cloud data often contains a large number of scattered points and isolated points, and these noise points will affect the accuracy of recognition. Therefore, perform filtering processing through methods such as bilateral filtering, Gaussian filtering, conditional filtering, direct filtering, random sample consensus filtering, VoxelGrid filtering, etc. to reduce noise and improve data quality.

[0169] 2. Key point extraction: Extract key points from the point cloud data. These key points usually have higher information content and can better represent the characteristics of the object. Common 3D point cloud key point extraction algorithms include ISS3D, Harris3D, NARF, SIFT3D, etc.

[0170] 3. Feature Description: To accurately describe a 3D point cloud, in addition to position information, some additional parameters need to be calculated, such as normal direction, curvature, texture features, etc. Commonly used feature description algorithms include normal and curvature calculation, eigenvalue analysis, PFH (Point Feature Histogram descriptor), FPFH (Fast Point Feature Histogram descriptor), 3D Shape Context, Spin Image, etc.

[0171] 4. Feature Enhancement: Enhance the extracted features through methods such as deep learning to improve the robustness and distinctiveness of the features, thereby further improving the accuracy of recognition.

[0172] 5. Point Cloud Registration: Point cloud registration is the process of aligning point cloud data obtained from different viewpoints or at different times. Commonly used point cloud registration algorithms include Normal Distribution Transform (NDT) and Iterative Closest Point (ICP) algorithms, etc. Through precise registration, the errors caused by viewpoint changes or motion can be reduced, and the accuracy of recognition can be improved.

[0173] 6. Alignment Optimization: During the registration process, optimization algorithms (such as variants of the ICP algorithm, such as Robust ICP, point to plane ICP, point to line ICP, etc.) can be used to improve the accuracy and efficiency of alignment.

[0174] 7. Point Cloud Segmentation: Point cloud segmentation is the process of dividing point cloud data into different regions or objects. Common segmentation methods include region extraction, line and plane extraction, semantic segmentation, and clustering, etc. Through segmentation, the objects of interest can be separated from the background, reducing the interference of irrelevant information and improving the accuracy of recognition.

[0175] 8. Classification Optimization: During the classification process, methods such as supervised classification and unsupervised classification can be adopted, combined with machine learning or deep learning algorithms to improve the accuracy and efficiency of classification.

[0176] 9. Deep Learning Algorithms: Using deep learning algorithms (such as convolutional neural networks, recurrent neural networks, etc.) to process and analyze point cloud data can extract richer feature information and improve the accuracy of recognition.

[0177] 10. Model Optimization: By adjusting the parameters and structure of the deep learning model (such as increasing the number of network layers, changing the activation function, introducing regularization, etc.), the generalization ability and recognition accuracy of the model can be further improved.

[0178] 11. Post-processing: Based on the recognition results, post-processing operations (such as removing redundant information, smoothing, filling holes, etc.) can be carried out to further improve the accuracy and readability of the recognition results.

[0179] 12. Visualization: Use visualization tools (such as the Plotter module and Viewer module in PCL) to visually display point cloud data and recognition results, which helps to intuitively analyze the recognition effect, discover potential problems and optimize them.

[0180] In summary, by optimizing methods and technologies in aspects such as data preprocessing, feature description and enhancement, point cloud registration and alignment, point cloud segmentation and classification, algorithm and model optimization, and post - processing and visualization, the accuracy of point cloud recognition can be significantly improved.

[0181] Calculate the straight - line equations of the four line segments of the hatch edge according to the coordinate values of the four corners. Take one hatch edge (line segment AB) as an example. The coordinates of point A are (x1, y1, z1), and the coordinates of point B are (x2, y2, z2). Subtract the coordinates of the two points to get the direction vector s=(m, n, p) of this straight line. Using the symmetric form of the straight - line equation, that is, each coordinate of the direction vector as the corresponding denominator, and the unknown minus the corresponding known number as the numerator, the space straight - line equation of this hatch edge can be obtained. According to the point - to - line distance formula, calculate the distance from the grab to the hatch edge AB. Use this method to calculate the real - time distances D1, D2, D3, and D4 from the grab to the four edges of the hatch in turn.

[0182] The following introduces how to calculate the real - time distances from the grab to the four edges of the hatch.

[0183] (1) Optimize the distance calculation algorithm

[0184] Use the point - to - line distance formula as the basis, but optimize it to improve calculation efficiency and accuracy. Numerical methods can be used to reduce rounding errors in the calculation process, or more efficient matrix operation methods can be adopted. Parallel computing technology can also be considered to further improve the calculation speed.

[0185] (2) Dynamically adjust parameters

[0186] Dynamically adjust the calculation parameters according to the motion state of the grab and the shape of the hatch edge.

[0187] When the grab is approaching the hatch edge, the sampling frequency can be increased to improve accuracy; when the grab is moving away from the hatch edge, the sampling frequency can be reduced to improve the calculation speed.

[0188] (3) Machine - learning - based prediction algorithm

[0189] Use machine - learning algorithms to predict the motion trajectory of the grab.

[0190] By training the model, it is possible to predict the future position of the grab, so as to calculate and adjust the distance between the grab and the hatch edge in advance. This helps to reduce the collision risk caused by calculation delay.

[0191] When calculating the real-time distance from the grab to the hatch edge using the point-to-line distance formula, improvements are made through methods such as optimizing the distance calculation algorithm, dynamically adjusting parameters, and machine learning-based prediction algorithms. These improved algorithms help to improve the accuracy and efficiency of the calculation, reduce the collision risk, and thus improve the safety and efficiency of the operation.

[0192] Set a collision prevention range L. When any one of D1, D2, D3, D4 is less than the set collision prevention range L, it is considered that there is a collision risk for the grab of the ship unloader, and an alarm is given in time to take corresponding anti-collision measures.

[0193] It can be understood that the methods in the traditional scheme can detect whether there will be a collision between the girder, the receiving plate and the ship's hold, but other positions, especially the grab mechanism, cannot be detected. According to the data such as the traveling of the large machine, the position of the fixed support platform, and the position of the grab trolley, the present disclosure uses a three-dimensional point cloud recognition algorithm to identify and track the grab features. During the up and down movement of the grab in the hatch area, the program calculates the real-time position coordinates of the grab, and then combines with the three-dimensional model of the ship's hold to calculate and judge whether there is a collision risk between the grab and the ship's hold, which can prevent the collision between the grab of the important mechanism on the sea side of the ship unloader and the ship's hold, so as to achieve the purpose of active avoidance.

[0194] This method can calculate and track the accurate position coordinates of the grab in real time according to the three-dimensional point cloud recognition algorithm. This high-precision positioning ability can accurately grasp the dynamic changes of the grab in the hatch area, providing a solid foundation for subsequent collision risk assessment. The anti-collision system using the method of calculating the distance from a point to a line can handle ship hatches of various specifications. By adjusting the algorithm parameters and model settings, it can adapt to ship's holds of different sizes and shapes, with high flexibility and adaptability, and has a wide application prospect, and can meet the operation requirements of different ports and terminals.

[0195] Example 6

[0196] The anti-collision warning device of the grab ship unloader according to the embodiment of the present application will be introduced below with reference to the accompanying drawings. For the sake of brevity, when introducing the device below, appropriate omissions will be made, and the relevant content can refer to the relevant descriptions in the above method and will not be repeated.

[0197] Figure 6 It is a schematic structural diagram of an anti-collision warning device of a grab ship unloader provided by an embodiment of the present disclosure.

[0198] As Figure 6 shown, the device 1000 includes the following units.

[0199] An acquisition unit 1001, configured to acquire cabin point cloud data;

[0200] A processing unit 1002, configured to construct a coordinate system based on the cabin point cloud data, where the origin of the coordinate system is the center point of the base of the grab unloader, and the directions of the coordinate system include the traveling direction of the trolley of the grab unloader and the traveling direction of the gantry of the grab unloader;

[0201] The processing unit 1002 is further configured to determine the initial coordinates of the grab according to the coordinate system;

[0202] A division unit 1003, configured to divide the target area into multiple sub-areas according to a preset step length when the initial coordinates of the grab are within the target area;

[0203] The acquisition unit 1001 is further configured to acquire the point cloud data of multiple sub-areas;

[0204] The processing unit 1002 is further configured to determine the real-time coordinates of the grab according to the point cloud data of two adjacent sub-areas among the multiple sub-areas;

[0205] The processing unit 1002 is further configured to calculate the real-time distance between the grab and the cabin according to the real-time coordinates of the grab;

[0206] An early warning unit 1004, configured to perform an early warning prompt operation when the real-time distance between the grab and the cabin is less than or equal to a preset distance threshold.

[0207] In some embodiments, the processing unit 1002 is further configured to determine the real-time coordinates of the grab according to the point cloud data of two adjacent sub-areas among the multiple sub-areas, including:

[0208] Determining the quantity difference between the point cloud data of two adjacent sub-areas among the multiple sub-areas according to the point cloud data of the multiple sub-areas;

[0209] Determining the real-time coordinates of the grab according to the quantity difference between the point cloud data of two adjacent sub-areas among the multiple sub-areas.

[0210] In some embodiments, the processing unit 1002 is further configured to determine the quantity difference between the point cloud data of two adjacent sub-areas among the multiple sub-areas according to the point cloud data of the multiple sub-areas, including:

[0211] Performing point cloud filtering processing on the point cloud data of the multiple sub-areas to obtain filtered point cloud data;

[0212] Performing key point extraction processing on the filtered point cloud data to obtain key point cloud data;

[0213] Performing point cloud registration processing on the key point cloud data to obtain target point cloud data;

[0214] Determine the difference in the number of point cloud data between two adjacent sub-regions among multiple sub-regions according to the target point cloud data.

[0215] In some embodiments, the processing unit 1002 is further configured to determine the real-time coordinates of the grab according to the difference in the number of point cloud data between two adjacent sub-regions among multiple sub-regions, including:

[0216] When the difference in the number of point cloud data between two adjacent sub-regions among multiple sub-regions is greater than or equal to a preset number threshold, determine the target sub-region corresponding to the grab from the two adjacent sub-regions, and the number of point cloud data in the target sub-region is greater than the number of point cloud data in the non-target sub-region;

[0217] Determine the real-time coordinates of the grab according to the point cloud data in the target sub-region.

[0218] In some embodiments, the processing unit 1002 is further configured to calculate the real-time distance between the grab and the cabin according to the real-time coordinates of the grab, including:

[0219] Determine the boundary point coordinates corresponding to the cabin according to the cabin point cloud data;

[0220] Fit the boundary line corresponding to the cabin according to the boundary point coordinates;

[0221] Determine the real-time distance between the grab and the cabin according to the distance between the real-time coordinates of the grab and the boundary line.

[0222] In some embodiments, the processing unit 1002 is further configured to calculate the real-time distance between the grab and the cabin according to the real-time coordinates of the grab, including:

[0223] Predict the real-time coordinates of the grab according to the pre-trained grab trajectory prediction model to obtain the predicted real-time coordinates;

[0224] Calculate the real-time distance between the grab and the cabin according to the predicted real-time coordinates.

[0225] In some embodiments, the acquisition unit is further configured to acquire the cabin point cloud data, including:

[0226] Scan the area corresponding to the cabin using at least two radars to obtain the scanned point cloud data, the horizontal field of view angle of the radar is 70.4°, and the vertical field of view angle of the radar is 77.2°;

[0227] Perform point cloud stitching processing on the scanned point cloud data to obtain the cabin point cloud data.

[0228] It should be noted that for the information interaction, execution process, etc. between the above units, since they are based on the same concept as the method embodiment of the present application, their specific functions and the technical effects brought, please refer to the method embodiment part for details, and will not be elaborated here.

[0229] Example 7

[0230] Based on the above embodiments, this embodiment provides a computer device 3000, including a memory 3200, a processor 3100, and a computer program 3210 stored in the memory. The processor 3100 executes the computer program 3210 to implement the steps of the method described in the above embodiments.

[0231] In some embodiments of this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. The computer program, when executed by a processor, implements the steps of the method described in the above embodiments.

[0232] In some embodiments of this embodiment, a computer program product is provided, including a computer program / instructions. The computer program, when executed by a processor, implements the steps of the method described in the above embodiments.

[0233] The processor 3100 may include, but is not limited to, for example, one or more processors or microprocessors, etc. Each processor may be implemented by an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a controller, a microcontroller, a microprocessor, or other electronic components, and is used to execute the method in the above embodiments.

[0234] The computer-readable storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof. The computer-readable storage medium may include, but is not limited to, for example, a random access memory (RAM), a read-only memory (ROM), a flash memory, an EPROM memory, an EEPROM memory, a register, a computer storage medium (such as a hard disk, a floppy disk, a solid state drive, a removable disk, a CD-ROM, a DVD-ROM, a Blu-ray disc, etc.).

[0235] The computer-readable storage medium may also store at least one computer-executable program / instructions, which are, for example, computer-readable instructions. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory, etc. The computer-readable storage medium may include, for example, read-only memory (ROM), hard disk, flash memory, etc. For example, the non-transitory computer-readable storage medium may be connected to a computing device such as a computer. Then, when the computing device runs the computer-readable instructions stored on the computer-readable storage medium, the various methods described above may be performed.

[0236] In addition, the computer device 3000 may further include (but is not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (such as a keyboard, a mouse, a speaker, etc.).

[0237] The processor 3100 may communicate with external devices via the I / O bus through a wired or wireless network.

[0238] In one embodiment, the at least one computer-executable instruction may also be compiled into or constitute a software product / computer program product, and when one or more computer-executable instructions are run by a processor, the steps of the various functions and / or methods in the embodiments described in this technology are performed.

[0239] In the embodiments provided in this disclosure, it should be understood that the disclosed devices and methods may also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the drawings show the possible architectures, functions, and operations of devices, methods, and computer program products according to multiple embodiments of this disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above-mentioned module, program segment, or part of code includes one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system for performing the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.

[0240] It should be noted that in this disclosure, the terms "including", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element limited by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the element.

[0241] Although the embodiments disclosed in this disclosure are as above, the above content is only an embodiment adopted for the convenience of understanding this disclosure and is not intended to limit this disclosure. Any person skilled in the art within the technical field to which this disclosure pertains may make any modifications and changes in the form of implementation and details without departing from the spirit and scope disclosed in this disclosure. However, the scope of patent protection of this disclosure shall still be subject to the scope defined by the appended claims.

Claims

1. A grab ship unloader anti-collision warning method, characterized in that: The grab ship unloader comprises a grab and a cabin, and the method comprises: Obtain cabin point cloud data; A coordinate system is constructed according to the cabin point cloud data, wherein the origin of the coordinate system is the center point of the base of the grab ship unloader, and the direction of the coordinate system includes the walking direction of the trolley of the grab ship unloader and the walking direction of the trolley of the grab ship unloader; Determining the initial coordinates of the grab bucket according to the coordinate system; When the initial coordinates of the grab bucket are within the target area, the target area is divided into a plurality of sub-areas according to a preset step length; Acquire point cloud data of the multiple sub-areas; Determine the real-time coordinates of the grab bucket according to the point cloud data of two adjacent sub-areas among the multiple sub-areas; Calculating the real-time distance between the grab bucket and the cabin according to the real-time coordinates of the grab bucket; When the real-time distance between the grab bucket and the cabin is less than or equal to a preset distance threshold, an early warning prompt operation is performed.

2. The method according to claim 1, characterized in that The method of determining the real-time coordinates of the grab bucket according to the point cloud data of two adjacent sub-areas among the multiple sub-areas includes: Determine, according to the point cloud data of the multiple sub-areas, a quantity difference of point cloud data of two adjacent sub-areas in the multiple sub-areas; The real-time coordinates of the grab bucket are determined according to a quantity difference of point cloud data of two adjacent sub-areas among the multiple sub-areas.

3. The method according to claim 2, characterized in that The step of determining the quantity difference of point cloud data of two adjacent sub-areas among the multiple sub-areas according to the point cloud data of the multiple sub-areas comprises: Performing point cloud filtering processing on the point cloud data of the multiple sub-areas to obtain filtered point cloud data; Performing key point extraction processing on the filtered point cloud data to obtain key point cloud data; Performing point cloud registration processing on the key point cloud data to obtain target point cloud data; Determine the quantity difference of the point cloud data of two adjacent sub-areas among the multiple sub-areas according to the target point cloud data.

4. The method according to any one of claims 1 to 3, characterized in that The step of determining the real-time coordinates of the grab bucket according to the quantity difference of the point cloud data of two adjacent sub-areas among the multiple sub-areas comprises: When the difference in the number of point cloud data of two adjacent sub-areas among the multiple sub-areas is greater than or equal to a preset number threshold, a target sub-area corresponding to the grab bucket is determined from the two adjacent sub-areas, and the number of point cloud data in the target sub-area is greater than the number of point cloud data in the non-target sub-area; The real-time coordinates of the grab bucket are determined according to the point cloud data in the target sub-area.

5. The method according to claim 4, characterized in that The calculating the real-time distance between the grab bucket and the cabin according to the real-time coordinates of the grab bucket comprises: Determine the coordinates of the boundary points corresponding to the cabin according to the cabin point cloud data; Fitting the boundary line corresponding to the cabin according to the coordinates of the boundary points; The real-time distance between the grab bucket and the cabin is determined according to the distance between the real-time coordinates of the grab bucket and the boundary line.

6. The method according to claim 1, characterized in that The calculating the real-time distance between the grab bucket and the cabin according to the real-time coordinates of the grab bucket comprises: Predict the real-time coordinates of the grab bucket according to the pre-trained grab bucket trajectory prediction model to obtain the predicted real-time coordinates; The real-time distance between the grab bucket and the cabin is calculated according to the predicted real-time coordinates.

7. The method according to claim 1, characterized in that The acquisition of cabin point cloud data comprises: Scanning an area corresponding to the cabin using at least two radars to obtain scanning point cloud data, wherein the horizontal field of view angle of the radar is 70.4°, and the vertical field of view angle of the radar is 77.2°; The scanned point cloud data is subjected to point cloud stitching processing to obtain the cabin point cloud data.

8. An electronic device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images