Ship loader operation control optimization method, device, equipment and medium

Abstract

The application relates to a ship loader operation control optimization method, device, equipment and medium, and relates to the field of ship loading control. The method comprises the following steps: determining a ship cabin three-dimensional center coordinate corresponding to a target ship cabin according to laser radar point cloud data corresponding to the target ship cabin; when a front-rear inclination angle or a left-right inclination angle is greater than a preset angle threshold value, or a vertical distance is less than a preset safety distance, a preset material falling point correction model is used to correct a first material falling reference point according to the front-rear inclination angle, the left-right inclination angle and a static offset correction amount of the first material falling reference point, so as to update the first material falling reference point and determine a second material falling reference point; a walking mechanism of a large car in the ship loader is driven to move horizontally along a wharf ground and / or the rotation angle of a cantilever in the ship loader is adjusted until a chute in the ship loader is aligned with the second material falling reference point, so that the target bulk cargo is continuously loaded to the target ship; and the above steps are repeated until the target ship is completely loaded. The application can maintain stable and efficient ship loading operation under complex operation conditions.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of ship loading control, and in particular to a ship loader operation control optimization method, a corresponding device, an electronic device and a computer readable storage medium. BACKGROUND

[0002] In the field of port bulk cargo (such as coal, ore, etc.) loading, compared with the high automation level achieved in container operation, bulk cargo loading still faces challenges such as poor working environment (high dust, strong interference), irregular cargo shape, etc., resulting in generally low automation level, significant limitations of traditional manual operation mode and existing technical solutions, and difficulty in meeting the needs of modern ports for safe, efficient and unmanned operation. The main technical defects include:

[0003] First, the manual operation mode is low in efficiency and high in risk. At present, most bulk cargo loading relies on pure manual operation. The ship loader driver needs to observe the stockpile shape through the monitoring screen and receive hand signals from the ship personnel to manually adjust the walking of the trolley, the slewing of the boom and the position of the chute to control the material dropping point.

[0004] Second, the manual judgment and operation response are lagging, and the operation posture needs to be frequently adjusted, resulting in prolonged loading period and difficulty in adapting to the operation demand of large throughput of the port.

[0005] Third, the balance state of the ship is determined by the experience of the driver, which is prone to cause uneven material dropping due to human error, and may cause risks such as ship imbalance, inclination and even overturning.

[0006] Fourth, the poor environment such as dust and strong light at the wharf easily affects the visual judgment of the manual operation, further amplifies the operation error, and increases the risk of equipment collision (chute and stockpile collision) or cargo falling into the sea.

[0007] Fifth, during the bulk cargo loading process, the ship is easily affected by factors such as water flow fluctuation, tide change and stockpile center of gravity shift, and is prone to dynamic posture changes such as inclination and ship drift.

[0008] Sixth, due to the inherent structural distribution of the ship, there is a natural center of gravity shift characteristic, which forms a static deviation of the material dropping point.

[0009] In view of the above, the applicant has made corresponding exploration to solve the problems of low efficiency and high risk of the manual operation mode in the prior art, and the dynamic posture changes such as inclination and ship drift caused by factors such as water flow fluctuation, tide change and stockpile center of gravity shift. SUMMARY

[0010] The purpose of the present application is to solve the above problems and provide a ship loader operation control optimization method, a corresponding device, an electronic device and a computer readable storage medium.

[0011] To meet various purposes of the present application, the present application adopts the following technical solutions:

[0012] A ship loader operation control optimization method is proposed to adapt to one of the purposes of the present application, comprising:

[0013] Step S10, collect laser radar point cloud data corresponding to a target ship cabin of a target ship, and determine a ship cabin three-dimensional center coordinate corresponding to the target ship cabin according to the laser radar point cloud data, wherein a space coordinate system in which the ship cabin three-dimensional center coordinate is located is constructed with a direction along a wharf shoreline to a bow direction as a horizontal axis direction, a direction along a vertical wharf shoreline to a sea side direction as a vertical axis direction, and a direction perpendicular to a wharf ground upward as a vertical axis direction;

[0014] Step S20, take the ship cabin three-dimensional center coordinate as a first material falling reference point of a chute in a ship loader, load target bulk cargo to the target ship based on the first material falling reference point to determine a material pile of the target bulk cargo, and continuously monitor a front and rear inclination angle, a left and right inclination angle of the target ship corresponding to the target ship, and a vertical distance between the chute and the material pile;

[0015] Step S30, when the front and rear inclination angle or the left and right inclination angle is greater than a preset angle threshold, or the vertical distance is less than a preset safety distance, update the first material falling reference point according to a preset material falling point correction model, a static offset correction amount of the first material falling reference point, and the front and rear inclination angle and the left and right inclination angle to determine a second material falling reference point, wherein the static offset correction amount is determined by a cabin position of the target ship;

[0016] Step S40, drive a cart traveling mechanism in the ship loader to move horizontally along a wharf ground and / or adjust a slewing angle of a cantilever in the ship loader until the chute in the ship loader is aligned with the second material falling reference point, so as to continue loading target bulk cargo to the target ship;

[0017] Step S50, repeat the above steps S10 to S40 until the target ship is completed to be loaded, so as to complete optimization of ship loader operation control.

[0018] Optionally, the step of updating the first material falling reference point according to the preset material falling point correction model, the front and rear inclination angle, the left and right inclination angle, and the static offset correction amount of the first material falling reference point to determine the second material falling reference point, comprises:

[0019] obtaining a horizontal axis coordinate of the first blanking reference point, a vertical axis coordinate of the first blanking reference point, a horizontal axis static offset correction of the first blanking reference point, a ship depth of the target ship, a fore-aft inclination angle of the target ship, a mass of the target bulk cargo, a total mass of the target ship, and an average vertical axis coordinate, wherein the ship depth represents a vertical distance between a bottom of a hold of the target ship and a deck of the target ship, and the average vertical axis coordinate represents an average value among the vertical axis coordinate of the first blanking reference point, a vertical axis coordinate of a center of a machinery room of the target ship, and a vertical axis coordinate of a center of a stockpile;

[0020] calculating a first product between a half of the ship depth and a sine value of the fore-aft inclination angle, calculating a first difference between the vertical axis coordinate of the first blanking reference point and the average vertical axis coordinate, calculating a first ratio between the mass of the target bulk cargo and the total mass of the target ship, and calculating a second product among the first ratio, the first difference, and the sine value of the fore-aft inclination angle;

[0021] determining the horizontal axis coordinate of the second blanking reference point according to a first sum among the horizontal axis coordinate of the first blanking reference point, the horizontal axis static offset correction of the first blanking reference point, the first product, and the second product.

[0022] Optionally, the step of updating the first blanking reference point to determine the second blanking reference point according to the fore-aft inclination angle, the left-right inclination angle, and the horizontal axis static offset correction of the first blanking reference point by using a preset blanking point correction model, comprises:

[0023] obtaining a longitudinal axis coordinate of the first blanking reference point, a longitudinal axis static offset correction of the first blanking reference point, a left-right inclination angle of the target ship, and a longitudinal axis dynamic offset of the first blanking reference point;

[0024] calculating a third product between a half of the ship depth and a sine value of the left-right inclination angle, calculating a second ratio between the vertical axis coordinate of the first blanking reference point and the average vertical axis coordinate, and calculating a fourth product among the first ratio, the longitudinal axis dynamic offset of the first blanking reference point, the second ratio, and the sine value of the left-right inclination angle;

[0025] determining the longitudinal axis coordinate of the second blanking reference point according to a second sum among the longitudinal axis coordinate of the first blanking reference point, the longitudinal axis static offset correction of the first blanking reference point, the third product, and the fourth product.

[0026] Optionally, the step of updating the first blanking datum point to determine a second blanking datum point according to the front and rear inclination angle, the left and right inclination angle, and the static offset correction amount of the first blanking datum point comprises:

[0027] Obtaining the draft of the target ship;

[0028] Calculating a second difference value between the draft of the target ship and half of the ship depth, and calculating a fifth product between the second difference value, the cosine value of the front and rear inclination angle, and the cosine value of the left and right inclination angle;

[0029] Determining the vertical axis coordinate of the second blanking datum point according to a third sum value between the vertical axis coordinate of the first blanking datum point and the fifth product;

[0030] Constructing the second blanking datum point according to the horizontal axis coordinate, the longitudinal axis coordinate, and the vertical axis coordinate of the second blanking datum point.

[0031] Optionally, the step of determining the first blanking datum point comprises:

[0032] Based on the laser scanning device installed on the cantilever of the ship loader, laser radar point cloud data corresponding to the target ship cabin of the target ship is collected;

[0033] The laser radar point cloud data is down-sampled by using a voxel center down-sampling method, and based on the down-sampled laser radar point cloud data, a preset statistical filtering algorithm is used to calculate the average distance between each point and a plurality of adjacent points, calculate the average value and the standard deviation of all average distances, calculate a distance threshold value according to the average value and the standard deviation, and detect whether the average distance is greater than the distance threshold value, and if so, the point is removed as an outlier.

[0034] The laser radar point cloud data after statistical filtering processing is fitted by using a RANSAC algorithm to extract cabin wall surface point cloud data corresponding to the target ship cabin, and the cabin length, the cabin width, the ship depth, and the vertical axis coordinate of the cabin bottom in the vertical axis direction of the target ship cabin are determined according to the cabin wall surface point cloud data.

[0035] A fourth sum value between the starting coordinate of the target ship cabin in the horizontal axis direction and half of the cabin length is used to determine the horizontal axis coordinate of the cabin three-dimensional center coordinate, a fifth sum value between the starting coordinate of the target ship cabin in the longitudinal axis direction and half of the cabin width is used to determine the longitudinal axis coordinate of the cabin three-dimensional center coordinate, and a sixth sum value between the vertical axis coordinate of the cabin bottom in the vertical axis direction and half of the ship depth is used to determine the vertical axis coordinate of the cabin three-dimensional center coordinate.

[0036] According to the horizontal axis coordinate, the longitudinal axis coordinate and the vertical axis coordinate in the three-dimensional center coordinates of the ship cabin, the first blanking reference point is determined.

[0037] Optionally, the step of driving the walking mechanism of the ship loader to move horizontally along the wharf ground and / or adjusting the rotation angle of the cantilever of the ship loader until the chute of the ship loader is aligned with the second blanking reference point comprises:

[0038] An RGB image of a moving area containing a target obstacle in the horizontal movement of the walking mechanism along the wharf ground and a corresponding moving area depth image are obtained;

[0039] A trained obstacle detection model is called to detect the target obstacle in the moving area RGB image to determine the two-dimensional center pixel coordinates corresponding to the target obstacle, wherein the basic network architecture of the obstacle detection model is a YOLOv7-tiny model, and the target obstacle includes workers or scattered goods;

[0040] The moving area RGB image and the moving area depth image are aligned, and the depth value in the moving area depth image corresponding to the two-dimensional center pixel coordinates of the target obstacle in the moving area RGB image is matched to determine the three-dimensional center coordinates of the target obstacle in the real space;

[0041] If the Euclidean distance between the three-dimensional center coordinates of the target obstacle and the three-dimensional center coordinates of the walking mechanism is less than a shutdown distance threshold, a shutdown instruction is sent to the ship loader to force the walking mechanism to stop moving; if the Euclidean distance between the three-dimensional center coordinates of the target obstacle and the three-dimensional center coordinates of the walking mechanism is between the shutdown distance threshold and a warning distance threshold, an audible and visual alarm is sounded to remind workers to avoid.

[0042] Optionally, the front and rear inclination angle represents the included angle between the horizontal axis of the ship body pointing to the bow direction along the wharf line and the horizontal reference plane parallel to the wharf ground when the target ship rotates around the longitudinal axis pointing to the sea side direction along the vertical wharf line;

[0043] The left and right inclination angle represents the included angle between the vertical axis of the ship body pointing to the sea side direction along the wharf line and the horizontal reference plane parallel to the wharf ground when the target ship rotates around the horizontal axis pointing to the bow direction along the wharf line;

[0044] The target bulk cargo includes coal, ore or grain.

[0045] Another object of the present application is to provide a ship loader operation control optimization device, which comprises:

[0046] a cabin center determining module configured to collect laser radar point cloud data corresponding to a target cabin of a target ship, and determine a cabin three-dimensional center coordinate corresponding to the target cabin according to the laser radar point cloud data, wherein a spatial coordinate system in which the cabin three-dimensional center coordinate is located is constructed with a direction along a wharf shoreline pointing to a bow direction as a horizontal axis direction, a direction along the vertical wharf shoreline pointing to a sea side direction as a vertical axis direction, and a direction upward perpendicular to a wharf ground as a vertical axis direction;

[0047] a first falling point determining module configured to take the cabin three-dimensional center coordinate as a first falling reference point of a spout drum in a ship loader, load target bulk cargo to the target ship based on the first falling reference point to determine a material pile of the target bulk cargo, and continuously monitor a front-rear inclination angle, a left-right inclination angle of the target ship, and a vertical distance between the spout drum and the material pile;

[0048] a second falling point determining module configured to, when the front-rear inclination angle or the left-right inclination angle is greater than a preset angle threshold, or the vertical distance is less than a preset safety distance, update the first falling reference point to determine a second falling reference point according to the front-rear inclination angle, the left-right inclination angle, and a static offset correction amount of the first falling reference point by using a preset falling point correction model, wherein the static offset correction amount is determined by a cabin position of the target ship;

[0049] a falling point adjusting module configured to drive a cart traveling mechanism in the ship loader to move horizontally along the wharf ground and / or adjust a rotation angle of a cantilever in the ship loader until the spout drum in the ship loader is aligned with the second falling reference point, so as to continue loading the target bulk cargo to the target ship;

[0050] an operation control optimizing module configured to repeat the above steps until the target ship is completed loading, so as to complete optimization of ship loader operation control.

[0051] An electronic device is provided to adapt to another object of the present application, comprising a central processing unit and a memory, the central processing unit is used to call and run a computer program stored in the memory to execute the steps of the ship loader operation control optimizing method described in the present application.

[0052] A computer readable storage medium is provided to adapt to another object of the present application, which stores a computer program implemented according to the ship loader operation control optimizing method in the form of computer readable instructions, when the computer program is called and run by a computer, the steps included in the corresponding method are executed.

[0053] Compared with the prior art, the present application is directed to the problems in the prior art that the artificial dominant operation mode is low in efficiency and high in risk, and the ship is affected by water flow fluctuation, tide change, center of gravity deviation of the material pile, and prone to dynamic attitude change such as inclination and ship body drift, etc. The present application includes but is not limited to the following beneficial effects:

[0054] Firstly, the ship loader operation control optimization method of the present application can effectively avoid the deviation of the center of gravity of the ship caused by uneven material falling, prevent the ship body from being overloaded, inclined or even overturned, and greatly reduce the safety risk of the ship during loading.

[0055] Secondly, the ship loader operation control optimization method of the present application can avoid collision between the chute and the material pile, protect the safety of the ship loader equipment and the goods, and at the same time reduce the problems such as goods scattering and dust pollution caused by collision.

[0056] Thirdly, the ship loader operation control optimization method of the present application considers the static offset correction amount of the engine room position of the target ship and compensates in the material falling point correction model, which makes the determination of the material falling reference point adapt to the natural deviation characteristics of the center of gravity of the ship caused by its own structure.

[0057] Fourthly, the ship loader operation control optimization method of the present application realizes the automation of the loading operation through the steps of repeated data collection, state monitoring, material falling point correction, equipment adjustment and continuous loading, without frequent manual intervention. Compared with the traditional manual operation mode, the time for manual judgment and operation is greatly reduced, the overall efficiency of the loading operation is greatly improved, and the port's large throughput operation demand can be better adapted.

[0058] Fifthly, the ship loader operation control optimization method of the present application can dynamically adapt to the changes of the ship under complex working conditions. During the loading process, the ship will be affected by factors such as water flow, tide, center of gravity deviation of the material pile, etc., and will produce dynamic attitude changes. The present application can monitor and respond to these dynamic attitude changes to adjust the material falling reference point, so that the loading operation can dynamically adapt to various attitude changes of the ship, and still maintain stable and efficient loading operation under complex port operation conditions. BRIEF DESCRIPTION OF DRAWINGS

[0059] The above and / or additional aspects and advantages of the present application will become apparent and more readily appreciated from the following description, taken in conjunction with the following drawings, in which:

[0060] Figure 1A schematic diagram of a network architecture of a ship loader operation control system in an embodiment of the present application;

[0061] Figure 2 A schematic diagram of a flow of a ship loader operation control optimization method in an embodiment of the present application;

[0062] Figure 3 A schematic diagram of a flow of determining a first blanking datum point in an embodiment of the present application;

[0063] Figure 4 A schematic diagram of a flow of determining a horizontal axis coordinate of a second blanking datum point in an embodiment of the present application;

[0064] Figure 5 A schematic diagram of a flow of determining a vertical axis coordinate of a second blanking datum point in an embodiment of the present application;

[0065] Figure 6 A schematic diagram of a flow of determining a vertical axis coordinate of a second blanking datum point in an embodiment of the present application;

[0066] Figure 7 A schematic diagram of a flow of obstacle detection on a moving area of a cart walking mechanism in an embodiment of the present application;

[0067] Figure 8 A schematic diagram of a principle block of a ship loader operation control optimization device in an embodiment of the present application;

[0068] Figure 9 A schematic diagram of a structure of a computer device in an embodiment of the present application. DETAILED DESCRIPTION

[0069] Embodiments of the present application are described in detail below with reference to the attached drawing figures, wherein the same or like reference numerals are used throughout the drawing figures to refer to the same or like elements or elements having the same or similar functionality. The embodiments described below are exemplary and are intended to be illustrative of the present application and are not to be construed as limiting of the present application.

[0070] It is to be understood that the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. It is further understood that the terms "comprise" (and any grammatical variations thereof, such as "comprises" and "comprising"), when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It is further understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements can be present. Furthermore, "connected" or "coupled" as used herein can include wirelessly connected or wirelessly coupled. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0071] The various data involved in the present application can be stored remotely on a server or locally on a terminal device, unless otherwise specified, as long as it is suitable for being called by the technical solutions of the present application.

[0072] Those skilled in the art should know that the various methods of the present application, although based on the same concept and described to present commonality among them, can be independently executed unless otherwise specified. Similarly, for each of the embodiments disclosed in the present application, they are based on the same inventive concept and therefore the same concept expressed and although the concept is expressed differently but only for the convenience of appropriate transformation of the concept should be understood equally.

[0073] The various embodiments to be disclosed in the present application can be flexibly constructed into new embodiments by cross-combining the related technical features of each embodiment unless it is explicitly indicated that they are mutually exclusive. For this variation, those skilled in the art should know.

[0074] Please refer to Figure 1 , the exemplary network architecture diagram of the ship loader operation control system is as shown in Figure 1As shown, the ship loader operation control optimization method of the present application can be implemented based on a ship loader operation control system, which includes a device layer, a link layer, a data layer, an intermediate layer, an application layer, etc. The device layer includes all constituent devices of the intelligent ship loading system, which can be divided into an intelligent sensing part and an intelligent control part, and other communication devices that unify all devices into the same network and communication protocol. The intelligent sensing part includes a ship cabin sensing device composed of a laser scanner, a gimbal, and a material level radar, an absolute value encoder, an angle tilt meter, an RFID positioning calibration device, and a positioning and anti-collision device composed of a laser anti-collision radar. The intelligent control part includes a computing processing center, a server, and a modified ship loader PLC system.

[0075] The link layer includes a Profitbus bus network and an industrial Ethernet network, and data is uniformly accessed to the Ethernet network through switches, protocol converters, and other devices. The link layer forms a transparent communication relationship between the lower layer basic devices in the form of hardware lines and communication protocols.

[0076] In the data layer, the data sources are a point cloud database composed of ship cabin and stockpile three-dimensional data collected by laser scanning devices and material level radars, ranging data collected by anti-collision detection devices, and calibration position data of RFID. The computing processing center processes the data, the server manages the data and establishes a database, and the self-control database, the point cloud processing database, and the anti-collision database are obtained. The computing processing center and the server share data, the server stores historical databases and system databases, and the client shares the data of the ship loading process by accessing the server.

[0077] The intermediate layer, i.e., according to the database, designs several modules to support the functions required by the application layer, mainly including a stockpile scanning module, an anti-collision module, a self-control module, a visualization module, etc. Among them,

[0078] The stockpile scanning module collects stockpile surface data by laser scanners and stockpile height data by material level radars, processes point cloud data including data preprocessing and feature data extraction based on Open3D, calculates the ship loading route strategy based on the scanned ship type, and transmits the calculated feature data such as stockpile length, width, and inclination angle, and stockpile height under the chute to the point cloud processing database.

[0079] The anti-collision module extracts data features based on data flow and point cloud data processing of the point cloud data collected by the anti-collision radar in the monitoring area of the trolley, cantilever, and chute, and calculates the distance between the obstacle (if existing in the monitoring area) and the probe in real time. If the distance is less than a certain distance, an early warning is sent to the visualization management platform to control the emergency stop.

[0080] The self-control module automatically controls the movement of the cart, cantilever mechanism and belt start-stop process of the ship loader based on the route planning generated by the cabin scanning module, the real-time attitude of the ship loader calculated by the positioning device and the anti-collision monitoring of the anti-collision module, so as to realize the automatic loading process of the ship loader.

[0081] The visualization module calls the real-time database to display the data on the design interface and the digital camera data, so as to realize the process monitoring function. The existing data such as coal flow, PLC state data in other management platforms of the wharf are combined to form a comprehensive data platform of the wharf. Based on KEPServer and the modified ship loader PLC system, a remote control hand operation platform of the ship loader is provided.

[0082] The application layer can realize real-time viewing of the characteristic data and the field picture of the loading process in the visualization management platform, so as to obtain the real-time situation of the field. The main functions include:

[0083] 1. Automatic control of loading: automatic scanning of ship type, generation of loading strategy, automatic loading according to the determined loading route and wharf loading procedures, remote control of the ship loader through the remote control hand operation platform or the hand operation interface.

[0084] 2. Anti-collision automatic monitoring: automatic and real-time monitoring of the collision risk of foreign matters in the anti-collision area of the cart, the anti-collision area of the cantilever and the anti-collision area of the chute, and issuing of collision warning or emergency stop treatment according to the distance from the foreign matters.

[0085] 3. Data management: management of data through MySQL software, realization of system data transparency and fault tracing function.

[0086] Based on the above example, please refer to Figure 2 In one embodiment of the ship loader operation control optimization method of the present application, the method comprises:

[0087] Step S10: collecting laser radar point cloud data corresponding to the target cabin of the target ship, and determining a cabin three-dimensional center coordinate corresponding to the target cabin according to the laser radar point cloud data, wherein a spatial coordinate system in which the cabin three-dimensional center coordinate is located is constructed with a direction along a wharf shoreline and pointing to a bow direction as a horizontal axis direction, a direction along a vertical wharf shoreline and pointing to a sea side direction as a vertical axis direction, and a direction perpendicular to a wharf ground and upward as a vertical axis direction;

[0088] The ship loader operation control system in the terminal device can collect laser radar point cloud data corresponding to a target ship cabin of a target ship, and determine a ship cabin three-dimensional center coordinate corresponding to the target ship cabin according to the laser radar point cloud data, wherein a space coordinate system in which the ship cabin three-dimensional center coordinate is located is constructed with a direction along a wharf shoreline pointing to a bow direction as a horizontal axis direction, a direction along a vertical wharf shoreline pointing to a sea side direction as a longitudinal axis direction, and a direction perpendicular to a wharf ground upward as a vertical axis direction.

[0089] Specifically, in a space coordinate system with a direction along a wharf shoreline pointing to a bow direction as a horizontal axis (X-axis) direction, a direction along a vertical wharf shoreline pointing to a sea side direction as a longitudinal axis (Y-axis) direction, and a direction perpendicular to a wharf ground upward as a vertical axis (Z-axis) direction, the three-dimensional geometric center coordinate of the target ship cabin is accurately obtained through the laser radar point cloud data, which provides an initial position basis for subsequent setting of a blanking reference point and is the basis for realizing accurate blanking of the ship loader. A laser scanning device installed on a cantilever of the ship loader can be used to scan the entire target ship cabin of the target ship, so as to ensure that complete three-dimensional point cloud data of the inner wall of the ship cabin, the bottom of the ship cabin, the deck and the like are obtained. After the ship loader obtains an automatic loading instruction, the ship loader is first moved to a preset ship cabin scanning point by driving a cart walking mechanism, and the laser scanning device is driven by a cloud platform to realize 360° three-dimensional scanning, thereby generating original laser radar point cloud data of the ship cabin.

[0090] In some embodiments, referring to Figure 3 , the step of determining the first blanking reference point comprises:

[0091] Step S101, based on a laser scanning device installed on a cantilever of the ship loader, laser radar point cloud data corresponding to a target ship cabin of a target ship is collected;

[0092] Step S102, a voxel center downsampling method is used to downsample the laser radar point cloud data, a preset statistical filtering algorithm is used to calculate an average distance of each point and a plurality of adjacent points based on the downsampled laser radar point cloud data, an average value and a standard deviation of all average distances are calculated, a distance threshold value is calculated and determined according to the average value and the standard deviation, and it is detected whether the average distance is greater than the distance threshold value, and if so, the point is removed as an outlier.

[0093] The laser radar point cloud data contains noise points (such as dust interference and equipment errors) and redundant points, which need to be processed by the voxel center downsampling and statistical filtering algorithm to retain the geometric features of the ship cabin while reducing the data volume and removing outliers.

[0094] Based on the voxel center downsampling method, the laser radar point cloud data is divided into a plurality of three-dimensional grids according to a preset voxel edge length (such as 5cm) to determine a plurality of non-empty voxels, and the voxel center coordinates of each non-empty voxel are selected to replace all points in the current non-empty voxel, which not only reduces the data amount, but also avoids the geometric structure distortion of the original laser radar point cloud data, greatly improves the subsequent calculation efficiency to reduce the time consumption of bulk cargo loading control. The calculation formula of the voxel center coordinates of each non-empty voxel is represented as:

[0095] ,

[0096] wherein, represents the center coordinates of the non-empty voxel in the X-axis direction; represents the center coordinates of the non-empty voxel in the Y-axis direction; represents the center coordinates of the non-empty voxel in the Z-axis direction; represents the minimum coordinate value of the laser radar point cloud data in the X-axis direction; represents the minimum coordinate value of the laser radar point cloud data in the Y-axis direction; represents the minimum coordinate value of the laser radar point cloud data in the Z-axis direction; represents the index of the voxel in the column direction; represents the index of the voxel in the row direction; represents the index of the voxel in the layer direction; represents the edge length of the voxel.

[0097] Further, the laser radar point cloud data after downsampling is traversed, and a preset statistical filtering algorithm is used to calculate the average distance between each point and its nearest neighbor points based on the laser radar point cloud data after downsampling; the average value and the standard deviation of all average distances are calculated, and the distance threshold is determined according to the average value and the standard deviation , wherein the calculation formula of the distance threshold is represented as:

[0098] ,

[0099] wherein, represents the distance threshold, which is used to judge whether the point is an outlier point; represents the average value of the average distance of all points and their nearest neighbor points; represents a proportion coefficient, which is a constant, used to adjust the size of the distance threshold and affect the strictness of outlier elimination; represents all points and their nearest The standard deviation of the average distance between neighboring points reflects the dispersion of the average distance.

[0100] Furthermore, it is detected whether the average distance is greater than the distance threshold. If the value is greater than 1, then the point is removed as an outlier to eliminate points that are not in harmony with the original point. The average distance between the neighboring points is greater than the distance threshold. point.

[0101] As shown in step S102 above, the original point cloud is divided into a three-dimensional mesh according to the preset voxel side length. The center coordinates of each non-empty voxel are selected to replace all points within the non-empty voxel. This reduces the amount of data and avoids geometric distortion of the original point cloud, greatly improving subsequent calculation efficiency and meeting the real-time requirements of ship loading. By traversing the downsampled lidar point cloud data, a preset statistical filtering algorithm is used to remove outliers from the downsampled lidar point cloud data. This can eliminate noise points caused by interference such as dock dust and light reflection, ensuring that the lidar point cloud data retains only valid points related to the ship's cabin structure.

[0102] Step S103: The RANSAC algorithm is used to fit the statistically filtered lidar point cloud data to extract the point cloud data of the cabin wall corresponding to the target cabin. The cabin length, cabin width, ship depth and the vertical axis coordinate of the bottom of the cabin in the vertical direction are determined based on the cabin wall point cloud data.

[0103] RANSAC (Random Sample Consensus) is an iterative algorithm that correctly estimates mathematical model parameters from a set of data containing "outliers." "Outliers" generally refer to noise in the data, such as mismatches in matching and outliers in the estimated curve. The statistically filtered lidar point cloud data is fitted to the ship's cabin wall using the RANSAC algorithm to extract key parameters needed to calculate the three-dimensional center coordinates, such as cabin length, cabin width, ship depth, and the vertical coordinates of the cabin bottom.

[0104] The RANSAC algorithm iteratively selects point cloud data that conforms to the planar model of the ship's cabin walls, eliminating noise points, and can accurately fit the fore and aft walls along the X-axis, the left and right walls along the Y-axis, and the bottom wall along the Z-axis. For example, the extreme difference of the X-axis coordinates of the fore and aft wall point clouds is the length of the cabin. The difference in extreme values ​​of the Y-axis coordinates of the point clouds on the left and right walls is the width of the cabin. The extreme difference between the Z-axis coordinates of the bottom bulkhead and the deck bulkhead is the ship's depth. The Z-axis coordinate mean value of the cabin bottom wall point cloud is the vertical axis coordinate of the cabin bottom in the vertical axis direction .

[0105] In step S104, the horizontal axis coordinate in the cabin three-dimensional center coordinate is determined according to the fourth sum value between the starting coordinate of the target cabin in the horizontal axis direction and half of the cabin length, the vertical axis coordinate in the cabin three-dimensional center coordinate is determined according to the fifth sum value between the starting coordinate of the target cabin in the vertical axis direction and half of the cabin width, and the vertical axis coordinate in the cabin three-dimensional center coordinate is determined according to the sixth sum value between the vertical axis coordinate of the cabin bottom in the vertical axis direction and half of the ship depth.

[0106] Specifically, the calculation formula of the horizontal axis coordinate in the cabin three-dimensional center coordinate of the target cabin is represented as:

[0107] ,

[0108] wherein, represents the horizontal axis coordinate in the cabin three-dimensional center coordinate of the target cabin; represents the starting coordinate of the target cabin in the horizontal axis direction, which can be determined by the most front end point coordinate of the cabin in the horizontal axis direction in the laser radar point cloud data; represents the cabin length;

[0109] The calculation formula of the vertical axis coordinate in the cabin three-dimensional center coordinate of the target cabin is represented as:

[0110] ,

[0111] wherein, represents the vertical axis coordinate in the cabin three-dimensional center coordinate of the target cabin; represents the starting coordinate of the target cabin in the vertical axis direction, which can be determined by the most shore side end point coordinate of the cabin in the vertical axis direction in the laser radar point cloud data; represents the cabin width.

[0112] The calculation formula of the vertical axis coordinate in the cabin three-dimensional center coordinate of the target cabin is represented as:

[0113] ,

[0114] wherein, represents the vertical axis coordinate in the cabin three-dimensional center coordinate of the target cabin; represents the vertical axis coordinate of the cabin bottom in the vertical axis direction; represents the ship depth.

[0115] Based on the above calculation formula, the horizontal axis (X-axis) coordinate, the longitudinal axis (Y-axis) coordinate and the vertical axis (Z-axis) coordinate in the three-dimensional center coordinate of the target ship cabin can be calculated and determined.

[0116] In step S105, the horizontal axis coordinate, the longitudinal axis coordinate and the vertical axis coordinate in the three-dimensional center coordinate of the ship cabin are used to determine the first blanking reference point.

[0117] The calculated , and are taken as the three-dimensional center coordinate of the target ship cabin, which is the first blanking reference point of the spout of the ship loader. When the spout of the ship loader initially blanks, it will be aligned with the first blanking reference point to achieve the preliminary accurate loading of the target bulk cargo into the target ship cabin.

[0118] As can be seen from the above embodiment, the three-dimensional center coordinate of the target ship cabin calculated by the above steps S101 to S105 is taken as the first blanking reference point of the spout of the ship loader, which can ensure that the initial blanking position is located in the center area of the ship cabin, avoiding the deviation of the blanking point to cause the deviation of the material pile, such as the material pile only accumulating in the bow, the stern, the sea side or the shore side, etc., which is the basis for uniform loading of bulk cargo and preventing the ship from tilting.

[0119] In step S20, the three-dimensional center coordinate of the ship cabin is taken as the first blanking reference point of the spout of the ship loader, and the target bulk cargo is loaded into the target ship based on the first blanking reference point to determine the material pile of the target bulk cargo, and the front and rear inclination angles, the left and right inclination angles of the target ship corresponding to the target ship, and the vertical distance between the spout and the material pile are continuously monitored.

[0120] The laser radar point cloud data corresponding to the target ship cabin of the target ship is collected, and the three-dimensional center coordinate of the target ship cabin is determined according to the laser radar point cloud data. Then, the three-dimensional center coordinate of the ship cabin is taken as the first blanking reference point of the spout of the ship loader, and the target bulk cargo is loaded into the target ship based on the first blanking reference point to determine the material pile of the target bulk cargo, and the front and rear inclination angles, the left and right inclination angles of the target ship corresponding to the target ship, and the vertical distance between the spout and the material pile are continuously monitored. The front and rear inclination angles represent the included angle between the horizontal axis of the ship body along the wharf shore line pointing to the bow direction and the horizontal reference plane parallel to the wharf ground when the target ship rotates around the longitudinal axis along the vertical wharf shore line pointing to the sea side direction. The left and right inclination angles represent the included angle between the vertical axis of the ship body along the wharf shore line pointing to the bow direction and the horizontal reference plane parallel to the wharf ground when the target ship rotates around the horizontal axis along the wharf shore line pointing to the bow direction. The target bulk cargo includes coal, ore or grain, etc.

[0121] Specifically, the target ship cabin corresponding three-dimensional center coordinates of the cabin are determined according to the laser radar point cloud data, and the three-dimensional center coordinates of the cabin are taken as the first falling reference point of the spout cone in the ship loader. The trolley walking mechanism (moving horizontally along the wharf ground) and the cantilever rotating mechanism (adjusting the angle of the cantilever) of the ship loader will act in coordination to make the spout cone align with the first falling reference point, and then the spout cone releases bulk cargo to form an initial pile in the cabin. The form (such as height and area) of the pile will change continuously during loading, providing a dynamic object for subsequent vertical distance monitoring. The vertical distance, that is, the vertical distance from the spout cone outlet to the top of the pile, needs to be kept within a safe range to avoid the spout cone colliding with the pile or bulk cargo splashing.

[0122] During the loading of the target ship, if the pile is unbalanced, for example, there is more cargo on the bow side and less cargo on the stern side, the ship will produce a front-to-back tilt around the vertical shore line pointing to the sea side (Y axis) to produce a front-to-back tilt angle; if the pile on the sea side or the shore side is uneven, it will produce a left-to-right tilt angle around the shore line pointing to the bow (X axis). Monitoring the front-to-back tilt angle and the left-to-right tilt angle is to determine whether the ship is excessively tilted due to uneven loading.

[0123] As the height of the pile rises, the vertical distance between the spout cone and the pile will gradually decrease. If the vertical distance is too small, the spout cone may collide with the pile, causing equipment damage or bulk cargo splashing; if the vertical distance is too large, the impact of the falling bulk cargo is easy to cause the pile to collapse or dust. Monitoring the vertical distance between the spout cone and the pile is to ensure the safety and standardization of the loading operation by adjusting the falling point or lifting the spout cone height when the distance is less than the preset safe distance.

[0124] As can be seen from the above step S20, the first falling reference point is used to start loading accurately, and the ship tilt and the vertical distance between the spout cone and the pile are detected in real time to provide data for subsequent adjustment, which not only ensures the accuracy of initial loading, but also provides a monitoring basis for subsequent dynamic optimization.

[0125] In step S30, when the front-to-back tilt angle or the left-to-right tilt angle is greater than a preset angle threshold, or the vertical distance is less than a preset safe distance, a preset falling point correction model is used to update the first falling reference point according to the front-to-back tilt angle, the left-to-right tilt angle and the static offset correction amount of the first falling reference point, to determine a second falling reference point, wherein the static offset correction amount is determined by the cabin position of the target ship.

[0126] The three-dimensional center coordinates of the hold are taken as a first feeding reference point of a spout tube in a ship loader, target bulk cargos are loaded into the target ship based on the first feeding reference point to determine a stockpile of the target bulk cargos, and after continuously monitoring a front and rear inclination angle, a left and right inclination angle of the target ship, and a vertical distance between the spout tube and the stockpile, when the front and rear inclination angle or the left and right inclination angle is greater than a preset angle threshold, or the vertical distance is less than a preset safety distance, a preset feeding point correction model is used to update the first feeding reference point according to the front and rear inclination angle, the left and right inclination angle, and a static offset correction amount of the first feeding reference point, to determine a second feeding reference point, wherein the static offset correction amount is determined by a cabin position of the target ship, the preset angle threshold includes 2° or 3°, for example, in the general safety standard of bulk cargo ship loading operation, the left and right inclination angle of the ship is usually not more than 3 degrees, and the front and rear inclination angle is not more than 2 degrees, to avoid excessive difference in draft at the bow and stern, causing stress concentration of the ship structure; the preset safety distance can be randomly taken between 0.3 m and 1.0 m, and a person skilled in the art can determine the preset safety distance according to the actual operation scene, which is not limited here.

[0127] The front and rear inclination angle and the left and right inclination angle of the target ship can be collected in real time by an inclination angle instrument, reflecting the dynamic attitude change of the ship caused by uneven loading. The feeding point correction model of the present application compensates the horizontal and vertical axis coordinates of the first feeding reference point based on the front and rear inclination angle and the left and right inclination angle, for example: if the ship is “inclined by the bow”, that is, the bow is low and the stern is high, the front and rear inclination angle is positive, and the feeding point needs to be offset to the stern direction (X axis negative direction) to balance the weight of the bow stockpile; if the ship is “inclined to the right”, that is, the sea side is low and the shore side is high, the left and right inclination angle is positive, and the feeding point needs to be offset to the shore side direction (Y axis negative direction) to reduce the pressure of the sea side stockpile.

[0128] The static offset correction amount is determined by the cabin position of the target ship. Since the ship cabin is usually a fixed heavy area, it will cause inherent gravity center offset of the ship, that is, even if the loading is uniform, one side of the cabin will be slightly low due to self weight. The cabin position can be identified by point cloud processing to determine the static offset correction amount in advance, for example, when the cabin is at the stern, the three-dimensional center coordinates of the hold need to be offset by 50 cm to the bow direction to balance the self weight of the cabin. The feeding point correction model of the present application fuses the static offset correction, which can avoid the problem of “uniform loading but still inclined” caused by the inherent gravity center offset of the ship.

[0129] For the risk of ship roll, the front and rear inclination angles and the left and right inclination angles of the target ship need to be detected, and the ship will roll over if the inclination is too large. The preset angle threshold is set to avoid this risk, and the preset angle threshold usually needs to be determined in combination with the stability parameters of the target ship to avoid the inclination angle exceeding the safe load range of the ship. The preset safe distance is set to prevent the chute from colliding with the material pile due to the increase in the height of the material pile, and to avoid equipment damage or bulk cargo splashing.

[0130] Further, when it is detected that the front and rear inclination angles or the left and right inclination angles of the target ship are greater than the preset angle threshold, or the vertical distance between the chute and the material pile is less than the preset safe distance, the risk of excessive inclination of the ship or the distance between the chute and the material pile during the loading operation, the real-time attitude parameters of the ship containing the front and rear inclination angles and the left and right inclination angles and the static offset correction amount are fused to dynamically update the initial first material falling reference point, and a safe and adaptive second material falling reference point is generated, so as to avoid accidents such as ship roll and equipment collision caused by improper material falling position, and to ensure the safety and stability of the loading operation.

[0131] Further, the corrected second material falling reference point is a new material falling position that can balance the inclination of the ship and avoid the risk of chute collision, which is obtained on the basis of the first material falling reference point through dynamic angle compensation and static center of gravity correction of the ship. For example, if the left and right inclination angles exceed the preset angle threshold (right inclination) and the static offset correction amount is 20 cm of bank offset, the longitudinal axis (Y axis) coordinate of the material falling point needs to be superimposed with the inclination compensation amount in the bank direction and the 20 cm static offset amount on the basis of the first material falling reference point to ensure that the ship attitude returns to the safe range after material falling.

[0132] As can be seen from the above step S30, by monitoring the risk trigger condition in real time, the material falling point correction model is actively called to update the material falling point, which avoids the inefficient mode of traditional manual risk discovery and shutdown adjustment, so as to achieve intelligent control of unmanned operation, and provides accurate position instructions for subsequent equipment action adjustment, ensuring the continuous progress of the loading operation under the premise of safety.

[0133] In some embodiments, referring to Figure 4 , the step of updating the first material falling reference point to determine the second material falling reference point according to the front and rear inclination angles, the left and right inclination angles, and the static offset correction amount of the first material falling reference point includes:

[0134] Step S31, obtaining the horizontal axis coordinate and the vertical axis coordinate of the first blanking reference point, the horizontal axis static offset correction amount of the first blanking reference point, the ship depth of the target ship, the fore-aft inclination angle of the target ship, the mass of the target bulk cargo, the total mass of the target ship, and the average vertical axis coordinate, wherein the ship depth represents the vertical distance between the bottom of the ship cabin and the deck of the target ship, and the average vertical axis coordinate represents the average value between the vertical axis coordinate of the first blanking reference point, the vertical axis coordinate of the center of the engine room of the target ship, and the vertical axis coordinate of the center of the material pile;

[0135] Step S32, calculating the first product between one-half of the ship depth and the sine value of the fore-aft inclination angle, calculating the first difference between the vertical axis coordinate of the first blanking reference point and the average vertical axis coordinate, calculating the first ratio between the mass of the target bulk cargo and the total mass of the target ship, and calculating the second product between the first ratio, the first difference, and the sine value of the fore-aft inclination angle;

[0136] Step S33, determining the horizontal axis coordinate of the second blanking reference point according to the first sum value between the horizontal axis coordinate of the first blanking reference point, the static offset correction amount of the first blanking reference point, the first product, and the second product.

[0137] Specifically, the calculation formula of the horizontal axis coordinate of the second blanking reference point is represented as:

[0138]

[0139] wherein, represents the horizontal axis coordinate of the second blanking reference point; represents the horizontal axis coordinate of the first blanking reference point, i.e., the horizontal axis coordinate of the three-dimensional center coordinate of the ship cabin; represents the horizontal axis static offset correction amount of the first blanking reference point, which is used to compensate for the deviation of the blanking point in the horizontal axis direction caused by static factors such as the position of the engine room; represents the ship depth of the target ship, i.e., the vertical distance between the bottom of the ship cabin and the deck; represents the fore-aft inclination angle of the target ship, which represents the included angle between the horizontal axis of the ship body pointing to the bow direction along the wharf shoreline and the horizontal reference plane parallel to the ground of the wharf when the target ship rotates around the longitudinal axis pointing to the sea side direction along the wharf shoreline; represents the mass of the target bulk cargo; represents the total mass of the target ship, which represents the total mass including the mass of the target bulk cargo and the mass of the ship itself; represents the vertical axis coordinate of the first blanking reference point; ​represents an average vertical axis coordinate, which is an average value among a vertical axis coordinate of the first material dropping reference point, a vertical axis coordinate of the center of the engine room of the target ship, and a vertical axis coordinate of the center of the material pile.

[0140] According to the above calculation formula, the horizontal axis coordinate of the first material dropping reference point is corrected according to the horizontal axis static deviation of the target ship caused by the engine room position, the fore-aft inclination angle , the bulk cargo quality ratio , and the vertical axis direction position difference , so as to avoid the deviation of the bulk cargo in the horizontal axis direction caused by the structure of the ship itself or the fore-aft inclination, the bulk cargo distribution, and the like during the loading process, to ensure the uniform and safe loading of the bulk cargo in the horizontal axis direction of the engine room, and to improve the accuracy and stability of the loading.

[0141] In a further embodiment, referring to Figure 5 , the step of updating the first material dropping reference point according to the fore-aft inclination angle, the left-right inclination angle, and the static deviation correction amount of the first material dropping reference point to determine the second material dropping reference point includes:

[0142] Step S301, obtaining the longitudinal axis coordinate of the first material dropping reference point, the longitudinal axis static deviation correction amount of the first material dropping reference point, the left-right inclination angle of the target ship, and the longitudinal axis dynamic deviation amount of the first material dropping reference point;

[0143] Step S302, calculating a third product between one-half of the ship's depth and the sine value of the left-right inclination angle, calculating a second ratio between the vertical axis coordinate of the first material dropping reference point and the average vertical axis coordinate, and calculating a fourth product between the first ratio, the longitudinal axis dynamic deviation amount of the first material dropping reference point, the second ratio, and the sine value of the left-right inclination angle;

[0144] Step S303, determining the longitudinal axis coordinate of the second material dropping reference point according to the second sum value between the longitudinal axis coordinate of the first material dropping reference point, the longitudinal axis static deviation correction amount of the first material dropping reference point, the third product, and the fourth product.

[0145] Specifically, the calculation formula of the longitudinal axis coordinate of the second material dropping reference point is represented as:

[0146] ,

[0147] wherein, represents the longitudinal axis coordinate of the second material dropping reference point; represents the longitudinal axis coordinate of the first material dropping reference point; represents a longitudinal axis static offset correction amount of the first material dropping reference point, used to compensate for the longitudinal axis direction material dropping point deviation caused by static factors such as engine room position; represents the ship depth of the target ship, that is, the vertical distance between the ship cabin bottom and the deck; represents the left-right inclination angle of the target ship, which represents the included angle between the longitudinal axis of the ship body pointing to the sea side direction perpendicular to the wharf line and the horizontal reference plane parallel to the wharf ground when the target ship rotates around the transverse axis pointing to the bow direction along the wharf line; represents the quality of the target bulk cargo; represents the total mass of the target ship, which represents the total mass including the quality of the target bulk cargo and the mass of the ship itself; represents the longitudinal axis dynamic offset amount of the first material dropping reference point, reflecting the longitudinal axis direction material dropping point dynamic deviation caused by dynamic factors such as loading of bulk cargo in the ship cabin; represents the vertical axis coordinate of the first material dropping reference point; represents the average vertical axis coordinate, which is the average value between the vertical axis coordinate of the first material dropping reference point, the vertical axis coordinate of the engine room center of the target ship, and the vertical axis coordinate of the material pile center.

[0148] As can be seen from the above calculation formula, according to the longitudinal axis static offset correction amount of the ship, the left-right inclination angle , the bulk cargo quality ratio , the longitudinal axis dynamic offset amount , and the vertical axis coordinate relationship , the longitudinal axis coordinate of the first material dropping reference point is adjusted, which can effectively solve the longitudinal axis static deviation of the ship caused by the structure such as the engine room, and the longitudinal axis direction material dropping point deviation caused by the left-right inclination and dynamic distribution of bulk cargo during loading, to ensure the rationality of the bulk cargo loading in the longitudinal axis direction of the ship cabin, and reduce the inclination risk caused by uneven loading in the longitudinal axis direction of the ship.

[0149] In a further embodiment, please refer to Figure 6 , the step of updating the first material dropping reference point according to the front-rear inclination angle, the left-right inclination angle, and the static offset correction amount of the first material dropping reference point to determine the second material dropping reference point by using a preset material dropping point correction model, includes:

[0150] Step S3001, obtaining the draft of the target ship;

[0151] Step S3002, calculating a second difference value between the draft of the target ship and half of the ship depth, and calculating a fifth product between the second difference value, the cosine value of the front-rear inclination angle, and the cosine value of the left-right inclination angle;

[0152] Step S3003, determining the vertical axis coordinate of the second blanking datum point according to a third sum between the vertical axis coordinate of the first blanking datum point and the fifth product;

[0153] Step S3004, constructing the second blanking datum point according to the horizontal axis coordinate, the vertical axis coordinate and the vertical axis coordinate of the second blanking datum point.

[0154] Specifically, the calculation formula of the vertical axis coordinate of the second blanking datum point is represented as:

[0155]

[0156] wherein, represents the vertical axis coordinate of the second blanking datum point; represents the vertical axis coordinate of the first blanking datum point; represents the draft of the target ship, i.e. the vertical distance from the water surface to the bottom of the ship cabin when the target ship is in the water; represents the molded depth of the target ship, i.e. the vertical distance between the bottom of the ship cabin and the deck.

[0157] As can be seen from the above calculation formula, the vertical axis coordinate of the blanking datum point is corrected according to the draft of the ship , the molded depth of the ship , the fore-aft inclination angle and the port-starboard inclination angle , which fully considers the attitude change of the ship in the vertical direction, so that the blanking position of the chute in the vertical axis direction can adapt to the actual state of the ship, avoiding the improper distance between the chute and the material pile due to the attitude change of the ship in the vertical direction, preventing the collision between the chute and the material pile, and controlling the dust generated by the falling of the bulk cargo, thereby ensuring the safety and environmental protection of the loading process.

[0158] In some embodiments, the calculation formula of the blanking point correction model is represented as:

[0159]

[0160] wherein, represents the horizontal axis coordinate of the second blanking datum point; represents the horizontal axis coordinate of the first blanking datum point, i.e. the horizontal axis coordinate of the three-dimensional center coordinate of the ship cabin; represents the horizontal axis static offset correction amount of the first blanking datum point, which is used to compensate for the deviation of the horizontal axis direction blanking point caused by static factors such as the engine room position; represents the molded depth of the target ship, i.e. the vertical distance between the bottom of the ship cabin and the deck. ​​represents a fore-aft inclination angle of the target ship, which represents an included angle between a longitudinal axis of the ship body pointing to a sea side direction along a wharf line and a horizontal reference plane parallel to a wharf ground when the target ship rotates around a longitudinal axis pointing to the sea side direction along the wharf line; represents a mass of the target bulk cargo; represents a total mass of the target ship, which represents a total mass including the mass of the target bulk cargo and a mass of the ship itself; represents a vertical axis coordinate of the first discharging reference point; represents an average vertical axis coordinate, which is an average value among the vertical axis coordinate of the first discharging reference point, a vertical axis coordinate of a center of the engine room of the target ship, and a vertical axis coordinate of a center of the bulk cargo pile; represents a longitudinal axis coordinate of the second discharging reference point; represents a longitudinal axis coordinate of the first discharging reference point; represents a longitudinal axis static offset correction amount of the first discharging reference point, which is used to compensate for a deviation of the discharging point in the longitudinal axis direction caused by static factors such as the position of the engine room; represents a left-right inclination angle of the target ship, which represents an included angle between a transverse axis of the ship body pointing to a bow direction along a wharf line and a horizontal reference plane parallel to a wharf ground when the target ship rotates around the transverse axis pointing to the bow direction along the wharf line; represents a longitudinal axis dynamic offset amount of the first discharging reference point, which reflects a dynamic deviation of the discharging point in the longitudinal axis direction caused by dynamic factors such as loading of the bulk cargo in the ship cabin; represents a vertical axis coordinate of the second discharging reference point; represents a draft of the target ship, which represents a vertical distance from a water surface to a bottom of the ship cabin when the target ship is in the water.

[0161] In step S40, a walking mechanism of the ship loader is driven to move horizontally along the wharf ground and / or a rotation angle of a boom of the ship loader is adjusted until a chute of the ship loader is aligned with the second discharging reference point, so as to continue loading the target bulk cargo to the target ship.

[0162] After the first discharging reference point is updated to determine the second discharging reference point according to the fore-aft inclination angle, the left-right inclination angle, and the static offset correction amount of the first discharging reference point by using a preset discharging point correction model, the walking mechanism of the ship loader is driven to move horizontally along the wharf ground and / or the rotation angle of the boom of the ship loader is adjusted until the chute of the ship loader is aligned with the second discharging reference point, so as to continue loading the target bulk cargo to the target ship.

[0163] Based on the second blanking reference point, the chute is precisely moved to a new safe blanking point for continuous loading by controlling the cart walking mechanism and the boom of the ship loader. By controlling the cart walking mechanism, the boom, and the like, the spatial position of the chute is adjusted to switch from aligning the first blanking reference point to aligning the second blanking reference point, so as to avoid the aggravation of the inclination of the ship or the collision of the chute due to the deviation of the blanking point, while ensuring that the target bulk cargo is continuously and accurately loaded into the target hold.

[0164] In some embodiments, the cart walking mechanism of the ship loader is usually installed on the wharf track, and the core function is to drive the entire ship loader to move horizontally along the ground of the wharf. The movement direction is directly matched with the spatial coordinate system defined in step S10. If the horizontal axis coordinate (pointing to the bow direction along the wharf shoreline) of the second blanking reference point deviates from the first blanking reference point, for example, due to the forward and backward inclination of the ship, the blanking point needs to be offset to the stern, and the cart walking mechanism will move in the direction "parallel to the wharf shoreline" to adjust the position of the chute in the horizontal axis direction. If the vertical axis coordinate (pointing to the sea side direction perpendicular to the wharf shoreline) of the second blanking reference point deviates, for example, due to the left and right inclination of the ship, the blanking point needs to be offset to the shoreline, and the cart walking mechanism will move in the direction "perpendicular to the wharf shoreline" to correct the position of the chute in the vertical axis direction. The movement of the cart walking mechanism can make the chute in the ship loader align with the horizontal axis coordinate and the vertical axis coordinate of the second blanking reference point in the horizontal plane, which is the core action to realize the horizontal position correction of the blanking point.

[0165] In some embodiments, the boom of the ship loader is a long rod structure connecting the cart body and the chute, which can rotate around the connection point (rotation center) of the boom and the cart. The adjustment of the rotation angle of the boom can further refine the horizontal position of the chute. When the cart walking mechanism moves to the approximate area, if there is still a small horizontal deviation between the chute and the second blanking reference point, for example, due to the small inclination angle of the ship, the cart does not need to move a long distance. By adjusting the rotation angle of the boom (clockwise or counterclockwise rotation), the chute can be rotated around the rotation center to accurately align with the horizontal coordinates of the second blanking reference point. For example, if the second blanking reference point needs to be fine-tuned by 0.5 meters to the sea side in the vertical axis direction, the cart does not need to move as a whole, and only by rotating the boom to a certain angle to the sea side, the chute can reach the target position, which has flexibility and efficiency.

[0166] In some embodiments, the gantry walking mechanism in the ship loader is driven to move horizontally along the wharf ground and / or the rotation angle of the cantilever in the ship loader is adjusted, when the second blanking reference point deviates from the first blanking reference point by a large amount, for example, the ship tilts more than a threshold value, and the blanking point needs to be deviated by a large amount, the combination of moving by the gantry walking mechanism and rotating the cantilever is used, the target area is quickly moved by the gantry walking mechanism first, and then the cantilever is fine-tuned for alignment, taking into account efficiency and accuracy. When the deviation is small, for example, only a small correction is needed, only the rotation angle of the cantilever needs to be adjusted, avoiding frequent starting and stopping of the gantry, reducing equipment wear and tear, and improving operation continuity.

[0167] In some embodiments, referring to Figure 7 , the gantry walking mechanism in the ship loader is driven to move horizontally along the wharf ground and / or the rotation angle of the cantilever in the ship loader is adjusted, until the chute in the ship loader is aligned with the second blanking reference point, the step of

[0168] Step S401, acquiring a moving area RGB image containing a target obstacle in the horizontal movement of the gantry walking mechanism along the wharf ground and a corresponding moving area depth image thereof;

[0169] The moving area RGB image can be collected by a monitoring camera added to the gantry walking mechanism of the ship loader, which presents a color picture in the moving area, and can identify the visual features such as color and texture of the target obstacle, such as the shape of the operating personnel or the scattered goods; the moving area depth image can be collected by a laser range finder or a collision avoidance radar, which shows the actual distance from each pixel point in the image to the imaging device, to solve the problem that the RGB image cannot judge the distance of the obstacle. Among them, the moving area RGB image and its corresponding moving area depth image need to be acquired in real time and synchronously, to ensure that the pixel positions of the same obstacle in the RGB image and the depth image correspond one by one, to lay a foundation for subsequent image alignment and coordinate matching, and to avoid deviation of obstacle positioning caused by asynchronous data.

[0170] Step S402, calling an obstacle detection model trained to a convergent state to perform target detection on the moving area RGB image to determine the two-dimensional center pixel coordinates corresponding to the target obstacle, wherein the basic network architecture of the obstacle detection model is a YOLOv7-tiny model, and the target obstacle includes operating personnel or scattered goods;

[0171] Since the edge computing device of the ship loader is usually an industrial industrial computer, the computing power and memory are limited, YOLOv7-tiny is a lightweight version of YOLOv7, the model structure is relatively simple, the parameter quantity and the calculation quantity are less, and the target obstacle in the image can be quickly detected under the premise of ensuring a certain detection accuracy, the demand of real-time detection of the target obstacle of the ship loader is met, and the slow detection speed is avoided. The YOLOv7-tiny model can avoid device overload and ensure that the detection frame rate is stable at 20 to 30 FPS, meeting the needs of real-time scanning and immediate response of the cart walking mechanism, therefore, the YOLOv7-tiny model is used as the basic network architecture of the target obstacle detection model of the application.

[0172] The trained YOLOv7-tiny model is called to detect the target in the moving area RGB image, so as to determine the two-dimensional center pixel coordinates corresponding to the target obstacle. The pixel coordinates corresponding to the geometric center of the target obstacle region framed by the model are calculated, such as the (x, y) value in the image coordinate system. For example, if it is identified that the operation personnel occupy the area of 100th to 200th column and 300th to 400th row in the image, the two-dimensional center pixel coordinates can be calculated as (150, 350), which is the bridge for subsequent association of depth information.

[0173] In step S403, the moving area RGB image and the moving area depth image are aligned, and the depth value in the moving area depth image corresponding to the two-dimensional center pixel coordinates of the target obstacle in the moving area RGB image is matched to determine the three-dimensional center coordinates of the target obstacle in the real space.

[0174] The camera intrinsic parameters (such as focal length and pixel size) and the calibration parameters are used to align the RGB image and the depth image in space, so that the position of each pixel point in the RGB image is completely matched with the position of the corresponding pixel point in the depth image, and the deviation caused by the installation angle of the imaging device and the lens distortion is eliminated.

[0175] According to the two-dimensional center pixel coordinates of the target obstacle obtained in step S402, the depth value of the corresponding pixel point in the aligned depth image is found, that is, the straight-line distance from the target obstacle to the laser range finder, and then the three-dimensional center coordinates of the cart walking mechanism of the ship loader are combined to calculate the three-dimensional center coordinates of the target obstacle in the real space, so as to realize the position mapping from the image to the reality.

[0176] Step S404, if the Euclidean distance between the three-dimensional center coordinates of the target obstacle and the three-dimensional center coordinates of the walking mechanism of the gantry is less than the shutdown distance threshold, a shutdown instruction is sent to the ship loader to force the walking mechanism of the gantry to stop moving; if the Euclidean distance between the three-dimensional center coordinates of the target obstacle and the three-dimensional center coordinates of the walking mechanism of the gantry is between the shutdown distance threshold and the warning distance threshold, an audible and visual alarm is issued to remind the operating personnel to avoid.

[0177] After determining the three-dimensional center coordinates of the target obstacle in the real space, the Euclidean distance between the three-dimensional center coordinates of the target obstacle and the three-dimensional center coordinates of the walking mechanism of the gantry is calculated. If the Euclidean distance between the three-dimensional center coordinates of the target obstacle and the three-dimensional center coordinates of the walking mechanism of the gantry is less than the shutdown distance threshold, it means that the obstacle has entered the dangerous area of the walking mechanism of the gantry, for example, the operating personnel is close to the track of the gantry, the scattered goods block the wheels, etc., a shutdown instruction is sent to the ship loader to force the walking mechanism of the gantry to stop moving; if the Euclidean distance between the three-dimensional center coordinates of the target obstacle and the three-dimensional center coordinates of the walking mechanism of the gantry is between the shutdown distance threshold and the warning distance threshold, an audible and visual alarm can be sent first to remind the operating personnel to avoid, and if the distance continues to decrease, the shutdown instruction is triggered again, which takes into account the safety and the continuity of the operation, wherein the shutdown distance threshold can be 0.8 meters, etc., and the warning distance threshold can be 1.5 meters, etc. The shutdown distance threshold and the warning distance threshold can be determined as needed by a person skilled in the art according to the actual business scene requirements, which are not limited here.

[0178] As can be seen from the above steps S401 to S404, the obstacle detection model of the present application can efficiently and accurately identify obstacles and determine two-dimensional center pixel coordinates, which are converted into real space three-dimensional coordinates through image alignment and camera perspective projection model, and then an audible and visual alarm or a shutdown instruction is issued according to the Euclidean distance classification, which not only adapts to the edge device computing power of the ship loader and ensures the detection response speed in complex environments, but also realizes accurate positioning and hierarchical safety control of obstacles, effectively avoids equipment collision and personnel risk, takes into account the safety and efficiency of the operation, and helps the ship loader to be unmanned and anti-collision.

[0179] Step S50, repeat the above steps S10 to S40 until the target ship is completed, to complete the optimization of the ship loader operation control.

[0180] By repeating the above steps S10 to S40, when the belt scale or metering system of the ship loader detects that the actual coal loading amount reaches the order preset value, the cycle is terminated, the ship loading operation is completed, at this time the system automatically performs the closing of the belt, the ship loader reset and other finishing actions, finally completes the optimization of the ship loader operation control, to realize the dynamic monitoring, risk identification, landing point correction and precise loading iteration until the target ship completes the bulk cargo loading, finally achieves the goal of unmanned, safe and efficient ship loading operation control.

[0181] From the above embodiments, compared with the prior art, the present application is directed to the problems of low efficiency and high risk of manual dominant operation mode in the prior art, and the ship is affected by water flow fluctuation, tidal change, material pile center of gravity deviation, etc., prone to dynamic attitude changes such as inclination and ship drift, etc. The present application includes but is not limited to the following beneficial effects:

[0182] Firstly, the ship loader operation control optimization method of the present application can effectively avoid the deviation of the center of gravity of the ship caused by uneven landing, prevent the ship from being loaded, inclined or even overturned, and greatly reduce the safety risk of the ship during loading.

[0183] Secondly, the ship loader operation control optimization method of the present application can avoid collision between the chute and the material pile, protect the safety of the ship loader equipment and the goods, and at the same time reduce the problems of goods scattering, dust pollution and other problems caused by collision.

[0184] Thirdly, the ship loader operation control optimization method of the present application considers the static offset correction amount of the determined engine room position of the target ship, and compensates in the landing point correction model, which makes the determination of the landing reference point adapt to the natural deviation characteristics of the center of gravity of the ship caused by its own structure.

[0185] Fourthly, the ship loader operation control optimization method of the present application realizes the automation of the ship loading operation by repeating the steps of data collection, state monitoring, landing point correction, equipment adjustment and continuous loading, without the need for frequent manual intervention. Compared with the traditional manual operation mode, the time of manual judgment and operation is greatly reduced, the overall efficiency of the ship loading operation is greatly improved, and the port's large throughput operation demand can be better adapted.

[0186] Fifthly, the ship loader operation control optimization method of the present application can dynamically adapt to the changes of the ship under complex working conditions. During the loading process, the ship will be affected by factors such as water flow, tide, and deviation of the center of gravity of the material pile, and will produce dynamic attitude changes. The present application can monitor and respond to these dynamic attitude changes in real time, adjust the material dropping reference point, dynamically adapt the loading operation to various attitude changes of the ship, and still maintain stable and efficient loading operation under complex port operation conditions.

[0187] Referring to Figure 8 , one of the purposes of the present application is to provide a ship loader operation control optimization device, which comprises a cabin center determination module 1100, a first material dropping point determination module 1200, a second material dropping point determination module 1300, a material dropping point adjustment module 1400, and an operation control optimization module 1500. Among them, the cabin center determination module 1100 is configured to collect laser radar point cloud data corresponding to the target cabin of the target ship, and determine the three-dimensional center coordinates of the target cabin corresponding to the target ship according to the laser radar point cloud data, wherein the space coordinate system in which the three-dimensional center coordinates of the cabin are located is constructed with the direction along the wharf shoreline as the horizontal axis direction, the direction along the vertical wharf shoreline as the vertical axis direction, and the direction perpendicular to the wharf ground as the vertical axis direction; the first material dropping point determination module 1200 is configured to take the three-dimensional center coordinates of the cabin as the first material dropping reference point of the chute in the ship loader, load the target bulk cargo into the target ship based on the first material dropping reference point to determine the material pile of the target bulk cargo, and continuously monitor the front and rear inclination angle, the left and right inclination angle of the target ship corresponding to the target ship, and the vertical distance between the chute and the material pile; the second material dropping point determination module 1300 is configured to detect that the front and rear inclination angle or the left and right inclination angle is greater than a preset angle threshold, or the vertical distance is less than a preset safety distance, and update the first material dropping reference point according to the front and rear inclination angle, the left and right inclination angle, and the static offset correction amount of the first material dropping reference point by using a preset material dropping point correction model to determine the second material dropping reference point, wherein the static offset correction amount is determined by the engine room position of the target ship; the material dropping point adjustment module 1400 is configured to drive the cart traveling mechanism of the ship loader to move horizontally along the wharf ground and / or adjust the rotation angle of the cantilever of the ship loader until the chute of the ship loader is aligned with the second material dropping reference point, so as to continue to load the target bulk cargo into the target ship; the operation control optimization module 1500 is configured to repeat the above steps until the target ship is fully loaded, so as to complete the optimization of the ship loader operation control.

[0188] On the basis of any embodiment of the present application, referring to Figure 9 , another embodiment of the present application further provides an electronic device, which can be realized by a computer device, such as Figure 9As shown, the internal structure diagram of the computer device is shown. The computer device includes a processor, a computer readable storage medium, a memory and a network interface connected by a system bus. Among them, the computer readable storage medium of the computer device stores an operating system, a database and computer readable instructions, the database can store control information sequence, and the computer readable instructions are executed by the processor to enable the processor to realize a ship loader operation control optimization method. The processor of the computer device is used to provide computing and control capability to support the operation of the entire computer device. The memory of the computer device can store computer readable instructions, and the computer readable instructions are executed by the processor to enable the processor to execute the ship loader operation control optimization method of the present application. The network interface of the computer device is used to connect and communicate with the terminal. Those skilled in the art can understand, Figure 9 The structure shown in the above embodiment is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components, or combine certain components, or have a different component arrangement.

[0189] The processor in the embodiment is used to execute the specific functions of each module in Figure 8 The memory stores the program codes and various data required for executing the above-mentioned modules or sub-modules. The network interface is used for data transmission between the user terminal or the server. The memory in the embodiment stores the program codes and data required for executing all modules in the ship loader operation control optimization device of the present application, and the server can call the program codes and data of the server to execute the functions of all modules.

[0190] The present application also provides a storage medium storing computer readable instructions, which are executed by one or more processors to enable the one or more processors to execute the steps of the ship loader operation control optimization method described in any embodiment of the present application.

[0191] The present application also provides a computer program product including computer programs / instructions, which are executed by one or more processors to realize the steps of the ship loader operation control optimization method described in any embodiment of the present application.

[0192] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments of the application can be completed by a computer program instructing relevant hardware, and the computer program can be stored in a computer readable storage medium. When the program is executed, the processes of the above-mentioned embodiments of the methods can be included. The storage medium can be a computer readable storage medium such as a magnetic disc, an optical disc, a read-only memory (ROM), or a random access memory (RAM).

[0193] The above only describes some embodiments of the application. It should be pointed out that those skilled in the art can make some improvements and refinements without departing from the principles of the application, and these improvements and refinements should also be considered as the protection scope of the application.

Claims

1. A ship loader operation control optimization method characterized by, The method comprises the following steps: Step S10, collecting laser radar point cloud data corresponding to a target ship cabin of a target ship, and determining a cabin three-dimensional center coordinate corresponding to the target ship cabin according to the laser radar point cloud data, wherein a space coordinate system in which the cabin three-dimensional center coordinate is located is constructed with a direction along a wharf shoreline to a bow direction as a horizontal axis direction, a direction along a vertical wharf shoreline to a sea side direction as a vertical axis direction, and a direction upward perpendicular to a wharf ground as a vertical axis direction; Step S20, taking the cabin three-dimensional center coordinate as a first material falling reference point of a spout of a ship loader, loading target bulk cargo to the target ship based on the first material falling reference point to accumulate a material pile of the target bulk cargo, and continuously monitoring a front-rear inclination angle, a left-right inclination angle of the target ship, and a vertical distance between the spout and the material pile; Step S30, when the front-rear inclination angle or the left-right inclination angle is greater than a preset angle threshold, or the vertical distance is less than a preset safety distance, updating the first material falling reference point according to a preset material falling point correction model and a static offset correction amount of the first material falling reference point based on the front-rear inclination angle, the left-right inclination angle, and the static offset correction amount to determine a second material falling reference point, wherein the static offset correction amount is determined by a cabin position of the target ship; The step of updating the first material falling reference point according to the preset material falling point correction model and the static offset correction amount of the first material falling reference point based on the front-rear inclination angle, the left-right inclination angle, and the static offset correction amount to determine a second material falling reference point comprises: obtaining a horizontal axis coordinate and a vertical axis coordinate of the first material falling reference point, a horizontal axis static offset correction amount of the first material falling reference point, a ship depth of the target ship, a front-rear inclination angle of the target ship, a mass of the target bulk cargo, a total mass of the target ship, and an average vertical axis coordinate, wherein the ship depth represents a vertical distance between a cabin bottom and a deck of the target ship cabin, and the average vertical axis coordinate represents an average value among the vertical axis coordinate of the first material falling reference point, a vertical axis coordinate of a cabin center of the target ship, and a vertical axis coordinate of a material pile center; calculating a first product of a half of the ship depth and a sine value of the front-rear inclination angle, a first difference between the vertical axis coordinate of the first material falling reference point and the average vertical axis coordinate, a first ratio between the mass of the target bulk cargo and the total mass of the target ship, and a second product of the first ratio, the first difference, and the sine value of the front-rear inclination angle; determining a horizontal axis coordinate of the second material falling reference point according to a first sum of the horizontal axis coordinate of the first material falling reference point, the horizontal axis static offset correction amount of the first material falling reference point, the first product, and the second product; Step S40, driving a cart traveling mechanism of the ship loader to move horizontally along a wharf ground and / or adjusting a slewing angle of a cantilever of the ship loader until a spout of the ship loader is aligned with the second material falling reference point, so as to continue loading the target bulk cargo to the target ship. Step S50, repeating the above steps S10 to S40 until the target ship is completed loading to complete the optimization of the ship loader operation control.

2. The ship loader operation control optimization method according to claim 1, characterized in that, The step of updating the first blanking reference point according to the front and rear inclination angle, the left and right inclination angle and the static offset correction amount of the first blanking reference point to determine the second blanking reference point comprises: Obtaining the longitudinal axis coordinate of the first blanking reference point, the longitudinal axis static offset correction amount of the first blanking reference point, the left and right inclination angle of the target ship, and the longitudinal axis dynamic offset amount of the first blanking reference point; Calculating a third product between one half of the ship depth and the sine value of the left and right inclination angle; calculating a second ratio between the vertical axis coordinate of the first blanking reference point and the average vertical axis coordinate; and calculating a fourth product between the first ratio, the longitudinal axis dynamic offset amount of the first blanking reference point, the second ratio and the sine value of the left and right inclination angle; Determining the longitudinal axis coordinate of the second blanking reference point according to the second sum value between the longitudinal axis coordinate of the first blanking reference point, the longitudinal axis static offset correction amount of the first blanking reference point, the third product and the fourth product.

3. The shipper operation control optimization method according to any one of claims 1 to 2, characterized in that, The step of updating the first blanking reference point according to the front and rear inclination angle, the left and right inclination angle and the static offset correction amount of the first blanking reference point to determine the second blanking reference point comprises: Obtaining the draft of the target ship; Calculating a second difference between the draft of the target ship and one half of the ship depth; and calculating a fifth product between the second difference, the cosine value of the front and rear inclination angle and the cosine value of the left and right inclination angle; Determining the vertical axis coordinate of the second blanking reference point according to the third sum value between the vertical axis coordinate of the first blanking reference point and the fifth product; According to the horizontal axis coordinate, the longitudinal axis coordinate and the vertical axis coordinate of the second blanking reference point, the second blanking reference point is constructed.

4. The ship loader operation control optimization method of claim 1, wherein, The step of determining the first blanking reference point comprises: Based on the laser scanning device mounted on the cantilever of the ship loader, laser radar point cloud data corresponding to the target ship cabin of the target ship is collected; The laser radar point cloud data is down-sampled by using the voxel center down-sampling method; based on the down-sampled laser radar point cloud data, the average distance of each point and a plurality of adjacent points is calculated by using a preset statistical filtering algorithm; the average value and the standard deviation of all average distances are calculated; the distance threshold value is calculated according to the average value and the standard deviation; it is detected whether the average distance is greater than the distance threshold value; if it is greater, the point is removed as an outlier; The laser radar point cloud data after statistical filtering processing is fitted by using the RANSAC algorithm to extract the cabin wall surface point cloud data corresponding to the target ship cabin; the cabin length, the cabin width, the ship depth and the vertical axis coordinate of the cabin bottom in the vertical axis direction of the target ship cabin are determined according to the cabin wall surface point cloud data. determining the horizontal axis coordinate in the cabin three-dimensional center coordinates according to a fourth sum value between the starting coordinate of the target cabin in the horizontal axis direction and half of the cabin length, determining the vertical axis coordinate in the cabin three-dimensional center coordinates according to a fifth sum value between the starting coordinate of the target cabin in the vertical axis direction and half of the cabin width, and determining the vertical axis coordinate in the cabin three-dimensional center coordinates according to a sixth sum value between the vertical axis coordinate of the cabin bottom in the vertical axis direction and half of the ship depth; determining the first blanking reference point according to the horizontal axis coordinate, the vertical axis coordinate and the vertical axis coordinate in the cabin three-dimensional center coordinates.

5. The ship loader operation control optimization method of claim 1, wherein, The steps of driving the walking mechanism of the ship loader to move horizontally along the wharf ground and / or adjusting the rotation angle of the cantilever of the ship loader until the chute of the ship loader is aligned with the second blanking reference point, comprising: obtaining a moving area RGB image containing a target obstacle in the horizontal movement of the walking mechanism along the wharf ground and a corresponding moving area depth image; calling an obstacle detection model trained to a convergent state to perform target detection on the moving area RGB image to determine the two-dimensional center pixel coordinates corresponding to the target obstacle, wherein the basic network architecture of the obstacle detection model is a YOLOv7-tiny model, and the target obstacle includes a working personnel or scattered goods; aligning the moving area RGB image and the moving area depth image, matching the depth value in the moving area depth image corresponding to the two-dimensional center pixel coordinates of the target obstacle in the moving area RGB image to determine the three-dimensional center coordinates of the target obstacle in the real space; if the Euclidean distance between the three-dimensional center coordinates of the target obstacle and the three-dimensional center coordinates of the walking mechanism is less than a shutdown distance threshold, sending a shutdown instruction to the ship loader to force the walking mechanism to stop moving; if the Euclidean distance between the three-dimensional center coordinates of the target obstacle and the three-dimensional center coordinates of the walking mechanism is between the shutdown distance threshold and a warning distance threshold, issuing an audible and visual alarm to remind the working personnel to avoid.

6. The ship loader operation control optimization method of claim 1, wherein, The front and rear inclination angles represent the included angle between the horizontal axis of the ship body pointing to the bow direction along the wharf line and the horizontal reference plane parallel to the wharf ground when the target ship rotates around the vertical axis pointing to the sea side direction along the wharf line; The left and right inclination angles represent the included angle between the vertical axis of the ship body pointing to the sea side direction perpendicular to the wharf line and the horizontal reference plane parallel to the wharf ground when the target ship rotates around the horizontal axis pointing to the bow direction along the wharf line; The target bulk cargo includes coal, ore or grain.

7. A ship loader operation control optimization device, characterized in that, The method comprises: A ship cabin center determining module is configured to collect laser radar point cloud data corresponding to a target ship cabin of a target ship, and determine a three-dimensional center coordinate of the target ship cabin according to the laser radar point cloud data, wherein a space coordinate system in which the three-dimensional center coordinate is located is constructed with a direction along a wharf shoreline to a bow direction as a horizontal axis direction, a direction along a vertical wharf shoreline to a sea side direction as a vertical axis direction, and a direction perpendicular to a wharf ground upward as a vertical axis direction; A first material falling point determining module is configured to take the three-dimensional center coordinate as a first material falling reference point of a spout drum in a ship loader, load target bulk cargo to the target ship based on the first material falling reference point to accumulate a material pile of the target bulk cargo, and continuously monitor a front-rear inclination angle, a left-right inclination angle of the target ship, and a vertical distance between the spout drum and the material pile; A second material falling point determining module is configured to, when detecting that the front-rear inclination angle or the left-right inclination angle is greater than a preset angle threshold, or the vertical distance is less than a preset safety distance, update the first material falling reference point according to a preset material falling point correction model, a front-rear inclination angle, a left-right inclination angle, and a static offset correction amount of the first material falling reference point to determine a second material falling reference point, wherein the static offset correction amount is determined by a cabin position of the target ship; The step of updating the first material falling reference point according to the preset material falling point correction model, the front-rear inclination angle, the left-right inclination angle, and the static offset correction amount of the first material falling reference point to determine the second material falling reference point includes: obtaining a horizontal axis coordinate and a vertical axis coordinate of the first material falling reference point, a horizontal axis static offset correction amount of the first material falling reference point, a ship depth of the target ship, a front-rear inclination angle of the target ship, a mass of the target bulk cargo, a total mass of the target ship, and an average vertical axis coordinate, wherein the ship depth represents a vertical distance between a cabin bottom and a deck of the target ship cabin, and the average vertical axis coordinate represents an average value among the vertical axis coordinate of the first material falling reference point, a vertical axis coordinate of a cabin center of the target ship, and a vertical axis coordinate of a material pile center; calculating a first product of a half of the ship depth and a sine value of the front-rear inclination angle, a first difference between the vertical axis coordinate of the first material falling reference point and the average vertical axis coordinate, a first ratio between the mass of the target bulk cargo and the total mass of the target ship, and a second product among the first ratio, the first difference, and the sine value of the front-rear inclination angle; determining a horizontal axis coordinate of the second material falling reference point according to a first sum among the horizontal axis coordinate of the first material falling reference point, the horizontal axis static offset correction amount of the first material falling reference point, the first product, and the second product; and The blanking point adjusting module is configured to drive a travelling mechanism of the ship loader to move horizontally along the ground of the wharf and / or adjust a rotation angle of a cantilever of the ship loader until a chute of the ship loader is aligned with the second blanking reference point, so as to continue loading the target bulk cargo to the target ship. The operation control optimizing module is configured to repeat the above steps until the target ship is completed with loading, so as to complete the optimization of the ship loader operation control.

8. An electronic device comprising a central processing unit and a memory, characterized in that The central processing unit is configured to call and run a computer program stored in the memory to perform the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer program is stored in the form of computer readable instructions and is implemented according to the method of any one of claims 1 to 6. When the computer program is called and run by a computer, the steps included in the corresponding method are performed.

Images