A crane pipe automatic positioning method and automatic positioning system

Abstract

The present invention discloses an automatic positioning method for a crane pipe, comprising acquiring an image of a filling area; analyzing the image of the filling area to obtain the number of filling vehicles in the filling area and an image of a current parking space; judging whether the number of filling vehicles is the same as a preset number based on the number of filling vehicles in the filling area, and judging whether each filling vehicle completely overlaps with the corresponding parking space based on the image of the current parking space; if any filling vehicle does not completely overlap with the corresponding parking space, and the number of filling vehicles is the same as the preset number, acquiring the actual coordinates of the filling port based on the filling port image, and generating a third path based on the actual coordinates of the filling port; if any filling vehicle does not completely overlap with the corresponding parking space, and the number of filling vehicles is different from the preset number, acquiring the actual coordinates of the filling port based on the filling port image, and generating a fourth path based on the actual coordinates of the filling port and the number of filling vehicles, and driving the crane pipe to fill each filling vehicle according to the fourth path. The present application also discloses an automatic positioning system for a crane pipe.

Description

Technical Field

[0001] The present invention relates to the technical field of crane pipe positioning, especially a method and automatic positioning system for crane pipes. Background Art

[0002] The automatic crane pipe positioning system is mainly composed of a programmable logic controller (PLC) controller, a gas phase curved arm, a liquid phase curved arm, a truss track, a motor, a truss trolley, a laser alignment sensor, a telescopic tube, a gas collecting cap, a liquid level and static electricity detection alarm, etc.

[0003] The automatic positioning truss crane pipe system sets up a truss track directly above the vehicle that needs to be filled, installs a mobile trolley and a crane pipe, uses a motor as a power source to drive the trolley and the crane pipe to move back and forth horizontally along the track, uses laser detection to perform non-contact measurement of the tank surface of the train tank car, and guides the crane pipe to automatically move to the center of the tank mouth through precise calculations by the PLC controller, thereby realizing automatic alignment of the crane pipe and the tank car. After accurately finding the tank mouth, the motor drives the steel wire rope to pull the internal telescopic tube of the crane pipe down to the bottom of the tank car, and the sealing cap is lowered to the tank mouth with the telescopic tube and compacted. After starting the pump, the interlock controls the electric valve to open the valve to allow oil to be filled. When the filling amount is reached, the interlock closes the electric valve and pulls the telescopic tube up to the tank mouth. After controlling the oil, it returns to its full position to complete the filling operation.

[0004] However, the automatic positioning truss crane system in the prior art only locates and identifies one vehicle that needs to be filled. If there are multiple vehicles that need to be filled, the current automatic positioning truss crane system does not perform path planning, resulting in low filling efficiency. Summary of the invention

[0005] In order to solve the problems existing in the prior art, the present invention provides a method of automatic positioning of crane tubes and an automatic positioning system.

[0006] The technical solution adopted by the present invention is: a method of automatic positioning of crane tubes, including the following contents:

[0007] Acquire an image of the filling area; wherein a plurality of preset parking spaces are provided in the filling area, a plurality of filling vehicles are parked in the filling area, the parking spaces are used for parking the filling vehicles, and the size of the parking spaces is the same as the size of the filling vehicles;

[0008] The image of the infusion area is analyzed to obtain the number of filling vehicles in the infusion area and the current parking space image; wherein the current parking space image is a parking space image with the filling vehicle parked;

[0009] Based on the number of filling vehicles in the filling area, determining whether the number of filling vehicles is the same as a preset number, and based on the current parking space image, determining whether each filling vehicle completely overlaps with the corresponding parking space;

[0010] If each filling truck is completely coincident with the corresponding parking space, and the number of filling trucks is the same as the preset number, the driving crane pipe will infuse each filling truck according to the preset first path;

[0011] If each filling vehicle completely overlaps with the corresponding parking space, and the number of filling vehicles is different from the preset number, a second path is generated based on the actual number of filling vehicles, and the crane is driven to fill each filling vehicle according to the second path;

[0012] If there is any filling vehicle that does not completely overlap with the corresponding parking space, and the number of filling vehicles is the same as the preset number, then identify the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space; obtain the actual coordinates of the filling port based on the filling port image, generate a third path based on the actual coordinates of the filling port, and drive the crane to fill each filling vehicle according to the third path;

[0013] If there is any filling vehicle that does not completely overlap with the corresponding parking space, and the number of filling vehicles is different from the preset number, the filling mouth image of the filling vehicle that does not completely overlap with the corresponding parking space is identified; the actual coordinates of the filling mouth are obtained based on the filling mouth image, and a fourth path is generated based on the actual coordinates of the filling mouth and the number of filling vehicles, and the crane is driven to fill each filling vehicle according to the fourth path.

[0014] Preferably, the perfusion area is a square area, and the parking space array is distributed within the perfusion area.

[0015] Preferably, based on the current parking space image, determining whether each filling vehicle and the corresponding parking space is completely coincident includes the following content:

[0016] Extract the edge contour of the current parking space image;

[0017] The edge contour of the current parking space image is compared with the edge contour of the preset parking space. If the edge contour of the current parking space completely overlaps with the edge contour of the preset parking space, each filling vehicle completely overlaps with the corresponding parking space.

[0018] Preferably, if each filling vehicle completely overlaps with the corresponding parking space, and the number of the filling vehicles is the same as the preset number, driving the crane to fill each filling vehicle according to the preset first path includes:

[0019] Obtain the parking directions of all filling vehicles. If all filling vehicles are in the first parking direction and the number of filling vehicles is the same as the preset number, drive the crane to fill each filling vehicle according to the preset first path;

[0020] If all filling trucks are in the first parking direction and the number of filling trucks is the same as the preset number, the driving crane pipe will infuse each filling truck according to the preset fifth path.

[0021] Preferably, acquiring the actual coordinates of the filling port based on the filling port image, generating a third path based on the actual coordinates of the filling port, and driving the crane to fill each filling vehicle according to the third path includes the following:

[0022] Based on the inlet image, the actual coordinates of the inlet mouth are obtained.

[0023] Comparing the actual coordinates of the silo mouth with the preset silo mouth standard coordinates, obtaining the offset coordinates of the silo mouth.

[0024] Based on the offset coordinates, determine whether the offset direction is on the first path; if the offset direction is on the first path, update the path points on the first path based on the actual coordinates of the filling port, and drive the crane to fill each filling vehicle according to the updated first path;

[0025] If the offset direction is not on the first path, a third path is generated based on the actual coordinates of the filling port, and the driving crane pipe is used to infuse each filling truck according to the third path.

[0026] Preferably, identifying the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space; acquiring the actual coordinates of the filling port based on the filling port image, generating a fourth path based on the actual coordinates of the filling port and the number of filling vehicles, and driving the crane to fill each filling vehicle according to the fourth path includes:

[0027] Analyze the images of the infusion area and identify filling trucks that do not exactly coincide with the corresponding parking space;

[0028] For each filling truck that does not exactly coincide, an image of its filling area is obtained;

[0029] Process the acquired stool image and extract the actual coordinates of the stool;

[0030] Set up a work area map to indicate all the filling port coordinates and the moving range of the crane; the filling port coordinates include the actual filling port coordinates of each incompletely overlapping filling vehicle and the standard filling port coordinates of the completely overlapping filling vehicle;

[0031] Create an empty path list to store the movement instructions of the crane tube;

[0032] According to the actual situation of the filling truck, a priority is assigned to the filling truck;

[0033] Using a path optimization algorithm, the fourth path is calculated based on the stool coordinates and the priority of the filling truck;

[0034] The drive crane pipe fills each filling truck according to the fourth path.

[0035] Preferably, the image of the acquired perfusion region comprises the following:

[0036] Assess the size and shape of the perfusion area, determine the number and position of the cameras, and overlaps in the acquisition areas between adjacent cameras;

[0037] Control the camera synchronization trigger to obtain the synchronous image taken;

[0038] The images taken synchronously are preprocessed and then stitched;

[0039] The stitched image is smoothed to obtain an image of the perfusion area.

[0040] A second aspect of the present application provides a crane tube automatic positioning system, including:

[0041] An image acquisition unit, the image acquisition unit is used to acquire an image of the perfusion area; wherein a plurality of preset parking spaces are provided in the perfusion area, a plurality of filling vehicles are parked in the filling area, the parking spaces are used for parking the filling vehicles, and the size of the parking spaces is the same as the size of the filling vehicles;

[0042] An analysis unit, the analysis unit is used to analyze the image of the perfusion area to obtain the number of filling vehicles in the perfusion area and the current parking space image; wherein the current parking space image is an image of a parking space where a filling vehicle is parked;

[0043] a judgment unit, the judgment unit being used to judge whether the number of filling vehicles is the same as a preset number based on the number of filling vehicles in the filling area, and the judgment unit being used to judge whether each filling vehicle completely overlaps with the corresponding parking space based on the current parking space image;

[0044] A control unit, wherein the control unit is used to drive the crane to fill each filling vehicle along a preset first path when each filling vehicle completely overlaps with the corresponding parking space and the number of filling vehicles is the same as a preset number;

[0045] The control unit is also used to generate a second path based on the actual number of filling vehicles when each filling vehicle completely overlaps with the corresponding parking space and the number of filling vehicles is different from the preset number, and drive the crane to fill each filling vehicle according to the second path;

[0046] The control unit is also used to identify the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space when any filling vehicle does not completely overlap with the corresponding parking space and the number of filling vehicles is the same as the preset number; obtain the actual coordinates of the filling port based on the filling port image, generate a third path based on the actual coordinates of the filling port, and drive the crane to fill each filling vehicle according to the third path;

[0047] The control unit is also used to identify the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space if there is any filling vehicle that does not completely overlap with the corresponding parking space and the number of filling vehicles is different from the preset number; obtain the actual coordinates of the filling port based on the filling port image, generate a fourth path based on the actual coordinates of the filling port and the number of filling vehicles, and drive the crane pipe to fill each filling vehicle according to the fourth path.

[0048] The beneficial effects of the present invention are: comprehensive acquisition and analysis of real-time images of the filling area. It can not only accurately determine the number and position of the filling vehicles, but also intelligently plan the optimal filling path according to different situations. When all the filling vehicles completely overlap with the parking spaces and the number is consistent with the preset, the system directly performs efficient filling according to the preset first path; even if there is a discrepancy in the number or position deviation, the system can quickly identify and adjust, and generate a new filling path based on the actual situation. This intelligent solution greatly improves the filling efficiency when multiple filling vehicles are operating at the same time, reduces unnecessary waiting and moving time, and effectively optimizes resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a flow chart of a method according to Embodiment 1 of the present invention;

[0050] Figure 2 This is a system block diagram of Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0051] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0052] Embodiment 1

[0053] like Figure 1 As shown, a method of automatic positioning of crane tubes includes the following contents:

[0054] Step 1: Acquire an image of the perfused area.

[0055] Among them, a plurality of preset parking spaces are provided in the filling area, and a plurality of filling vehicles are parked in the filling area. The parking space is used for filling vehicle parking, and the size of the parking space is the same as the size of the filling vehicle.

[0056] It should be noted that the perfusion area is a square area, and the parking spaces in the perfusion area are distributed in an array. The array distribution can ensure that the parking spaces are arranged evenly and orderly in the perfusion area, and the parking spaces distributed in an array can more easily track and identify the status of each parking space. Through the design of parking spaces distributed in an array, the standardization and modularization of parking spaces can be achieved. The square area itself has regular boundaries and shapes, so the parking spaces distributed in an array can better adapt to this shape. By arranging the parking spaces in an array in the square area, it can be ensured that each parking space has a similar size and position relationship.

[0057] Among them, the square pouring area is built with horizontal and vertical crane pipe mobile racks and industrial camera mobile racks in the air. Through the horizontal and vertical mobile racks, the crane pipes and industrial cameras can move in the X and Y directions. The movement methods all adopt existing technical means, such as moving through the coordination of truss tracks, motors, truss trolleys, etc.

[0058] Reference, the image of the acquired perfusion area includes the following:

[0059] Evaluate the size and shape of the perfusion area, determine the number and position of cameras, and ensure that there is overlap between the acquisition areas of adjacent cameras. Specifically, ensure that each area in the perfusion area is covered by at least one camera and that there is an appropriate overlap between the cameras. Select a high-definition camera and choose whether the infrared function is required based on the on-site environment. Install the camera and adjust the angle and focal length to ensure that the image is clear and covers the perfusion area. There is an appropriate overlap between the cameras, which means that when multiple cameras monitor an area, in order to ensure that there is no blind spot in the monitoring and facilitate subsequent image stitching, there needs to be a certain overlap between the images taken by adjacent cameras. When images from multiple cameras are stitched together, the overlapping area can be used as a benchmark for image registration and fusion, making it easier to merge into a complete image. The size of the overlapping area also needs to be determined based on the specific situation. If the overlapping area is too small, it may result in blind spots in monitoring or inaccurate image stitching; if the overlapping area is too large, it may increase unnecessary redundant information and calculations. Generally speaking, the optimal overlapping area size can be determined through experiments or simulations.

[0060] Control the synchronous triggering of the cameras to obtain synchronously captured images. Specifically, set all cameras to acquire images at the same frame rate. Ensure that the cameras are triggered synchronously to obtain images captured almost simultaneously. Adjust camera parameters such as exposure time, white balance, etc. to obtain the best image quality. If the cameras are not triggered synchronously, the judgment of the vehicle position and number may be inaccurate due to the time difference. Synchronous triggering can minimize this time difference, thereby reducing the error caused by time asynchrony.

[0061] The synchronously captured images are preprocessed and then stitched. Specifically, the original images are denoised and enhanced using image preprocessing techniques (such as Gaussian filtering, histogram equalization, etc.). Feature points are extracted from each image using feature point extraction algorithms such as SIFT and SURF. Feature point matching is performed using algorithms such as RANSAC and FLANN to determine the correspondence between images. Images are aligned to the same coordinate system using perspective transformation or affine transformation. Multiple images are merged into a complete perfusion area image using weighted averaging or multi-band fusion methods. The stitched images are smoothed to obtain an image of the perfusion area. The stitched images are smoothed to eliminate seams and noise. Color correction technology can also be used to ensure color consistency across the entire image.

[0062] Step 2: Analyze the image of the infusion area to obtain the number of filling vehicles in the infusion area and the current parking space image.

[0063] The current parking space image is an image of a parking space where a tanker truck is parked. Specifically, an edge detection algorithm (such as a Canny edge detector) is used to extract the contour of the tanker truck and the boundary information of the parking space in the image.

[0064] Features of the tanker trucks are extracted from the images of the perfusion area, and image recognition algorithms (such as machine learning or deep learning models) are used to identify the tanker trucks in the images. These models can be trained on large amounts of labeled image data to learn how to identify tanker trucks and count the number of tanker trucks identified.

[0065] By using image segmentation or region growing algorithms, combined with known features of parking spaces (such as lines, marks, etc.), the parking space area in the image is located, and in the located parking space area, the parking space with the filling vehicle is selected, that is, the current parking space image. This can be achieved by comparing whether there are filling vehicles in the parking space area.

[0066] Step three: based on the number of filling vehicles in the filling area, determine whether the number of filling vehicles is the same as a preset number, and based on the current parking space image, determine whether each filling vehicle completely overlaps with the corresponding parking space.

[0067] Specifically, first, the number of filling vehicles preset in the filling area needs to be determined. This number can be determined based on factors such as the size of the filling area, production capacity, and work plan. For example, assuming that the preset number of filling vehicles is N, the number of filling vehicles counted in step 2 is compared with the preset number N. Determine whether the number of filling vehicles is the same as the preset number.

[0068] To determine whether each filling truck completely overlaps with the corresponding parking space, it is necessary to extract the edge contour of the current parking space image; compare the edge contour of the current parking space image with the edge contour of the preset parking space. If the edge contour of the current parking space completely overlaps with the edge contour of the preset parking space, then each filling truck completely overlaps with the corresponding parking space. Specifically, first, there is a preset parking space edge contour template, which is obtained when there is no filling truck parked in the parking space. This template should accurately reflect the boundary and shape of the parking space. When the filling truck is parked in the parking space, image processing technology (such as edge detection algorithm) is used to extract the edge contour of the current parking space image containing the filling truck. This contour will include the boundary of the filling truck and the part that may overlap with the parking space. The edge contour of the front parking space is completely consistent with the edge contour of the preset parking space. This filling truck is completely parked in the parking space, that is, the two completely overlap.

[0069] If each filling vehicle completely overlaps with the corresponding parking space, and the number of filling vehicles is the same as the preset number, then execute step 4. If each filling vehicle completely overlaps with the corresponding parking space, and the number of filling vehicles is different from the preset number, then execute step 5. If any filling vehicle does not completely overlap with the corresponding parking space, and the number of filling vehicles is the same as the preset number, then execute step 6. If any filling vehicle does not completely overlap with the corresponding parking space, and the number of filling vehicles is different from the preset number, then execute step 7.

[0070] Step 4: Drive the crane pipe to infuse each filling truck according to the preset first path.

[0071] If each filling truck completely overlaps with the corresponding parking space, and the number of filling trucks is the same as the preset number, it means that the filling truck is parked at the preset position, the filling port is also at the preset position, and the number of filling ports is also the preset number. At this time, the crane can be directly driven to fill each filling truck according to the preset first path, without wasting time on planning the path.

[0072] For example, a filling area has six parking spaces preset, each of which corresponds to a filling vehicle, and each filling vehicle has a preset filling port position. When five filling vehicles are accurately parked in their respective parking spaces and their filling ports are all located at the preset positions, the system will recognize this state and immediately start the crane to perform filling according to the preset first path.

[0073] Among them, the preset first path aims to reduce the time of crane movement. For example, there are 6 preset parking spaces, which are arranged in an array in the square filling area, 2 rows and 3 columns or 3 columns and 2 rows, numbered P1 to P6, and each parking space corresponds to a filling truck.

[0074] The starting position of the crane tube is set to S and the end position is set to E (maybe the parking or recycling position of the crane tube).

[0075] The filling port of each filling vehicle has fixed coordinates on the parking space, such as the filling port coordinates of P1 are (x1, y1), the filling port coordinates of P2 are (x2, y2), and so on. Calculate the distance of each segment of the preliminary path (such as S to P1, P1 to P2, etc.), which is usually based on the distance formula between two points in the coordinate system (such as Euclidean distance). Add the distances of all segments to get the total length of the preliminary path.

[0076] The preset first path is S→P1(x1,y1)→P2(x2,y2)→P3(x3,y3)→P4(x4,y4)→P5(x5,y5)→P6(x6,y6)→E, where the arrow indicates the direction of the movement of the crane tube.

[0077] Furthermore, considering that the parking direction of the filling vehicle may be reversed, each filling vehicle completely overlaps with the corresponding parking space. For example, if each filling vehicle completely overlaps with the corresponding parking space, and the number of filling vehicles is the same as the preset number, driving the crane to fill each filling vehicle according to the preset first path includes:

[0078] Obtain the parking direction of all filling trucks. If all filling trucks are in the first parking direction and the number of filling trucks is the same as the preset number, drive the crane to fill each filling truck according to the preset first path; if all filling trucks are in the first parking direction and the number of filling trucks is the same as the preset number, drive the crane to fill each filling truck according to the preset fifth path.

[0079] For example, OK, we can use coordinates to more accurately describe the preset fifth path. Assume that the filling area is a rectangular area, assume that the first parking direction is with the front of the vehicle facing S, and the coordinates of each filling port are as follows (with the lower left corner as the origin, the right is the positive direction of the x-axis, and the upward direction is the positive direction of the y-axis): P1 (x1, y1), P2 (x2, y1), P3 (x3, y1), P4 (x1, y2), P5 (x2, y2), P6 (x3, y2). When all filling vehicles are parked in the opposite direction, that is, when the front of the vehicle faces E, the position of the filling port will move to the other side of the vehicle. We assume that the relative position of each filling port on the vehicle is fixed, but the direction is opposite. In this case, we need to determine the new coordinates of each filling port. If the width of the vehicle is w, and we assume that the outlets are horizontally symmetrical on the side of the vehicle facing E and the side facing S, then the new coordinates can be calculated as follows: New coordinates of P1: (x1+w, y1) New coordinates of P2: (x2+w, y1) New coordinates of P3: (x3+w, y1) New coordinates of P4: (x1+w, y2) New coordinates of P5: (x2+w, y2), New coordinates of P6: (x3+w, y2), Now, it is necessary to preset the fifth path so that the crane can traverse these new outlet coordinates in order from S and finally reach E. Since we want to minimize the moving distance of the crane, a possible strategy is to traverse the farthest outlet first (P6 in this example), and then gradually approach the starting position S. Therefore, the fifth path may be as follows: S→(x3+w,y2) (i.e., the new coordinates of P6)→(x2+w,y2) (i.e., the new coordinates of P5)→(x1+w,y2) (i.e., the new coordinates of P4)→(x3+w,y1) (i.e., the new coordinates of P3)→(x2+w,y1) (i.e., the new coordinates of P2)→(x1+w,y1) (i.e., the new coordinates of P1)→E.

[0080] Since the first path and the fifth path are preset based on the parking status, quantity and position of the filling vehicles, once these two common states are identified through image monitoring, the crane can be directly driven to position and pour according to the corresponding path, saving time in planning the path.

[0081] Step 5: Generate a second path based on the actual number of filling trucks, and drive the crane pipe to infuse each filling truck according to the second path.

[0082] When each filling vehicle completely overlaps with the corresponding parking space and the number of filling vehicles is different from the preset number, a second path is generated based on the actual number of filling vehicles, and the crane is driven to fill each filling vehicle according to the second path.

[0083] When each filling truck completely overlaps with the corresponding parking space and the number of filling trucks is different from the preset number, the distance between two adjacent filling trucks may change. At this time, it is necessary to replan the path to form a second path, and position and fill each filling truck according to the second path.

[0084] Specifically, assume that the number of actual parked filling trucks is less than the preset number, for example, there are only 4 trucks, parked at positions P1, P2, P4, and P5. Determine the actual filling port coordinates: P1 (x1, y1), P2 (x2, y1), P4 (x1, y2), P5 (x2, y2), and calculate the distance between each pair of filling ports to determine the access order. For example, calculate the distance from P1 to P2, P1 to P4, P2 to P4, and so on. You can choose to use a greedy algorithm, a traveling salesman problem (TSP) algorithm, etc., starting from the starting position of the crane (assuming it is S), and add each filling port coordinate to the path in turn according to the calculated access order. Ensure that the path is continuous and that each filling port is visited only once. The end point of the path can be the starting position S or a preset end position E. The path can be further optimized using a local search algorithm (such as 2-opt, 3-opt, etc.). The goal of optimization is to reduce the moving distance or movement time of the crane tube and output the planned second path for use by the crane tube control system.

[0085] Step six, identifying the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space; obtaining the actual coordinates of the filling port based on the filling port image, generating a third path based on the actual coordinates of the filling port, and driving the crane pipe to fill each filling vehicle according to the third path.

[0086] If there is any filling vehicle that does not completely overlap with the corresponding parking space, and the number of filling vehicles is the same as the preset number, the filling mouth image of the filling vehicle that does not completely overlap with the corresponding parking space is identified; the actual coordinates of the filling mouth are obtained based on the filling mouth image, and a third path is generated based on the actual coordinates of the filling mouth, and the crane pipe is driven to fill each filling vehicle according to the third path.

[0087] Specifically, obtaining the actual coordinates of the filling port based on the filling port image, generating a third path based on the actual coordinates of the filling port, and driving the crane to fill each filling vehicle according to the third path includes the following:

[0088] The actual coordinates of the Guankou are obtained based on the Guankou image. For reference, first, the system uses a camera or similar visual device to capture an image of the Guankou of the filling vehicle that does not completely overlap with the corresponding parking space. Through image processing technology (such as edge detection, feature extraction, etc.), the system can identify the position of the Guankou in the image. Then, the system converts the position of the Guankou in the image into coordinates in the actual world coordinate system, that is, the actual coordinates of the Guankou, through visual positioning technology (such as camera calibration, image coordinate conversion, etc.).

[0089] Based on the offset coordinates, it is determined whether the offset direction is on the first path. If the offset direction is on the first path, the path points on the first path are updated based on the actual coordinates of the filling port, and the crane pipe is driven to fill each filling vehicle according to the updated first path. For reference, the actual coordinates of the filling port are compared with the preset standard coordinates of the filling port to obtain the offset coordinates of the filling port. For reference, the system pre-stores the standard coordinates of each filling port (that is, the coordinates of the filling port if the filling vehicle is completely parked in the corresponding parking space). The system will compare the actual coordinates of the filling port with the corresponding preset standard coordinates, calculate the offset coordinates of the filling port (that is, the difference between the actual coordinates and the standard coordinates), and the system analyzes the direction of the filling port offset to determine whether it is on the originally planned first path. This usually involves calculating the relative position relationship between the offset coordinates and the path points on the first path. If the offset direction is on the first path, it means that the crane pipe can still reach the filling port when moving according to the first path. At this time, the system will update the relevant path points on the first path according to the actual coordinates of the filling port to ensure that the crane pipe can accurately align with the filling port.

[0090] For example, if the port is offset 10 cm to the left, and there is a waypoint on the first path that is also to the left of the port, the system will move that waypoint 10 cm to the left to match the actual location of the port.

[0091] If the offset direction is not on the first path, a third path is generated based on the actual coordinates of the filling port, and the crane is driven to fill each filling vehicle according to the third path. If the offset direction is not on the first path, it means that a new path needs to be replanned. The system will calculate an optimal third path based on the actual coordinates of all filling ports using a path planning algorithm (such as the Dijkstra algorithm, etc.). The third path ensures that the crane can visit all filling ports in sequence and minimizes the crane's moving distance and number of turns.

[0092] Finally, the system sends the generated third path to the crane control system. The crane control system drives the crane to move according to the third path and fills each filling vehicle in turn. During the filling process, the system uses displacement sensors or other devices to monitor the progress and accuracy of the filling to ensure that each filling vehicle receives the correct amount of filling.

[0093] Step seven, identifying the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space; obtaining the actual coordinates of the filling port based on the filling port image, generating a fourth path based on the actual coordinates of the filling port and the number of filling vehicles, and driving the crane to fill each filling vehicle according to the fourth path.

[0094] If any filling truck does not exactly coincide with the corresponding parking space, and the number of filling trucks is different from the preset number, then step seven is performed.

[0095] Among them, identifying the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space; acquiring the actual coordinates of the filling port based on the filling port image, generating a fourth path based on the actual coordinates of the filling port and the number of filling vehicles, and driving the crane to fill each filling vehicle according to the fourth path includes:

[0096] Analyze the image of the filling area and identify the filling vehicles that do not completely overlap with the corresponding parking spaces. For example, the system captures the image of the entire filling area through a high-definition camera. Using image processing techniques such as edge detection and shape matching, the system is able to identify the filling vehicles that do not completely overlap with the corresponding parking spaces.

[0097] For each incompletely overlapping filling vehicle, obtain an image of its filling port area, process the obtained filling port image, and extract the actual filling port coordinates. For example: for each identified incompletely overlapping filling vehicle, further focus on its filling port area and obtain a high-definition image of the filling port. Then, through image recognition technology (such as template matching, feature point detection, etc.), the system can accurately extract the actual coordinates of the filling port.

[0098] Set up a work area map, indicating all the filling port coordinates and the moving range of the crane; the filling port coordinates include the actual filling port coordinates of each incompletely overlapping filling vehicle and the standard filling port coordinates of the completely overlapping filling vehicle. For example, set up a work area map that contains the location information of all filling ports. For the completely overlapping filling vehicles, use their preset filling port standard coordinates; for the incompletely overlapping filling vehicles, use the actual filling port coordinates just extracted. In addition, the work area map must also indicate the moving range of the crane.

[0099] Create an empty path list that stores the movement instructions of the crane tube.

[0100] According to the actual situation of the filling vehicle, a priority is assigned to the filling vehicle. For example, based on the actual situation (such as the waiting time of the filling vehicle, order priority, etc.), the system assigns a priority to each filling vehicle. This helps to give priority to high-priority filling vehicles during route planning.

[0101] Using a path optimization algorithm, a fourth path is calculated based on the filling port coordinates and the priority of the filling vehicle. For example, a path optimization algorithm (such as a genetic algorithm, a particle swarm optimization algorithm, etc.) is used to calculate the fourth path of the crane. The algorithm takes into account factors such as the coordinates of all filling ports, the movement range of the crane, the priority of the filling vehicle, and possible obstacles. The goal is to find a path that covers all filling ports while minimizing the distance and time of the crane movement.

[0102] Drive the crane to fill each filling vehicle according to the fourth path. Send the calculated fourth path to the crane control system. The crane control system drives the crane to move according to the path instruction and fills each filling vehicle in turn. During the filling process, the system will monitor the position and status of the crane in real time through sensors and other equipment to ensure the accuracy and safety of the filling.

[0103] Embodiment 2, as Figure 2 As shown, this embodiment provides a crane tube automatic positioning system, including:

[0104] An image acquisition unit, the image acquisition unit is used to acquire an image of the perfusion area; wherein a plurality of preset parking spaces are provided in the perfusion area, a plurality of filling vehicles are parked in the filling area, the parking spaces are used for parking the filling vehicles, and the size of the parking spaces is the same as the size of the filling vehicles;

[0105] An analysis unit, the analysis unit is used to analyze the image of the perfusion area to obtain the number of filling vehicles in the perfusion area and the current parking space image; wherein the current parking space image is an image of a parking space where a filling vehicle is parked;

[0106] a judgment unit, the judgment unit being used to judge whether the number of filling vehicles is the same as a preset number based on the number of filling vehicles in the filling area, and the judgment unit being used to judge whether each filling vehicle completely overlaps with the corresponding parking space based on the current parking space image;

[0107] A control unit, wherein the control unit is used to drive the crane to fill each filling vehicle along a preset first path when each filling vehicle completely overlaps with the corresponding parking space and the number of filling vehicles is the same as a preset number;

[0108] The control unit is also used to generate a second path based on the actual number of filling vehicles when each filling vehicle completely overlaps with the corresponding parking space and the number of filling vehicles is different from the preset number, and drive the crane to fill each filling vehicle according to the second path;

[0109] The control unit is also used to identify the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space when any filling vehicle does not completely overlap with the corresponding parking space and the number of filling vehicles is the same as the preset number; obtain the actual coordinates of the filling port based on the filling port image, generate a third path based on the actual coordinates of the filling port, and drive the crane to fill each filling vehicle according to the third path;

[0110] The control unit is also used to identify the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space if there is any filling vehicle that does not completely overlap with the corresponding parking space and the number of filling vehicles is different from the preset number; obtain the actual coordinates of the filling port based on the filling port image, generate a fourth path based on the actual coordinates of the filling port and the number of filling vehicles, and drive the crane pipe to fill each filling vehicle according to the fourth path.

[0111] The above-mentioned embodiments only express the specific implementation of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A crane pipe automatic positioning method, characterized in that: Includes the following: Acquire an image of the filling area; wherein a plurality of preset parking spaces are provided in the filling area, a plurality of filling vehicles are parked in the filling area, the parking spaces are used for parking the filling vehicles, and the size of the parking spaces is the same as the size of the filling vehicles; Analyze the image of the perfusion area to obtain the number of filling vehicles in the perfusion area and the current parking space image; wherein the current parking space image is an image of a parking space where a filling vehicle is parked; Based on the number of filling vehicles in the filling area, determining whether the number of filling vehicles is the same as a preset number, and based on the current parking space image, determining whether each filling vehicle completely overlaps with the corresponding parking space; If each filling vehicle completely overlaps with the corresponding parking space, and the number of filling vehicles is the same as the preset number, the crane is driven to fill each filling vehicle according to the preset first path; If each filling vehicle completely overlaps with the corresponding parking space, and the number of filling vehicles is different from the preset number, a second path is generated based on the actual number of filling vehicles, and the crane is driven to fill each filling vehicle according to the second path; If there is any filling vehicle that does not completely overlap with the corresponding parking space, and the number of filling vehicles is the same as the preset number, then identify the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space; obtain the actual coordinates of the filling port based on the filling port image, generate a third path based on the actual coordinates of the filling port, and drive the crane to fill each filling vehicle according to the third path; If there is any filling vehicle that does not completely overlap with the corresponding parking space, and the number of filling vehicles is different from the preset number, the filling mouth image of the filling vehicle that does not completely overlap with the corresponding parking space is identified; the actual coordinates of the filling mouth are obtained based on the filling mouth image, and a fourth path is generated based on the actual coordinates of the filling mouth and the number of filling vehicles, and the crane is driven to fill each filling vehicle according to the fourth path.

2. The method for automatically positioning a crane pipe according to claim 1, characterized in that: The perfusion area is a square area, and the parking spaces in the perfusion area are distributed in an array.

3. The method for automatically positioning a crane pipe according to claim 2, characterized in that: Based on the current parking space image, judging whether each filling vehicle completely overlaps with the corresponding parking space includes the following: Extract the edge contour of the current parking space image; The edge contour of the current parking space image is compared with the edge contour of the preset parking space. If the edge contour of the current parking space completely overlaps with the edge contour of the preset parking space, each filling vehicle completely overlaps with the corresponding parking space.

4. The method for automatically positioning a crane pipe according to claim 3, characterized in that: If each filling vehicle completely overlaps with the corresponding parking space, and the number of the filling vehicles is the same as the preset number, driving the crane to fill each filling vehicle according to the preset first path includes: Obtain the parking directions of all filling vehicles. If all filling vehicles are in the first parking direction and the number of filling vehicles is the same as the preset number, drive the crane to fill each filling vehicle according to the preset first path; If all the filling vehicles are in the first parking direction, and the number of the filling vehicles is the same as the preset number, the crane is driven to fill each filling vehicle along the preset fifth path.

5. The method for automatically positioning a crane pipe according to claim 3, characterized in that: Acquiring the actual coordinates of the filling port based on the filling port image, generating a third path based on the actual coordinates of the filling port, and driving the crane to fill each filling vehicle according to the third path includes the following: Acquire the actual coordinates of Guankou based on the Guankou image, Compare the actual coordinates of the filling port with the preset standard coordinates of the filling port to obtain the offset coordinates of the filling port, Based on the offset coordinates, determine whether the offset direction is on the first path; if the offset direction is on the first path, update the path points on the first path based on the actual coordinates of the filling port, and drive the crane to fill each filling vehicle according to the updated first path; If the offset direction is not on the first path, a third path is generated based on the actual coordinates of the filling port, and the lifting pipe is driven to fill each filling vehicle according to the third path.

6. The method for automatically positioning a crane pipe according to claim 3, characterized in that: Identifying a filling port image of a filling vehicle that does not completely overlap with a corresponding parking space; acquiring actual filling port coordinates based on the filling port image, generating a fourth path based on the actual filling port coordinates and the number of filling vehicles, and driving the crane to fill each filling vehicle according to the fourth path includes: Analyze the image of the filling area and identify the filling vehicles that do not completely overlap with the corresponding parking spaces; For each incompletely overlapping filling vehicle, obtain an image of its filling port area; Process the acquired Guankou image and extract the actual coordinates of Guankou; Set up a work area map to indicate all the filling port coordinates and the moving range of the crane; the filling port coordinates include the actual filling port coordinates of each incompletely overlapping filling vehicle and the standard filling port coordinates of the completely overlapping filling vehicle; Create an empty path list to store the crane's movement instructions; Assign a priority to the filling truck according to its actual situation; Using the path optimization algorithm, the fourth path is calculated based on the filling port coordinates and the priority of the filling vehicle; The crane is driven to fill each filling vehicle according to the fourth path.

7. The method for automatically positioning a crane pipe according to claim 6, characterized in that: The method of acquiring an image of the perfusion area includes the following contents: Assess the size and shape of the perfusion area and determine the number and location of cameras. The acquisition areas of adjacent cameras overlap. Control the synchronous triggering of the cameras to obtain synchronously captured images; The synchronously shot images are pre-processed and then stitched; The stitched images were smoothed to obtain images of the perfusion area.

8. A crane pipe automatic positioning system, characterized in that: include: An image acquisition unit, the image acquisition unit is used to acquire an image of the perfusion area; wherein a plurality of preset parking spaces are provided in the perfusion area, a plurality of filling vehicles are parked in the filling area, the parking spaces are used for parking the filling vehicles, and the size of the parking spaces is the same as the size of the filling vehicles; An analysis unit, the analysis unit is used to analyze the image of the perfusion area to obtain the number of filling vehicles in the perfusion area and the current parking space image; wherein the current parking space image is an image of a parking space where a filling vehicle is parked; a judgment unit, the judgment unit being used to judge whether the number of filling vehicles is the same as a preset number based on the number of filling vehicles in the filling area, and the judgment unit being used to judge whether each filling vehicle completely overlaps with the corresponding parking space based on the current parking space image; A control unit, wherein the control unit is used to drive the crane to fill each filling vehicle along a preset first path when each filling vehicle completely overlaps with the corresponding parking space and the number of filling vehicles is the same as a preset number; The control unit is also used to generate a second path based on the actual number of filling vehicles when each filling vehicle completely overlaps with the corresponding parking space and the number of filling vehicles is different from the preset number, and drive the crane to fill each filling vehicle according to the second path; The control unit is also used to identify the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space when any filling vehicle does not completely overlap with the corresponding parking space and the number of filling vehicles is the same as the preset number; obtain the actual coordinates of the filling port based on the filling port image, generate a third path based on the actual coordinates of the filling port, and drive the crane to fill each filling vehicle according to the third path; The control unit is also used to identify the filling port image of the filling vehicle that does not completely overlap with the corresponding parking space if there is any filling vehicle that does not completely overlap with the corresponding parking space and the number of filling vehicles is different from the preset number; obtain the actual coordinates of the filling port based on the filling port image, generate a fourth path based on the actual coordinates of the filling port and the number of filling vehicles, and drive the crane pipe to fill each filling vehicle according to the fourth path.

Images