Precise parking method, system and equipment for port crane scene based on lifting appliance identification and storage medium

Abstract

The invention discloses a precise parking method, system and equipment for a port crane scene based on lifting appliance identification, and a storage medium, and aims to directly align an unmanned container truck and a lifting appliance without depending on a guide system of a port crane, improve the operation success rate, directly take the lifting appliance as an alignment reference, do not need to model contour information of the crane, and improve the parking efficiency. The offset of the lifting appliance of the crane does not need to be calibrated, a large amount of function deployment and debugging cost is saved, the method can be transferred to various crane operation scenes such as quay crane rail cranes, box area rail cranes and box area tire cranes, and the generalization ability is high; the lifting appliance position identification comprises two parts of lifting appliance position detection and lifting appliance position prediction, the fact that the lifting appliance can stop swinging during hovering is considered, it is not needed to assume that the lifting machine is kept static in the whole process, and box loading and unloading operation can be automatically carried out after the lifting machine moves; and the abnormal working condition that the lifting appliance swings irregularly or is too large in amplitude in extreme weather can be dealt with.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent driving technology and relates to precise parking of intelligent driving, and specifically to a precise parking method, system, equipment and storage medium for port crane scenarios based on spreader recognition. Background Art

[0002] With the development and progress of unmanned driving technology, unmanned container trucks can automatically complete multiple tasks such as stacking and loading and unloading ships in port scenarios, thereby improving work efficiency, reducing operating costs, and reducing traffic accidents caused by human factors. In the automated operation of the port scene, whether the unmanned container truck can park accurately under the crane is very critical, which directly determines the operation efficiency and degree of automation. However, the general unmanned container truck automatic driving positioning system is difficult to meet the parking accuracy requirements of the crane operation.

[0003] In the prior art, the publication number CN114237213B discloses an automatic driving system for port vehicles. This solution uses the global positioning function of the port vehicle to achieve low-precision parking, and relies on the port machinery guidance system for secondary movement to achieve precise parking. However, this solution needs to rely on an external guidance system, which increases the complexity of the system and the implementation cost, and cannot achieve precise parking when the port machinery guidance system is unavailable or fails. The publication number CN116086467A discloses a target detection and positioning system for port docking scenarios based on unmanned container trucks. This solution achieves precise parking through global fusion positioning and high-precision map matching. The fusion positioning method navigates the vehicle to the vicinity of the port machinery, and then performs relative position matching based on the port machinery position information and contour information to obtain the precise parking position of the vehicle; however, this method relies on high-precision maps and matching algorithms, which increases the deployment cost, and the effect of the matching algorithm may be greatly reduced in a dynamic environment. Summary of the invention

[0004] In view of the above problems, the main purpose of the present invention is to design a precise parking method, system, equipment and storage medium for port crane scenarios based on spreader recognition, so as to solve the problem that precise parking depends on external guidance systems or complex matching algorithms.

[0005] In order to achieve the above purpose, the present invention adopts the following technical solutions: A precise parking method for port crane scenarios based on spreader recognition includes the following steps: Through the fixed-point navigation of the unmanned container truck, the vehicle is navigated to the crane operation point; The point cloud data of the spreader is obtained through the laser radar on the unmanned container truck, and the position of the spreader is identified based on the point cloud data to determine the center position of the spreader; The unmanned container truck detects whether there are boxes on the vehicle and determines the task type of the vehicle, which includes loading and unloading boxes. For packing tasks, high-precision parking is performed based on the difference between the center position of the spreader and the expected packing position; For unloading tasks, the position of the box on the vehicle is identified, and high-precision parking is performed based on the difference between the center position of the spreader and the position of the box on the vehicle.

[0006] As a further description of the present invention, at least two laser radars are arranged, respectively located at the front end and the rear end of the unmanned container truck, and their collection angles are upward.

[0007] As a further description of the present invention, spreader position identification includes spreader position detection and spreader position prediction; Spreader position detection, through the spreader point cloud at a certain moment, calculate the spreader position at that moment; Spreader position prediction, based on the historically detected spreader center positions, predicts the center position of the spreader swing.

[0008] As a further description of the present invention, the spreader position detection is performed using a plane fitting-based method, comprising the following steps: The lifting equipment point cloud P obtained by laser radar scanning is denoised in the height direction to obtain the denoised point cloud P'; Analyze the point cloud P' in the X dimension and select the point closest to the edge in the X dimension; Use the plane fitting method to fit the selected edge point set to obtain the plane equation of the end face of the spreader; According to the fitted plane equation, the position of the end face of the spreader is calculated; Output the position coordinates of the end face of the spreader.

[0009] As a further description of the present invention, the spreader position prediction includes the following steps: The historical detected spreader center position data is used as input to analyze the spreader's swing pattern, including the swing period and amplitude; Calculate the swing period of the spreader based on the empirical length of the spreader cable; Collect the spreader center position data within a swing cycle, including the maximum position and the minimum position; Calculate the center position of the spreader swing according to the maximum position and the minimum position; Based on the center position and the swing period, the future position of the spreader is predicted, and the predicted center position of the spreader is output.

[0010] As a further description of the present invention, the unmanned container truck determines the task type of the vehicle by detecting whether the vehicle has a box, including the following steps: The laser radar on the unmanned container truck is used to obtain the point cloud within 3m above the vehicle, and the point cloud is extracted on the plane. If the plane extraction result contains a normal vector close to the driving direction of the unmanned container truck, it is considered that there are boxes on the truck and the task of the vehicle is to unload the boxes. If the plane extraction result does not contain a normal vector close to the plane of the unmanned container truck's driving direction, it is considered that there are no boxes on the truck and the vehicle's task is to load the boxes.

[0011] A precise parking system for port crane scenarios based on spreader recognition, the system includes a fixed-point navigation module, a spreader recognition module, a task judgment module, and a high-precision parking control module; The fixed-point navigation module, based on the navigation system of the unmanned container truck, combined with the high-precision map or path planning algorithm of the port, accurately navigates the unmanned container truck to the crane operation point; The spreader identification module acquires the point cloud data of the spreader through the laser radar of the unmanned container truck to detect and predict the spreader position; The task judgment module detects whether the unmanned container truck is carrying boxes through its own system or sensor. If the unmanned container truck is carrying boxes, the task of the unmanned container truck is to unload boxes; otherwise, it is to load boxes. The high-precision parking control module includes a loading parking control submodule and an unloading parking control submodule; the loading parking control submodule calculates and controls the unmanned container truck to perform high-precision parking according to the difference between the spreader position and the expected loading position; the unloading parking control submodule performs high-precision parking according to the difference between the spreader position and the box position.

[0012] An electronic device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus, and the memory is used to store a computer program; The processor is used to execute the above method by running the computer program stored in the memory.

[0013] A computer-readable storage medium stores a computer program, wherein the computer program implements the above method when executed by a processor.

[0014] Compared with the prior art, the technical effects of the present invention are: The present invention provides a precise parking method, system, device and storage medium for port crane scenarios based on spreader recognition. By directly aligning the unmanned container truck with the spreader, high-precision parking is achieved without relying on the guidance system of the port crane, thereby improving the success rate of the operation. The spreader is directly used as the alignment reference, and there is no need to model the contour information of the crane, and there is no need to calibrate the offset of the crane spreader, thus saving a lot of function deployment and debugging costs. Moreover, it can be migrated to various crane operation scenarios such as quay crane rail crane, container area rail crane and container area tire crane, and there is no special requirement for the type and layout of the laser radar, and it can be directly migrated. It can be moved to unmanned container trucks with different LiDAR configurations, with strong generalization ability. Spreader position recognition includes two parts: spreader position detection and spreader position prediction. Considering the fact that the spreader will stop when hovering, there is no need to assume that the crane remains stationary throughout the whole process. After the crane moves, it can still automatically perform loading and unloading operations, which is suitable for spreader position recognition in multiple environments. By modeling and analyzing the swing law of the spreader, and then using the spreader detection algorithm and the spreader prediction algorithm to obtain the position of the spreader, while ensuring the accuracy of the spreader position recognition, it can cope with abnormal conditions such as irregular or excessive swing of the spreader in extreme weather. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the method flow of the present invention; Figure 2 A side view of the laser radar arrangement of the present invention; Figure 3 A top view of the laser radar scanning spreader of the present invention; Figure 4 It is a schematic diagram of the detection and prediction results of the hanger position of the present invention. DETAILED DESCRIPTION

[0016] The present invention is described in detail below in conjunction with the accompanying drawings: In one embodiment of the present invention, a precise parking method for a port crane scenario based on spreader recognition is disclosed, referring to Figure 1-4 As shown, precise parking is performed by using laser radar to identify the spreader and the box. While achieving high-precision parking effect, it has stability, strong generalization ability, and can be quickly deployed. Specifically, the method includes the following steps: Through the fixed-point navigation of the unmanned container truck, the vehicle is navigated to the crane operation point; The point cloud data of the spreader is obtained through the laser radar on the unmanned container truck, and the position of the spreader is identified based on the point cloud data to determine the center position of the spreader; The unmanned container truck detects whether there are boxes on the vehicle and determines the task type of the vehicle, which includes loading and unloading boxes. For packing tasks, high-precision parking is performed based on the difference between the center position of the spreader and the expected packing position; For unloading tasks, the position of the box on the vehicle is identified, and high-precision parking is performed based on the difference between the center position of the spreader and the position of the box on the vehicle.

[0017] In this embodiment, at least two laser radars are provided, which are located at the front and rear ends of the unmanned container truck, and the acquisition angle is upward, so that partial areas of the spreader and the box can be scanned, such as Figure 2 Specifically, the front and rear ends of the spreader are all within the scanning range of the laser radar, and at least one third of the front and rear ends of the container are within the scanning range of the laser radar.

[0018] In this embodiment, low-precision parking is first performed through the fixed-point navigation function of the automatic driving, the unmanned container truck is navigated to the vicinity of the crane operation point, and then the spreader above the unmanned container truck is identified; specifically, the spreader position identification includes two parts: spreader position detection and spreader position prediction.

[0019] Spreader position detection: Calculate the spreader position at a certain moment through the spreader point cloud at that moment; Figure 3 As shown, the point cloud of the spreader obtained by the laser radar scanning is denoted as P. The point cloud P is actually the scanning result of the laser radar on the side surface of the spreader, which basically conforms to the distribution law of the plane. Therefore, this embodiment chooses to use a method based on plane fitting to detect the position of the spreader. Specifically, it includes the following steps: The lifting equipment point cloud P obtained by laser radar scanning is denoised in the height direction to obtain the denoised point cloud P'; Analyze the point cloud P' in the X dimension and select the point closest to the edge in the X dimension; Use the plane fitting method to fit the selected edge point set to obtain the plane equation of the end face of the spreader; According to the fitted plane equation, the position of the end face of the spreader is calculated; Output the position coordinates of the end face of the spreader.

[0020] Spreader position prediction: Based on the historically detected spreader center position, predict the spreader swing center position. During actual loading and unloading operations, the spreader is not absolutely stationary in the longitudinal direction, but swings periodically. Especially in scenes with strong winds, the swing phenomenon is particularly obvious. Therefore, only detecting the spreader position at a certain moment is not enough to achieve high-precision alignment. Therefore, this embodiment predicts the spreader swing center position through spreader position prediction. Specifically, it includes the following steps: The historical detected spreader center position data is used as input to analyze the spreader's swing law, including the swing period and amplitude. The spreader's swing can be approximated as a simple pendulum motion, and its swing period is: ; in, is the swing period (in seconds), is the pendulum length (in meters), is the acceleration due to gravity (unit: m / s²); from this formula, it can be found that the swing period of a simple pendulum motion is only related to the swing length, and has nothing to do with whether there is a container on the spreader; Calculate the swing period of the spreader based on the empirical length of the spreader cable; Collect the spreader center position data within a swing cycle, including the maximum position and the minimum position; Calculate the center position of the spreader swing according to the maximum position and the minimum position; Based on the center position and the swing period, the future position of the spreader is predicted, and the predicted center position of the spreader is output.

[0021] For port cranes, the spreader is connected to the crane by a cable, and when the spreader swings, the other end of the cable is fixed to the crane. Therefore, the swing of the spreader can be regarded as a simple pendulum motion, that is, the swing period and amplitude will not change without external force, and the swing period has nothing to do with the amplitude and the mass of the pendulum ball, but only with the pendulum length.

[0022] like Figure 4 As shown, it is the point cloud data for a section of the sling swing, taking the vehicle system as the reference coordinate system, and the sling recognition effect diagram; from the figure, it can be seen that the real-time position of the sling performing a single pendulum motion can be accurately detected, and thus this embodiment can more accurately predict the center position of the sling swing.

[0023] In this embodiment, the unmanned container truck determines the task type of the vehicle by detecting whether the vehicle has a box, including the following steps: The laser radar on the unmanned container truck is used to obtain the point cloud within 3m above the vehicle, and the point cloud is extracted on the plane. If the plane extraction result contains a normal vector close to the driving direction of the unmanned container truck, it is considered that there are boxes on the truck and the task of the vehicle is to unload the boxes. If the plane extraction result does not contain a normal vector close to the plane of the unmanned container truck's driving direction, it is considered that there are no boxes on the truck and the vehicle's task is to load the boxes.

[0024] In another embodiment of the present invention, a precise parking system for port crane scenarios based on spreader recognition is disclosed, the system comprising a fixed-point navigation module, a spreader recognition module, a task judgment module, and a high-precision parking control module; The fixed-point navigation module, based on the navigation system of the unmanned container truck, combined with the high-precision map or path planning algorithm of the port, accurately navigates the unmanned container truck to the crane operation point; The spreader identification module acquires the point cloud data of the spreader through the laser radar of the unmanned container truck to detect and predict the spreader position; The task judgment module detects whether the unmanned container truck is carrying boxes through its own system or sensor. If the unmanned container truck is carrying boxes, the task of the unmanned container truck is to unload boxes; otherwise, it is to load boxes. The high-precision parking control module includes a loading parking control submodule and an unloading parking control submodule; the loading parking control submodule calculates and controls the unmanned container truck to perform high-precision parking according to the difference between the spreader position and the expected loading position, ensuring that the box can be accurately aligned with the spreader for loading; the unloading parking control submodule performs high-precision parking according to the difference between the spreader position and the box position, so that the spreader can accurately grab the box for unloading.

[0025] Through the above content, the precise parking method and system of the present invention are disclosed. Compared with the prior art, the present invention has the following advantages: 1. The present invention directly aligns the unmanned container truck with the spreader without relying on the guidance system of the port machinery, thus achieving high-precision parking and improving the success rate of the operation; 2. The present invention directly uses the spreader as the alignment reference, and there is no need to model the profile information of the crane, and there is no need to calibrate the offset of the crane spreader, which saves a lot of function deployment and debugging costs; 3. The present invention focuses on the crane's spreader rather than the crane itself, and can be migrated to various crane operation scenarios such as quay crane rail crane, container area rail crane and container area tire crane. It has no special requirements on the type and layout of the lidar, and can be directly migrated to unmanned container trucks with different lidar configurations, with strong generalization ability.

[0026] 4. The spreader position recognition of the present invention includes two parts: spreader position detection and spreader position prediction. Considering the fact that the spreader may stop when hovering, there is no need to assume that the crane remains stationary throughout the whole process. After the crane moves, the container loading and unloading operations can still be performed automatically, which is applicable to the spreader position recognition in multiple environments; 5. The present invention analyzes the swing pattern of the spreader, and then uses the spreader detection algorithm and the spreader prediction algorithm to obtain the position of the spreader. While ensuring accurate spreader position identification, it can cope with abnormal conditions such as irregular spreader swing or excessive amplitude in extreme weather.

[0027] In another embodiment of the present invention, an electronic device is also included. The electronic device may include a processor and a memory storing computer program instructions.

[0028] Specifically, in this embodiment, the processor may include a central processing unit (CPU), or a specific integrated circuit, or may be configured as one or more integrated circuits of this embodiment; the memory may include a large-capacity memory for data or instructions, including but not limited to a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a tape, or a universal serial bus (USB) drive, or a combination of two or more of these; where appropriate, the memory may include a removable or non-removable (or fixed) medium; in a specific embodiment, the memory is a non-volatile solid-state memory. In a specific embodiment, the memory includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM) or a flash memory, or a combination of two or more of these.

[0029] The processor implements the precise alignment method disclosed above in the present invention by reading and executing computer program instructions stored in the memory.

[0030] It should also be noted that the electronic device of this embodiment may also include a communication interface and a communication bus. Among them, the processor, memory, and communication interface are connected through the communication bus and communicate with each other. The communication interface is mainly used to realize the communication between the various units, modules, devices or equipment in the embodiment of the present invention.

[0031] The communication bus mentioned above includes hardware, software or a combination of hardware and software, coupling the components of the online data flow device to each other. Where appropriate, the communication bus may include one or more buses.

[0032] In addition, in combination with the precise alignment method in the above embodiment, an embodiment of the present invention may provide a computer storage medium for implementation, wherein the computer storage medium stores computer program instructions; the computer program instructions are executed by a processor to implement the precise alignment method.

[0033] It should be clear that the present invention is not limited to the methods, systems, and devices disclosed above, but also includes various changes, modifications, and additions made by those skilled in the art based on the ideas of the present invention, or changes in the order of steps.

[0034] When the present invention is implemented in hardware, it can be an electronic circuit, a dedicated integrated circuit, appropriate firmware, plug-in, function card, etc.; when implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks, and the programs or code segments can be stored in a machine-readable medium, or uploaded on a transmission medium or communication link through a data signal carried in a carrier. "Machine-readable medium" may include any medium capable of storing or transmitting information, such as: electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, optical disks, hard disks, optical fiber media, radio frequency links, etc. The code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0035] The above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Other modifications or equivalent substitutions made to the technical solution of the present invention by ordinary technicians in this field should be included in the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solution of the present invention.

Claims

1. A precise parking method for port crane scenarios based on spreader recognition, characterized in that: The steps include: Through the fixed-point navigation of the unmanned container truck, the vehicle is navigated to the crane operation point; The point cloud data of the spreader is obtained through the laser radar on the unmanned container truck, and the position of the spreader is identified based on the point cloud data to determine the center position of the spreader; The unmanned container truck detects whether there are boxes on the vehicle and determines the task type of the vehicle, which includes loading and unloading boxes. For packing tasks, high-precision parking is performed based on the difference between the center position of the spreader and the expected packing position; For unloading tasks, the position of the box on the vehicle is identified, and high-precision parking is performed based on the difference between the center position of the spreader and the position of the box on the vehicle.

2. According to claim 1, a precise parking method for port crane scenarios based on spreader recognition is characterized by: At least two laser radars are set up, located at the front and rear ends of the unmanned container truck, and their collection angles are upward to scan spreaders and boxes.

3. According to claim 1, a precise parking method for port crane scenarios based on spreader recognition is characterized by: Spreader position identification includes spreader position detection and spreader position prediction; Spreader position detection, through the spreader point cloud at a certain moment, calculate the spreader position at that moment; Spreader position prediction, based on the historically detected spreader center positions, predicts the center position of the spreader swing.

4. According to claim 3, a precise parking method for port crane scenarios based on spreader recognition is characterized in that: The spreader position detection is performed using a plane fitting-based method, which includes the following steps: The lifting equipment point cloud P obtained by laser radar scanning is denoised in the height direction to obtain the denoised point cloud P'; Analyze the point cloud P' in the X dimension and select the point closest to the edge in the X dimension; Use the plane fitting method to fit the selected edge point set to obtain the plane equation of the end face of the spreader; According to the fitted plane equation, the position of the end face of the spreader is calculated; Output the position coordinates of the end face of the spreader.

5. According to claim 3, a precise parking method for port crane scenarios based on spreader recognition is characterized in that: Spreader position prediction includes the following steps: The historical detected spreader center position data is used as input to analyze the spreader's swing pattern, including the swing period and amplitude; Calculate the swing period of the spreader based on the empirical length of the spreader cable; Collect the spreader center position data within a swing cycle, including the maximum position and the minimum position; Calculate the center position of the spreader swing according to the maximum position and the minimum position; Based on the center position and the swing period, the future position of the spreader is predicted, and the predicted center position of the spreader is output.

6. According to claim 1, a precise parking method for port crane scenarios based on spreader recognition is characterized by: The unmanned container truck determines the mission type of the vehicle by detecting whether there is a box on the vehicle, including the following steps: The laser radar on the unmanned container truck is used to obtain the point cloud within 3m above the vehicle, and the point cloud is extracted on the plane. If the plane extraction result contains a normal vector close to the driving direction of the unmanned container truck, it is considered that there are boxes on the truck and the task of the vehicle is to unload the boxes. If the plane extraction result does not contain a normal vector close to the plane of the unmanned container truck's driving direction, it is considered that there are no boxes on the truck and the vehicle's task is to load the boxes.

7. A precise parking system for port crane scenarios based on spreader recognition according to any one of claims 1 to 6, characterized in that: The system includes a fixed-point navigation module, a spreader identification module, a task judgment module, and a high-precision parking control module; The fixed-point navigation module, based on the navigation system of the unmanned container truck, combined with the high-precision map or path planning algorithm of the port, accurately navigates the unmanned container truck to the crane operation point; The spreader identification module acquires the point cloud data of the spreader through the laser radar of the unmanned container truck to detect and predict the spreader position; The task judgment module detects whether the unmanned container truck is carrying boxes through its own system or sensor. If the unmanned container truck is carrying boxes, the task of the unmanned container truck is to unload boxes; otherwise, it is to load boxes. The high-precision parking control module includes a loading parking control submodule and an unloading parking control submodule; the loading parking control submodule calculates and controls the unmanned container truck to perform high-precision parking according to the difference between the spreader position and the expected loading position; The unloading parking control submodule performs high-precision parking according to the difference between the spreader position and the box position.

8. An electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein: The processor, the communication interface and the memory communicate with each other via the communication bus, wherein the memory is used to store a computer program; The processor is configured to execute the method according to any one of claims 1 to 6 by running the computer program stored in the memory.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program implements the method according to any one of claims 1 to 6 when executed by a processor.

Images