Auxiliary positioning method and system for hoisting components in prefabricated scenarios

Abstract

The present disclosure relates to an auxiliary positioning method and system for hoisting components in an assembled scenario. The system includes a hoisting positioning device and an image sensor. The hoisting positioning device includes a spreader, a motor, and a controller. The image sensor is configured to obtain a video of the hoisting target position at the construction site in real time. The controller is configured to control the operation of the motor to drive the movement of the spreader. The spreader is used for hoisting a target precast component. Meanwhile, the controller receives the video transmitted by the image sensor, identifies and determines the component position of the target precast component and the hoisting target position from the video based on a target recognition model. When the component position is aligned with the hoisting target position, the target precast component is installed to perform the installation operation of the next precast component. Wherein, the target recognition model is obtained by training the target detection YOLO model based on sample videos.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the technical field of construction hoisting automation, and particularly to a method and system for assisting in positioning hoisted components in an assembled scenario. Background Art

[0002] Currently, approximately 70% of the time for building prefabricated buildings is spent on component assembly. Prefabricated structural components are usually transported by a crane to a position near the assembly location for preliminary positioning. Subsequently, several workers cooperate to adjust the horizontal attitude of the hoisted component to align it with the target installation area, and cooperate with the crane to place the component in the correct position and orientation. Due to the large size, heavy weight of the components, and being susceptible to factors such as insufficient stiffness, deformation of the crane boom and ropes, limited precision of traditional crane motion control, and harsh outdoor environments, the positioning accuracy and stability of the components during hoisting are poor. Therefore, it is usually necessary to repeatedly perform lifting steps, which will seriously affect the construction efficiency. Summary of the Invention

[0003] To solve the above technical problems or at least partially solve the above technical problems, embodiments of the present disclosure provide a method and system for assisting in positioning hoisted components in an assembled scenario.

[0004] In a first aspect, embodiments of the present disclosure provide a system for assisting in positioning hoisted components in an assembled scenario. The system includes a hoisting positioning device and an image sensor. The hoisting positioning device at least includes a spreader, a motor, and a controller, and is pre-installed at a preset installation position at a construction site;

[0005] The image sensor is configured to obtain a video of the hoisting target position at the construction site in real time;

[0006] The controller is configured to control the operation of the motor to drive the movement of the spreader. The spreader is used to hoist a target prefabricated component; at the same time, receive the video transmitted by the image sensor, identify and determine the component position of the target prefabricated component and the hoisting target position from the video based on a target recognition model, and install the target prefabricated component when the component position is aligned with the hoisting target position to perform the installation operation of the next prefabricated component; wherein, the target recognition model is obtained by training a target detection YOLO model based on a sample video.

[0007] In one embodiment, the controller is further configured to calculate the working duration of the motor according to the installation position of the hoisting positioning device, the hoisting target position, and the output power of the motor before the component position is aligned with the hoisting target position, generate a control instruction based on the working duration, and control the operation of the motor based on the control instruction to drive the movement of the spreader, so that the target prefabricated component moves and aligns with the hoisting target position.

[0008] In one embodiment, the controller is further configured to identify and determine the positioning error distance of the hoisting target position from the video based on the target recognition model, and when the positioning error distance is less than or equal to a preset error distance, calculate the working duration of the motor according to the installation position of the hoisting positioning device, the hoisting target position, and the output power of the motor.

[0009] In one embodiment, the installation position of the hoisting positioning device is determined by the following method:

[0010] According to the construction design drawings of the prefabricated building, mark the coordinates of the hoisting target positions of each precast component, group all the hoisting target positions in the form of regions according to the maximum working range of the hoisting positioning device, find a point within each group of regions such that the sum of the distances to all the hoisting target positions is the shortest, and install the hoisting positioning device at this point; wherein, the installation order of the precast components within each group of regions is determined according to the installation task priority; or the installation order of component hoisting is determined according to the order of the adjacent position relationships of the precast components.

[0011] In one embodiment, the image sensor includes at least one camera and / or depth sensor.

[0012] In one embodiment, the YOLO model includes a first backbone network, a second backbone network, a feature extraction network, and a detection head; the output ends of the first backbone network and the second backbone network are both connected to the input end of the feature extraction network; the training process of the target recognition model includes:

[0013] Obtain a sample video, where the sample video includes multiple frames of two-dimensional images and multiple frames of depth images collected by the at least one camera and depth sensor, each frame of two-dimensional image corresponds to a frame of depth image and is labeled with the component position and the hoisting target position;

[0014] Input the multiple frames of two-dimensional images into the first backbone network, and input the multiple frames of depth images into the second backbone network to iteratively train the YOLO model until the loss function of the YOLO model is less than or equal to a preset value and the training ends.

[0015] In one embodiment, there are multiple hoisting target positions, each hoisting target position is located at the construction site and marked with at least one physical mark, each hoisting target position corresponds to a precast component, and all the precast components are used to assemble and form a prefabricated building structure.

[0016] Second aspect, embodiments of the present disclosure provide an auxiliary positioning method for hoisting components in an assembled scenario. This method is applied to an auxiliary positioning system for hoisting components, which system includes a hoisting positioning device and an image sensor. The hoisting positioning device includes a spreader, a motor, and a controller, and is pre-installed at a preset installation position at a construction site. This method is executed by the controller and includes the following steps:

[0017] Control the image sensor to continuously obtain a video of the hoisting target position at the construction site;

[0018] Control the operation of the motor to drive the movement of the spreader. The spreader is used to hoist a target precast component. At the same time, receive the video transmitted by the image sensor, identify and determine the component position of the target precast component and the hoisting target position from the video based on a target recognition model. When the component position is aligned with the hoisting target position, install the target precast component to perform the installation operation of the next precast component. Among them, the target recognition model is obtained by training the target detection YOLO model based on sample videos.

[0019] Third aspect, embodiments of the present disclosure provide a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the auxiliary positioning method for hoisting components in the assembled scenario of the above embodiments.

[0020] Fourth aspect, embodiments of the present disclosure provide an automatic hoisting device, including:

[0021] An image sensor;

[0022] A hoisting positioning device, the hoisting positioning device includes a spreader, a motor, and a processor; and a memory for storing a computer program;

[0023] Wherein, the processor is configured to execute the auxiliary positioning method for hoisting components in the assembled scenario of the above embodiments by executing the computer program.

[0024] The technical solutions provided by the embodiments of the present disclosure have the following advantages compared with the prior art:

[0025] The hoisting member auxiliary positioning method and system in the prefabricated scenario provided by the embodiments of the present disclosure. The system includes a hoisting positioning device and an image sensor. The hoisting positioning device at least includes a spreader, a motor, and a controller, and is pre-installed at a preset installation position at the construction site. The image sensor acquires the video of the hoisting target position at the construction site in real time. The controller controls the operation of the motor to drive the movement of the spreader. The spreader is used to hoist the target prefabricated member. At the same time, it receives the video transmitted by the image sensor, and based on the target recognition model, determines the member position of the target prefabricated member and the hoisting target position from the video. When the member position is aligned with the hoisting target position, the target prefabricated member is installed to perform the installation operation of the next prefabricated member. Among them, the target recognition model is obtained by training the target detection YOLO model based on the sample video. The solution of this embodiment provides a scheme for auxiliary positioning of hoisting members in the prefabricated scenario. The motor is used to control the spreader to tow the hoisting member. The target recognition model based on the video target detection positioning algorithm model, that is, the YOLO model, determines the member position and the hoisting target position in real time through the collected video, and accordingly accurately judges whether the member is accurately positioned to perform the installation operation of the prefabricated member. In this way, the attitude and position of the hoisted member can be automatically and accurately positioned, and the positioning accuracy of the member during hoisting is high, and the member assembly can be completed precisely and efficiently, that is, the efficiency and accuracy of the hoisting member positioning are improved, and further the construction efficiency and quality of the prefabricated building are improved. In addition, this hoisting positioning assistance scheme can replace manual labor to complete high-intensity and high-risk hoisting tasks, avoid personnel operation risks, and ensure construction safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings herein are incorporated into the specification and constitute a part of this specification, showing the embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure.

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0028] Figure 1 It is a schematic diagram of the hoisting member auxiliary positioning system in the prefabricated scenario of the embodiments of the present disclosure;

[0029] Figure 2 It is a schematic diagram of the hoisting member auxiliary positioning process in the prefabricated scenario of the embodiments of the present disclosure;

[0030] Figure 3 It is a flowchart of the method for the target recognition model to recognize and process the video in the embodiments of the present disclosure;

[0031] Figure 4 This is a flowchart of a method for assisting in positioning a hoisting member in an assembled scenario according to an embodiment of the present disclosure. Detailed implementation manners

[0032] In order to more clearly understand the above objects, features, and advantages of the present disclosure, the solutions of the present disclosure will be further described below. It should be noted that, without conflict, the embodiments of the present disclosure and the features in the embodiments may be combined with each other.

[0033] In the following description, many specific details are set forth in order to fully understand the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only a part of the embodiments of the present disclosure, rather than all the embodiments.

[0034] It should be understood that in the following text, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" may mean: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B may be singular or plural. The character " / " generally means that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or its similar expression means any combination of these items, including any combination of single item (one) or plural items (ones). For example, at least one (one) of a, b, or c may mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c may be single or multiple.

[0035] Figure 1 This is a schematic diagram of a system for assisting in positioning a hoisting member in an assembled scenario according to an embodiment of the present disclosure. The system may include a hoisting positioning device 20 and an image sensor 10. The hoisting positioning device 20 may at least include a spreader 201, a motor 202, and a controller 203, and is pre-installed at a preset installation position at the construction site. Exemplarily, the hoisting positioning device may be an automatic hoisting robot, such as a hoisting robot with 6 degrees of freedom, which at least includes a spreader such as a hook, a motor such as a servo motor, and a controller such as a processor, etc. The motor may be connected to the hook through an actuator (such as a transmission mechanism, a reducer, etc.), and control the actuator (not shown in the figure) to drive the hook to move. In this embodiment, the specific structure of the hoisting positioning device is not limited, and the focus is on the improvement of its control method.

[0036] The image sensor 10 is used to obtain the video of the hoisting target position at the construction site in real time. The hoisting target position is the installation position of the target precast component. In one embodiment, there are multiple hoisting target positions, and each hoisting target position corresponds to a precast component. All precast components are used to assemble an assembled building structure. Each hoisting target position is located at the construction site and is marked with at least one physical mark such as a reference frame mark. During the construction process, as each precast component is gradually installed, physical marks can be set on the already installed precast component as the hoisting target position for the installation of the next precast component.

[0037] The controller 203 is used to control the operation of the motor 202 to drive the actuator to drive the movement of the lifting tool such as a hook, and the lifting tool is used to hoist the target precast component; the controller 203 simultaneously receives the video transmitted by the image sensor 10, and based on the target recognition model, identifies and determines the component position of the target precast component and the hoisting target position from the video. When the component position is aligned with the hoisting target position, the target precast component is installed to perform the installation operation of the next precast component; wherein, the target recognition model is obtained by training the target detection YOLO model based on the sample video.

[0038] Exemplarily, after the target precast component is hoisted on the lifting tool such as a hook, the controller 203 controls the operation of the motor 202 to drive the actuator to drive the movement of the lifting tool such as a hook, so as to move the target precast component. At the same time, the controller 203 receives the video transmitted by the image sensor 10, and based on the target recognition model, identifies and determines the component position of the target precast component such as the three-dimensional (3D) coordinate position and the hoisting target position such as the 3D coordinate position from the video, and performs real-time matching monitoring on the determined component position and the hoisting target position. When the component position is aligned with the hoisting target position, the movement of the hook is stopped, so that the installer can install the target precast component, and then continue to perform the installation operation of the next precast component. The operation process is the same as the hoisting and position recognition and matching process of the previous precast component.

[0039] Among them, the YOLO model can be the YOLOv11 model, but it is not limited to this, and models such as YOLOv10 are also applicable. The YOLOv11 model has demonstrated significant performance improvements in multiple aspects, especially in object detection tasks. Its enhanced feature extraction ability: The YOLOv11, through improved backbone and neck networks, introduces C3K2 and C2PSA modules, significantly enhancing the feature extraction ability and making the model perform better in complex tasks (such as multi-object detection, occlusion handling, etc.); in addition, it optimizes speed and efficiency. The YOLOv11 adopts a more efficient architecture and training process, improving the processing speed while maintaining high accuracy. Therefore, the object recognition model trained accordingly can more accurately identify the component positions and lifting target positions of the target precast components in the video, thereby improving the efficiency and accuracy of component lifting positioning, and further enhancing the construction efficiency and quality of prefabricated buildings.

[0040] The solution of this embodiment provides a solution for assisting in the positioning of lifted components in a prefabricated scenario. The motor is used to control the spreader to tow the lifted component. The object recognition model trained based on the video object detection and positioning algorithm model, i.e., the YOLO model, determines the component position and the lifting target position in real time through the collected video. Based on this, it accurately judges whether the component has been precisely positioned for the installation operation of the precast component. In this way, the attitude and position of the lifted component can be automatically and accurately positioned, with high positioning accuracy during the lifting process of the component, and the component assembly can be completed precisely and efficiently, that is, the efficiency and accuracy of component lifting positioning are improved, and further the construction efficiency and quality of prefabricated buildings are enhanced. In addition, this lifting positioning assistance solution can replace manual labor to complete high-intensity and high-risk lifting tasks, avoiding personnel operation risks and ensuring construction safety.

[0041] To further improve the construction efficiency and quality of prefabricated buildings, based on the above embodiment, in one embodiment, referring to Figure 2 as shown, the controller 203 is further configured to calculate the working duration of the motor according to the installation position of the lifting positioning device, the lifting target position, and the output power of the motor before the component position is aligned with the lifting target position, generate a control instruction based on the working duration, and control the operation of the motor based on the control instruction to drive the movement of the spreader, so that the target precast component moves to be aligned with the lifting target position.

[0042] Exemplarily, the controller 203 can calculate the working duration required for the servo motor to control the spreader to pull the precast component to be lifted to the target lifting position based on the positional relationship between the installation position of the lifting equipment itself and the target lifting position, such as the distance relationship (e.g., the straight-line distance), the azimuth relationship (e.g., the angles in the horizontal and vertical directions), the height difference relationship, and the spatial relative position relationship, etc., in combination with the output power of the servo motor. Based on the calculation result, i.e., the working duration, an action control instruction for equipment control is generated, and the operation of the servo motor is controlled based on the control instruction to drive the actuator to drive the movement of the spreader, so that the target precast component is moved and aligned to the target lifting position. Among them, the movement trajectory can be generated according to the positional relationship, and then the working duration can be determined by combining the power-time conversion method. In some cases, the adjustment time can also be extended or shortened through incremental PID or adaptive control. In this way, the shortest time required for lifting the component can be accurately calculated in advance, thereby improving the efficiency and accuracy of component lifting and positioning, and further improving the construction efficiency and quality of the prefabricated building.

[0043] In one embodiment, the controller 203 is further configured to identify and determine the positioning error distance of the target lifting position from the video based on the target recognition model. When the positioning error distance is less than or equal to the preset error distance, the working duration of the motor is calculated according to the installation position of the lifting and positioning equipment, the target lifting position, and the output power of the motor. In this way, the influence of the positioning error can be avoided, and the shortest time required for lifting the component can be calculated more accurately, thereby improving the efficiency and accuracy of component lifting and positioning, and further improving the construction efficiency and quality of the prefabricated building.

[0044] Based on any one of the above embodiments, in one embodiment, with reference to Figure 2 as shown, the installation position of the lifting and positioning equipment is determined in the following manner:

[0045] According to the construction design drawings of the prefabricated building, mark the coordinates of the target lifting position of each precast component, group all the target lifting positions in the form of regions according to the maximum working range of the lifting and positioning equipment, find a point within each group of regions such that the sum of the distances to all the target lifting positions is the shortest, and install the lifting and positioning equipment at this point; among them, the installation order of the precast components within each group of regions is determined according to the installation task priority; or the installation order of component lifting is determined according to the order of the adjacent position relationship of the precast components.

[0046] Exemplarily, for example, there are 100 prefabricated components, corresponding to 100 hoisting target positions. The maximum working range of the hoisting positioning device, such as a hoisting robot, is the maximum distance range that the spreader can move. Project this maximum distance range onto the ground. Group the 100 hoisting target positions within the projected area. For example, 10 adjacent hoisting target positions are grouped into one group and a sub-region is divided. Each sub-region includes the corresponding 10 adjacent hoisting target positions. Find a point within each group of regions, that is, the sub-region, such that the sum of the distances from this point to all 100 hoisting target positions is the shortest. This problem can be transformed into a geometric median problem with constraints. For example, it can be solved by introducing boundary constraints through the Lagrange multiplier method, which can be understood with reference to the prior art and will not be elaborated here. Installing the hoisting positioning device at this point can make the distance from the spreader on the hoisting positioning device to each hoisting target position relatively short when moving, thereby reducing the time required for each hoisting of the component, further improving the efficiency of component hoisting positioning, and further improving the construction efficiency of the prefabricated building.

[0047] In one embodiment, the image sensor includes at least one camera and / or depth sensor. The camera can collect RGB images, that is, two-dimensional images, and the depth sensor can collect depth images.

[0048] Correspondingly, in one embodiment, in order to further improve the construction efficiency and quality of the prefabricated building, in one embodiment, referring to Figure 3 as shown, improvements are made to the architecture and training method of the YOLO model such as YOLOv11. Specifically, the YOLO model includes a first backbone network, a second backbone network, a feature extraction network (such as the neck network Neck), and a detection head (Head); the output ends of the first backbone network and the second backbone network are both connected to the input end of the feature extraction network. That is, a backbone network (Backbone) is added to the original architecture of the YOLO model. The training process of the target recognition model includes:

[0049] Obtain a sample video, where the sample video includes multiple frames of two-dimensional images and multiple frames of depth images collected by the at least one camera and depth sensor. Each frame of two-dimensional image corresponds to a frame of depth image and is labeled with the component position and the hoisting target position;

[0050] Input the multiple frames of two-dimensional images into the first backbone network, and input the multiple frames of depth images into the second backbone network to iteratively train the YOLO model until the loss function of the YOLO model is less than or equal to a preset value and the training ends.

[0051] Exemplarily, in this embodiment, for the added second backbone network, since it processes depth images containing more feature information, the number of convolutional layers in the second backbone network is greater than that in the first backbone network, so as to better extract the features of depth images, combine with the features of two-dimensional images extracted by the first backbone network for comprehensive training, so that the position determination result of the trained model is more accurate, that is, the component position and the hoisting target position of the target precast component in the video can be identified and determined more accurately, thereby further improving the efficiency and accuracy of component hoisting positioning, and further improving the construction efficiency and quality of prefabricated buildings.

[0052] Among them, the first backbone network extracts the first feature map of the two-dimensional image, the second backbone network extracts the second feature map of the depth image, and then the feature extraction network such as the neck network extracts the feature maps of the first feature map and the second feature map and fuses the feature maps into the detection head to output the predicted component position and the corresponding predicted hoisting target position of the precast component, and updates the loss function of the YOLO model based on the difference between the predicted component position and the predicted hoisting target position and the corresponding labeled component position and hoisting target position.

[0053] It should be noted that although several modules or units of a device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units. The components shown as modules or units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of the present disclosure. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0054] The embodiment of the present disclosure provides a method for assisting in positioning hoisted components in a prefabricated scenario. This method is applied to a system for assisting in positioning hoisted components. The system includes a hoisting positioning device and an image sensor. The hoisting positioning device includes a spreader, a motor, and a controller, and is pre-installed at a preset installation position at the construction site; refer to Figure 4 As shown in, this method is executed by the controller and may include the following steps:

[0055] Control the image sensor to continuously obtain the video of the hoisting target position at the construction site;

[0056] Control the operation of the motor to drive the movement of the spreader, which is used for hoisting the target precast component; at the same time, receive the video transmitted by the image sensor, identify and determine the component position of the target precast component and the hoisting target position from the video based on the target recognition model, and when the component position is aligned with the hoisting target position, install the target precast component to perform the installation operation of the next precast component. Wherein, the target recognition model is obtained by training the target detection YOLO model based on the sample video.

[0057] In one embodiment, before the component position is aligned with the hoisting target position, the controller calculates the working duration of the motor according to the installation position of the hoisting positioning device, the hoisting target position and the output power of the motor, generates a control instruction based on the working duration, and controls the operation of the motor based on the control instruction to drive the movement of the spreader, so that the target precast component moves and aligns with the hoisting target position.

[0058] In one embodiment, the controller identifies and determines the positioning error distance of the hoisting target position from the video based on the target recognition model. When the positioning error distance is less than or equal to the preset error distance, the controller calculates the working duration of the motor according to the installation position of the hoisting positioning device, the hoisting target position and the output power of the motor.

[0059] In one embodiment, the installation position of the hoisting positioning device is determined by the following method:

[0060] According to the construction design drawings of the prefabricated building, mark the coordinates of the hoisting target position of each precast component, group all the hoisting target positions in the form of regions according to the maximum working range of the hoisting positioning device, find a point in each group of regions so that the sum of the distances to all the hoisting target positions is the shortest, and install the hoisting positioning device at this point; wherein, the installation order of the precast components in each group of regions is determined according to the installation task priority; or the installation order of component hoisting is determined according to the adjacent position relationship of the precast components.

[0061] In one embodiment, the image sensor includes at least one camera and / or depth sensor.

[0062] In one embodiment, the YOLO model includes a first backbone network, a second backbone network, a feature extraction network and a detection head; the output ends of the first backbone network and the second backbone network are both connected to the input end of the feature extraction network; the training process of the target recognition model includes:

[0063] Obtain a sample video, where the sample video includes multiple frames of two-dimensional images and multiple frames of depth images collected by the at least one camera and depth sensor, and each frame of two-dimensional image corresponds to a frame of depth image and is labeled with the component position and the hoisting target position;

[0064] Input the multiple frames of two-dimensional images into the first backbone network, and input the multiple frames of depth images into the second backbone network to iteratively train the YOLO model until the loss function of the YOLO model is less than or equal to a preset value and the training ends.

[0065] In one embodiment, the number of convolutional layers in the second backbone network is greater than the number of convolutional layers in the first backbone network. The first backbone network extracts a first feature map of the two-dimensional image, and the second backbone network extracts a second feature map of the depth image. Then, a feature extraction network such as a neck network extracts the feature maps of the first feature map and the second feature map and fuses the feature maps into the detection head to output the predicted component position of the precast component and the corresponding predicted hoisting target position. Based on the difference between the predicted component position and the predicted hoisting target position and the corresponding labeled component position and hoisting target position, the loss function of the YOLO model is updated.

[0066] In one embodiment, there are multiple hoisting target positions, each hoisting target position is located at the construction site and is marked with at least one physical mark, each hoisting target position corresponds to a precast component, and all precast components are used to assemble an assembled building structure.

[0067] It should be noted that although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that these steps must be executed in this specific order, or that all the steps shown must be executed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc. Additionally, it is also easy to understand that these steps may be executed synchronously or asynchronously, for example, in multiple modules / processes / threads.

[0068] Regarding the method in the above embodiments, the specific manner of performing operations for each step and the corresponding technical effects have been described in detail in the corresponding embodiments of the system, and will not be elaborated here in detail.

[0069] The embodiment of the present disclosure also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the hoisting component auxiliary positioning method in the assembled scenario described in any one of the above embodiments.

[0070] Exemplarily, the readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0071] The computer-readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, in which the readable program code is carried. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium may also be any readable medium other than the readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination of the above.

[0072] The embodiments of the present disclosure further provide an automatic hoisting device such as an automatic hoisting robot, including: an image sensor; a hoisting positioning device, where the hoisting positioning device includes a spreader, a motor, and a processor; and a memory for storing a computer program; wherein the processor is configured to execute the hoisting member auxiliary positioning method in the prefabricated scenario of the above embodiments by executing the computer program.

[0073] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including", or any other variant thereof is intended to cover a non-exclusive inclusion, such that a process, method, article, or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such a process, method, article, or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article, or device including the element.

[0074] The above are only specific embodiments of the present disclosure, enabling those skilled in the art to understand or implement the present disclosure. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to these embodiments described herein, but rather will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An auxiliary positioning system for hoisting components in an assembled scenario, characterized in that, The system includes a hoisting positioning device and an image sensor. The hoisting positioning device at least includes a lifting tool, a motor, and a controller, and is pre-installed at a preset installation position on the construction site; The image sensor is used to obtain the video of the hoisting target position on the construction site in real time; The controller is used to control the operation of the motor to drive the movement of the lifting tool. The lifting tool is used to hoist the target precast component. At the same time, it receives the video transmitted by the image sensor, and based on the target recognition model, it identifies and determines the component position of the target precast component and the hoisting target position from the video. When the component position is aligned with the hoisting target position, the target precast component is installed to perform the installation operation of the next precast component. Among them, the target recognition model is obtained by training the target detection YOLO model based on the sample video; Among them, the image sensor includes at least one camera and / or depth sensor; the YOLO model includes a first backbone network, a second backbone network, a feature extraction network, and a detection head; the output ends of the first backbone network and the second backbone network are both connected to the input end of the feature extraction network; the training process of the target recognition model includes: Obtain a sample video, which includes multiple two-dimensional images and multiple depth images collected by the at least one camera and depth sensor. Each two-dimensional image corresponds to a depth image and is labeled with the component position and the hoisting target position; Input the multiple two-dimensional images into the first backbone network, and input the multiple depth images into the second backbone network to iteratively train the YOLO model until the loss function of the YOLO model is less than or equal to a preset value and the training ends.

2. The system according to claim 1, characterized in that, The controller is further used to calculate the working duration of the motor according to the installation position of the hoisting positioning device, the hoisting target position, and the output power of the motor before the component position is aligned with the hoisting target position, generate a control instruction based on the working duration, and control the operation of the motor based on the control instruction to drive the movement of the lifting tool, so that the target precast component moves and aligns with the hoisting target position.

3. The system according to claim 2, wherein The controller is further used to identify and determine the positioning error distance of the hoisting target position from the video based on the target recognition model. When the positioning error distance is less than or equal to the preset error distance, calculate the working duration of the motor according to the installation position of the hoisting positioning device, the hoisting target position, and the output power of the motor.

4. The system according to any one of claims 1 to 3, characterized in that, The installation position of the hoisting positioning device is determined by the following method: According to the construction design drawings of the prefabricated building, mark the coordinates of the hoisting target position of each precast component, group all the hoisting target positions in the form of regions according to the maximum working range of the hoisting positioning device, find a point in each group of regions so that the sum of the distances to all the hoisting target positions is the shortest, and install the hoisting positioning device at this point. Among them, the installation order of the precast components in each group of regions is determined according to the installation task priority.

5. The system according to any one of claims 1 to 3, characterized in that There are multiple hoisting target positions, each of which is located at the construction site and is marked with at least one physical marker. Each hoisting target position corresponds to a precast component, and all the precast components are used to assemble an assembled building structure.

6. An auxiliary positioning method for hoisting components in an assembled scenario, characterized in that, This method is applied to a hoisting component auxiliary positioning system, which includes a hoisting positioning device and an image sensor. The hoisting positioning device includes a spreader, a motor and a controller, and is pre-installed at a preset installation position at the construction site. This method is executed by the controller and includes the following steps: Control the image sensor to continuously acquire videos of the hoisting target positions at the construction site. Control the operation of the motor to drive the movement of the spreader, where the spreader is used for hoisting the target precast component. At the same time, receive the videos transmitted by the image sensor, and identify and determine the component position of the target precast component and the hoisting target position from the videos based on the target recognition model. When the component position is aligned with the hoisting target position, install the target precast component to perform the installation operation of the next precast component. Among them, the target recognition model is obtained by training the YOLO model for target detection based on sample videos. Among them, the image sensor includes at least one camera and / or depth sensor; the YOLO model includes a first backbone network, a second backbone network, a feature extraction network, and a detection head; the output ends of the first backbone network and the second backbone network are both connected to the input end of the feature extraction network; the training process of the target recognition model includes: Obtain sample videos, where the sample videos include multiple frames of two-dimensional images and multiple frames of depth images collected by the at least one camera and depth sensor. Each frame of two-dimensional image corresponds to a frame of depth image, and both are marked with the component position and the hoisting target position. Input the multiple frames of two-dimensional images into the first backbone network, and input the multiple frames of depth images into the second backbone network to iteratively train the YOLO model until the loss function of the YOLO model is less than or equal to a preset value to end the training.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method for auxiliary positioning of hoisting components in the assembled building scenario according to claim 6.

8. An automatic hoisting device, characterized in that, Including: An image sensor; A hoisting positioning device, where the hoisting positioning device includes a spreader, a motor, and a processor; And a memory for storing a computer program; Among them, the processor is configured to execute the method for auxiliary positioning of hoisting components in the assembled building scenario according to claim 6 by executing the computer program.

Images