Construction site unmanned inspection method and system for dynamically generating calibration target

Abstract

The invention discloses a construction site unmanned inspection method and system for dynamically generating a calibration target, and relates to the field of construction site inspection, and the construction site unmanned inspection method for dynamically generating the calibration target comprises the steps: periodically obtaining the dynamic target information of a construction site; identifying the dynamic target information, and judging whether the content corresponding to the dynamic target information meets the construction standard or not according to the identification result; and under the condition that the construction standard is not met, making a rectification strategy according to associated dynamic target information associated with the dynamic target information so as to generate a rectification scheme, and sending the rectification scheme to a preset terminal. The method has the effect of improving the patrol efficiency of the construction site.

Description

Technical Field

[0001] This application relates to the technical field of construction site inspection, and in particular to a method and system for unmanned construction site inspection that dynamically generates calibration targets. Background Technology

[0002] On-site inspections are a core management tool to ensure project safety, quality, schedule, and cost control. Systematic inspections help to identify and resolve problems in a timely manner, thereby reducing risks.

[0003] Existing construction sites typically have inspector positions. Inspectors patrol designated areas at specified times to maintain construction order. For example, in terms of safety, they inspect personnel protective equipment, edge and opening protection, temporary power supply, and fire-fighting facilities; in terms of quality, they verify material acceptance, construction process implementation, and concealed works records; in terms of progress, they compare planned and actual progress, coordinate resources, and optimize work processes; in terms of environmental protection, they supervise dust control, noise management, and waste sorting and disposal; and in terms of comprehensive management, they ensure compliant construction, contract fulfillment, and effective communication, etc.

[0004] Based on the above inspection content, the inspectors will record the inspection results, issue rectification notices for any problems found, and follow up on the closed loop. However, this process often requires a lot of labor and is time-consuming and laborious. Summary of the Invention

[0005] To improve the efficiency of on-site inspections, this application provides a method and system for unmanned on-site inspections that dynamically generate calibration targets.

[0006] Firstly, this application provides a method for dynamically generating calibration targets for unmanned inspection of construction sites, employing the following technical solution: A method for unmanned inspection of construction sites that dynamically generates calibrated targets, the method comprising: periodically acquiring dynamic target information of the construction site; identifying the dynamic target information and determining whether the content corresponding to the dynamic target information conforms to construction standards based on the identification results; if it does not conform to construction standards, formulating a rectification strategy based on associated dynamic target information related to the dynamic target information to generate a rectification plan, and sending the rectification plan to a preset terminal.

[0007] By adopting the above technical solution, this application can identify dynamic targets at the construction site and determine whether the corresponding dynamic targets meet the construction standards based on the identification results. If they do not meet the construction standards, a rectification strategy is formulated based on the related dynamic targets that are associated with the dynamic targets, and a rectification plan is sent to the preset terminal so that the construction site can rectify the non-compliance with the construction standards more quickly, thereby improving the inspection efficiency of the construction site.

[0008] Optionally, the periodic acquisition of dynamic target information at the construction site includes: dividing the construction site into multiple target areas; marking each target area according to a preset safety level, so that when acquiring dynamic target information of each target area, the patrol duration of each target area is allocated according to its respective safety level; and acquiring dynamic target information of multiple target areas according to preset patrol trajectory points.

[0009] By adopting the above technical solution, this application sets the patrol duration for different target areas based on security levels. Within a cycle, more patrol time is allocated to areas with relatively lower security levels, which can improve the data collection fault tolerance rate in areas with lower security levels, thereby improving the effectiveness of maintaining order at construction sites.

[0010] Optionally, the step of identifying the dynamic target information and determining whether the content corresponding to the dynamic target information conforms to the construction standards based on the identification result includes: when the dynamic target information is identified as construction worker information, identifying the wearing information of one or more protective gears of the construction worker; if the identification result is that the worker is not wearing protective gear or is not fully wearing protective gear, it is determined that the worker does not conform to the construction standards; if the identification result is that the worker is wearing protective gear, it is determined that the worker conforms to the construction standards. By adopting the above technical solutions, construction workers who are not wearing or are not wearing protective gear can be quickly identified, so as to generate rectification plans more quickly and improve the inspection efficiency of construction sites.

[0011] Optionally, in the case of non-compliance with construction standards, the step of formulating a rectification strategy based on associated dynamic target information related to the dynamic target information to generate a rectification plan, and sending the rectification plan to a preset terminal, includes: calculating the optimal prompting path for the corresponding construction personnel based on the associated dynamic target information related to the dynamic target information; transmitting the protective gear wearing prompting information through the terminal on the optimal prompting path to prompt construction personnel who are not wearing or are not fully wearing protective gear.

[0012] By adopting the above technical solution, when generating rectification plans for construction workers who are not wearing or are not fully wearing protective gear, the optimal prompt path can be found quickly, thereby improving rectification efficiency.

[0013] Optionally, it also includes: when calculating the optimal prompt path for the corresponding construction worker, determining whether there is a communication barrier between the communication endpoint of the optimal prompt path and the construction worker; if so, correcting the optimal prompt path according to a preset correction rule.

[0014] Optionally, the step of identifying the dynamic target information and determining whether the construction content corresponding to the dynamic target information meets the construction standards based on the identification result further includes: when the dynamic target information is identified as environmental information of the construction site, identifying the dust concentration information of the construction site; if the identification result is not within the preset dust emission concentration range, it is determined that it does not meet the construction standards; if the identification result is within the preset dust emission concentration range, it is determined that it meets the construction standards.

[0015] Optionally, in the case of non-compliance with construction standards, formulating a rectification strategy based on associated dynamic target information related to the dynamic target information to generate a rectification plan, and sending the rectification plan to a preset terminal, includes: calculating the duration and opening degree of the spray device based on the associated dynamic target information related to the dynamic target information, wherein the spray device is used to reduce the dust concentration at the construction site; generating a control command based on the duration and opening degree of the spray device, and sending the control command to a terminal for controlling the spray device.

[0016] By adopting the above technical solution, when the dust concentration does not meet the construction standards, a rectification strategy is formulated based on the dynamic target information related to the dust concentration to generate a rectification plan, thereby improving the inspection efficiency of the construction site.

[0017] Secondly, this application provides an unmanned inspection system for construction sites that dynamically generates calibration targets, employing the following technical solution: An unmanned inspection system for construction sites that dynamically generates calibrated targets includes: a data acquisition unit for periodically acquiring dynamic target information at the construction site; an identification unit for identifying the dynamic target information and determining whether the content corresponding to the dynamic target information conforms to construction standards based on the identification results; and a strategy unit for formulating rectification strategies based on associated dynamic target information related to the dynamic target information when the construction standards are not met, thereby generating a rectification plan and sending the rectification plan to a preset terminal.

[0018] By adopting the above technical solution, dynamic target information is collected, identified, and judged through the cooperation of the acquisition unit, identification unit, and strategy unit. If the judgment result is that the target does not meet the construction standards, the strategy unit generates a rectification plan to improve the inspection efficiency of the construction site.

[0019] In summary, this application has the following beneficial technical effects: This application can identify dynamic targets at the construction site and determine whether the corresponding dynamic targets meet the construction standards based on the identification results. If they do not meet the construction standards, a rectification strategy is formulated based on the associated dynamic targets that are related to the dynamic targets, so as to form a rectification plan and send it to the preset terminal, so that the construction site can rectify the content that does not meet the construction standards more quickly, thereby improving the inspection efficiency of the construction site. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating an exemplary embodiment of the present application of a method for dynamically generating calibration targets for unmanned inspection of construction sites.

[0021] Figure 2 This is a flowchart illustrating the acquisition of dynamic target information in an exemplary embodiment of this application.

[0022] Figure 3 This is a flowchart illustrating the determination of whether dynamic target information conforms to construction standards, as shown in an exemplary embodiment of this application.

[0023] Figure 4 This is a flowchart illustrating the optimal prompt path generation method in an exemplary embodiment of this application.

[0024] Figure 5 This is a plan view illustrating the optimal prompt path determination method in an exemplary embodiment of this application.

[0025] Figure 6 This is a plan view illustrating an optimal prompt path determination method, as shown in another exemplary embodiment of this application.

[0026] Figure 7 This is a flowchart illustrating a different exemplary embodiment of the present application, showing the determination process for whether dynamic target information conforms to construction standards.

[0027] Figure 8 This is a logic diagram illustrating the control method of the spray device according to an exemplary embodiment of this application.

[0028] Figure 9 This is the frame of an unmanned construction site inspection system that dynamically generates calibration targets, as illustrated in an exemplary embodiment of this application. Detailed Implementation

[0029] The following is in conjunction with the appendix Figure 1-9 This application will be described in further detail.

[0030] This application discloses a method for unmanned inspection of construction sites that dynamically generates calibration targets.

[0031] Reference Figure 1The unmanned inspection method for dynamically generating calibration targets disclosed in this application may include at least the following execution steps: S110. Periodically acquire dynamic target information of the construction site; S120. Identify dynamic target information and determine whether the content corresponding to the dynamic target information meets the construction standards based on the identification results. S130. In cases where construction standards are not met, a rectification strategy is formulated based on the associated dynamic target information that is related to the dynamic target information, in order to generate a rectification plan and send the rectification plan to the preset terminal.

[0032] To illustrate, in this embodiment, the periodic acquisition of dynamic information from the construction site, including image data, can be achieved using aerial drones or pre-installed cameras at fixed locations within the construction site. Furthermore, non-image-related dynamic information can be acquired through data collection using various sensors.

[0033] For example, when using a drone, the shooting time can be set in advance in the drone's control program. For instance, multiple time periods can be set between 8:00 AM and 10:00 AM, with each time period being a cycle. Within a cycle, the drone completes the pre-set aerial photography content.

[0034] For example, when using camera devices, spherical cameras can be used to achieve shooting angles similar to those of aerial drones, with multiple spherical cameras distributed at designated locations on the construction site, such as the four corners of a rectangular area. Similarly, shooting can be conducted according to pre-set shooting schedules.

[0035] Since different devices have different advantages, this application does not limit the devices used to acquire dynamic information at the construction site, but only provides an example.

[0036] Furthermore, dynamic information at the construction site can include one or more of the following: environmental status information, personnel safety information, equipment safety information, and material management information. Environmental status information can include dust concentration at the construction site; personnel safety information can include one or more of the following: information on workers wearing safety helmets, safety ropes, and worker location safety; equipment safety information can include one or more of the following: information on protective netting safety and scaffolding safety; and material management information can include one or more of the following: information on steel management and timber management. Similarly, the above examples are for illustrative purposes only and do not constitute specific limitations.

[0037] Based on the periodically acquired dynamic target information, the dynamic target information is further identified. This identification can be based on image recognition principles, analyzing the content within the captured image.

[0038] To illustrate, after acquiring the captured image, a convolutional neural network (CNN) is used to perform object detection on the image, and the trained model automatically divides the image region and identifies construction workers, construction equipment, buildings, etc.

[0039] For example, the YOLOv7-BEV algorithm can perform object detection directly in the fused bird's-eye view (BEV), significantly improving efficiency and accuracy.

[0040] For example, when identifying construction workers in an image, the detected human-shaped regions are compared with pedestrian features (such as head and limb proportions) in a pre-stored database, and the motion trajectory is used to predict whether it is a real person. If abnormal features are detected (such as people on posters or people on construction site warning signs), the algorithm will optimize the model by adding negative samples to reduce the probability of misjudgment.

[0041] After identifying dynamic target information, this application can also determine whether the content corresponding to the dynamic target information conforms to construction standards based on the identification results. Specifically, taking identified construction workers as an example, image recognition is also used to divide the identified construction workers into multiple regions according to their body structure, such as the upper body and lower body. The upper and lower body are further subdivided into specific body regions. Based on this, the head and waist feature information of the construction workers is found. Then, based on the head and waist feature information, it is determined whether the construction workers are wearing safety helmets and safety harnesses. The determination method can be obtained by comparing with pre-recorded personnel feature images, or by image recognition, such as the identification of safety helmets and safety harnesses. If they are wearing them, it is considered that they conform to construction standards; otherwise, it is considered that they do not conform to construction standards.

[0042] For example, dust concentration can be collected directly from the dust monitoring equipment at the construction site, or it can be collected through a separately installed dust concentration sensor.

[0043] After judging the identification results of the above dynamic target information, if they do not meet the construction standards, the technical solution of this application can formulate a rectification strategy based on the associated dynamic target information that is related to the dynamic target information, generate a rectification plan, and send the rectification plan to the preset terminal.

[0044] It should be noted that the associated dynamic target information can be of the same type as the dynamic target information or not. For example, if a construction worker is not wearing a safety helmet, the associated dynamic target information for that worker could be the dynamic target information of a coworker or construction safety officer, or it could be the dynamic target information of a warning device. Based on this, when the associated dynamic target information is the target information corresponding to a coworker or construction safety officer, the strategy could be to first notify the coworker or construction safety officer, then have the coworker or construction safety officer remind the worker, and finally send a rectification plan generated from this strategy to the coworker's or construction safety officer's mobile phone, walkie-talkie, or other communication device, so that the coworker or construction safety officer can remind the worker. When the associated dynamic target information is the dynamic target information of a warning device, it could indicate that there is a broadcast device near the worker's current location. The strategy could be to directly remind the worker based on this broadcast device, and then send a rectification plan generated from this strategy to the control terminal of the broadcast device.

[0045] According to the above implementation method, this application can identify dynamic targets at the construction site and determine whether the corresponding dynamic targets meet the construction standards based on the identification results. If they do not meet the construction standards, a rectification strategy is formulated based on the associated dynamic targets that are related to the dynamic targets, so as to form a rectification plan and send it to the preset terminal, so that the construction site can rectify the content that does not meet the construction standards more quickly, thereby improving the inspection efficiency of the construction site.

[0046] In some embodiments of this application, based on the above embodiments, refer to Figure 2 To improve the reliability of dynamic target information when periodically acquiring it from the construction site, at least the following execution steps may be included: S210. Divide the construction site into multiple target areas; S220. Mark each target area according to the preset security level, so that when acquiring dynamic target information of each target area, the patrol duration of each target area is allocated according to its respective security level. S230. Based on the preset patrol trajectory points, acquire dynamic target information for multiple target areas respectively.

[0047] Specifically, multiple target areas can be planar or three-dimensional. For example, when the target area is planar, a planar image can be generated based on the overhead view of the construction site, and this image can be divided into multiple target areas, such as building area, adjacent building area, flat area, slope area, well drilling area, etc.

[0048] For example, a three-dimensional area can be achieved through Building Information Modeling (BIM). This involves integrating the entire lifecycle data of all buildings on the construction site using a 3D digital model, collecting building data in advance, and using parametric modeling to achieve information collaboration and sharing across design, construction, and operation stages. The target area is then divided based on the established model.

[0049] Based on this, after dividing the target area, this application can mark each target area according to a preset security level. For example, the security level can be distinguished by a number from 1 to 5, where a larger number indicates a higher security level and fewer corresponding security risks, and vice versa.

[0050] For example, flat areas have fewer safety hazards, so the safety level of such areas can be set to 5.

[0051] For example, areas adjacent to buildings may have more safety hazards; therefore, the safety level for such areas can be set to 3. For example, some areas are adjacent to buildings, slopes, and well drilling areas. Therefore, the safety level of such areas can be set to 1.

[0052] Among them, the rules for determining the safety level of the target area can be written into the program in advance, and the target area can be marked based on the collected construction images.

[0053] Furthermore, this application can establish a mapping relationship between security levels and patrol durations in advance. After determining the security level of a target area, the patrol duration for each target area can be mapped accordingly. For example, if an area has fewer security risks and a higher security level, its patrol duration can be set shorter compared to other target areas with different security levels.

[0054] Based on this, this application can perform patrol work according to the pre-set patrol trajectory points. For example, the security level of area A is 1, the security level of area B is 2, and the patrol trajectory points are from A to B. During the patrol, since the security level of area A is lower than that of area B, the patrol time of area A is longer than that of area B.

[0055] In the above implementation plan, this application sets patrol durations for different target areas based on security levels. Within a cycle, more patrol durations are allocated to areas with relatively lower security levels, which can improve the data collection fault tolerance rate in areas with lower security levels, thereby improving the effectiveness of maintaining order at construction sites.

[0056] In some embodiments of this application, reference is made to Figure 3 The process of identifying dynamic target information and determining whether the content corresponding to the dynamic target information conforms to construction standards based on the identification results can include at least the following steps: S310. When the dynamic target information is identified as construction personnel information, the wearing information of one or more protective gears of the construction personnel shall be identified. S320. If the identification result is that protective gear is not worn or not fully worn, it is determined that the construction standard is not met. S321. If the identification result indicates that protective gear is being worn, it is determined that the construction standard is met.

[0057] Specifically, when the dynamic target information is identified as construction personnel information, the wearing of protective gear by the construction personnel is identified based on the specific method described in the above embodiments. If the identification result is that the construction personnel are not wearing protective gear or are not wearing protective gear completely, it is determined that they do not meet the construction standards.

[0058] It should be noted that "not wearing protective gear" refers to not wearing the designated protective gear in designated areas on the construction site. For example, in flat areas, the designated protective gear is a safety helmet; in high-altitude areas, the designated protective gear is a safety helmet and a safety rope.

[0059] Based on this, the following example steps can be used to illustrate how to confirm the identification results of protective gear wearing information. After obtaining the construction worker information, first determine the area where the construction worker is located. For each area of ​​the construction site, this application can pre-set the protective gear that needs to be worn. Then confirm the protective gear that needs to be worn in the area. Compare the confirmation result with the protective gear wearing information of the construction worker identified by the construction worker information to obtain the identification result.

[0060] To illustrate the identification steps described above, for example, if the area where the construction worker is located is a flat area where no construction work is required, then the only protective gear needed in this area is a safety helmet. Therefore, if the construction worker is not wearing a safety helmet in this area, it can be identified that the construction worker is not wearing protective gear.

[0061] For example, if the area where construction workers are located is a special material handling area, where special materials could be steel, paint, etc., the steel used on construction sites often has a rough surface, and handling it barehanded can easily cut hands and cause diseases such as tetanus. Similarly, some paints may be corrosive, and handling them barehanded can also lead to hand infections. Therefore, when construction workers are carrying out handling work in this area, in addition to wearing safety helmets, they also need to wear gloves. Thus, there are four scenarios regarding the protective gear worn by construction workers in this area: the first is wearing both a safety helmet and gloves; the second is not wearing a safety helmet but wearing gloves; the third is wearing a safety helmet but not gloves; and the fourth is not wearing a safety helmet and not wearing gloves. Only the first scenario indicates that the construction worker is wearing protective gear; the second and third scenarios indicate that the construction worker is not fully wearing protective gear; and the fourth scenario indicates that the construction worker is not wearing protective gear.

[0062] Through the above-described implementation methods, this application can more accurately obtain the personal protective equipment (PPE) status of construction workers by identifying different protective gear wearing conditions in different areas. Simultaneously, it can quickly identify construction workers who are not wearing or are not fully wearing protective gear, enabling faster generation of rectification plans and thus improving the efficiency of on-site inspections.

[0063] In some embodiments of this application, reference is made to Figure 4 In cases where construction standards are not met, a rectification strategy is formulated based on the associated dynamic target information to generate a rectification plan, which is then sent to a preset terminal. This process may include at least the following execution steps: S410. Based on the associated dynamic target information that is related to the dynamic target information, calculate the optimal prompt path for the corresponding construction personnel. S420. The protective gear wearing reminder information is transmitted through the terminal on the optimal reminder path to remind construction workers who are not wearing or are not wearing protective gear completely.

[0064] Specifically, if the identified dynamic target information is the dynamic information of construction workers, the system searches for communication devices around the workers who are not wearing protective gear or are not fully wearing protective gear. These devices include walkie-talkies located at the construction site, registered telephones, and fixed-location warning devices. The device with the shortest distance to the worker within the search range is selected as the optimal alert device, and a protective gear wearing alert is sent to this device to remind the worker who is not wearing or is not fully wearing protective gear.

[0065] Exemplary illustration, see reference Figure 5The system searches for communication devices around a construction worker O who is not wearing protective gear or whose protective gear is not fully worn, with the search results including coworker phone numbers A, B, and C. The distance between worker O and coworker phone number A is denoted by OA, the distance between worker O and coworker phone number B by OB, and the distance between worker O and coworker phone number C by OC, where OA is greater than OB, and OB is greater than OC. Therefore, coworker phone number C is determined as the optimal notification device. The optimal notification path can be represented as the communication path between the protective gear wearing notification device and the optimal notification device.

[0066] It should be noted that if construction workers who are not wearing protective gear or are not fully wearing protective gear have their own communication devices that can provide alerts, they can be directly selected as the best alerting device without searching for them.

[0067] Based on the optimal prompt path calculated in the above embodiments, this application can further determine whether there is a communication barrier between the communication endpoint of the optimal prompt path and the construction worker; if so, the optimal prompt path is corrected according to a preset correction rule.

[0068] Specifically, to further illustrate with the above example, after determining coworker phone C as the optimal notification device, communication is established between the protective gear wearing notification information sending device and the optimal notification device. If communication fails to establish successfully, such as due to coworker phone C being switched off or having unpaid bills, the optimal notification path is corrected according to preset correction rules. For example, coworker phone C is excluded during the communication device search, and coworker phone B is selected as the new optimal notification device.

[0069] In addition, the communication barrier in this application can be a building barrier. This application can also identify building barriers, as illustrated above. Figure 6 If there are building obstacles, such as building walls or gaps, between the worker's phone C and the construction worker O, preventing the worker carrying the phone C from transmitting the protective gear wearing instructions to the construction worker O, then the optimal instruction path still needs to be corrected according to the preset correction rules.

[0070] In the above implementation method, when generating rectification plans for construction workers who are not wearing or are not fully wearing protective gear, the optimal prompt path can be found quickly, thereby improving rectification efficiency.

[0071] In some embodiments of this application, reference is made to Figure 7 The process of identifying dynamic target information and determining whether the construction content corresponding to the dynamic target information meets the construction standards based on the identification results may include at least the following execution steps: S710. When dynamic target information is identified as environmental information of the construction site, the dust concentration information of the construction site shall be identified. S720. If the identification result is not within the preset dust emission concentration range, it is determined that it does not meet the construction standards. S721. If the identification result is within the preset dust emission concentration range, it is determined to meet the construction standards.

[0072] Specifically, the dust concentration collection method of this application can be implemented based on the content described in the above embodiments, that is, obtaining the dust concentration from the environmental monitoring equipment set up at the construction site. One method is to directly establish communication with the environmental monitoring equipment, and another method is to capture images from the display module on the monitoring equipment. Of course, a separate sensor can also be set up to obtain the dust concentration.

[0073] The dust emission concentration range can include both the range corresponding to PM2.5 and the range corresponding to PM10. When both meet the emission standards, it is determined to comply with the construction standards; otherwise, it is determined to fail to comply with the construction standards.

[0074] Reference Figure 8 When obtaining the aforementioned dust concentration, the following steps may also be included: S810. Based on the associated dynamic target information that is related to the dynamic target information, calculate the duration and opening degree of the spray device. The spray device is used to reduce the dust concentration at the construction site. S820: Generate control commands based on the duration and opening degree of the spray device, and send the control commands to the terminal used to control the spray device.

[0075] Specifically, the associated dynamic target information related to dust concentration may include wind direction information and wind speed information. When the wind speed is less than a preset wind speed threshold, the spray nozzle of the spray device is controlled to face the side opposite to the wind direction and sprays according to the first opening degree. When the wind speed is greater than or equal to the preset wind speed threshold, the spray nozzle of the spray device is controlled to face the side opposite to the wind direction and sprays according to the second opening degree, wherein the second opening degree is greater than the first opening degree.

[0076] It should be noted that the spraying device may also include multiple spray nozzles, which can be arranged around the construction site. When the wind speed is less than a preset wind speed threshold, all spray nozzles of the spraying device are controlled to spray at the first opening. When the wind speed is greater than or equal to the preset wind speed threshold, the spraying device is controlled to spray at the second opening of the spray nozzle on the side opposite to the wind direction, while the other spray nozzles spray at the first opening. The second opening is greater than the first opening.

[0077] Furthermore, based on periodically acquired dust concentration information, a shutdown command for the spray device is generated when the dust concentration meets construction standards. This application can also record the generation process of spray device control commands over a period of time, referred to here as a historical time period. The dust concentration information at the time of each start and stop command generation within this historical time period is recorded. A function model is established based on the duration between the start and stop commands and the changes in dust concentration. Data after the historical time period is used as training data for this function model until a preset number of training iterations are reached. After this, the generation of the spray device shutdown command is no longer determined by periodically acquired dust concentration, but rather by the current dust concentration when the spray device generates an start command, determining the spray device's operating duration accordingly.

[0078] The application itself can obtain the associated dynamic target information related to dust concentration in the trajectory of dust concentration change, so as to calculate the opening duration and opening degree of the spray device, thereby realizing the automatic control of the spray device.

[0079] In the above implementation, when the dust concentration does not meet the construction standards, a rectification strategy is formulated based on the dynamic target information related to the dust concentration to generate a rectification plan, thereby improving the inspection efficiency of the construction site.

[0080] The following describes an implementation system embodiment of this application, which can be used to execute the unmanned construction site inspection method for dynamically generating calibration targets as described in the above embodiments of this application. For details not disclosed in the implementation system embodiments of this application, please refer to the above embodiments of the unmanned construction site inspection method for dynamically generating calibration targets.

[0081] This application also discloses an unmanned inspection system for construction sites that dynamically generates calibration targets.

[0082] Reference Figure 9 The unmanned inspection system for construction sites that dynamically generates calibrated targets includes: a data acquisition unit 910, used to periodically acquire dynamic target information at the construction site; an identification unit 920, used to identify the dynamic target information and determine whether the content corresponding to the dynamic target information conforms to the construction standards based on the identification results; and a strategy unit 930, used to formulate rectification strategies based on related dynamic target information that is associated with the dynamic target information when the construction standards are not met, in order to generate a rectification plan and send the rectification plan to a preset terminal.

[0083] There is a communication relationship between the acquisition unit 910, the identification unit 920, and the strategy unit 930 to enable data transmission between the units. This can be implemented based on a wired transmission module or a wireless transmission module.

[0084] In some embodiments of this application, based on the aforementioned scheme, the acquisition unit 910 is further configured to: periodically acquire dynamic target information of the construction site, including: dividing the construction site into multiple target areas; marking each target area according to a preset safety level, so as to allocate the patrol duration of each target area according to its respective safety level when acquiring dynamic target information of each target area; and acquiring dynamic target information of multiple target areas according to preset patrol trajectory points.

[0085] In some embodiments of this application, based on the aforementioned scheme, the identification unit 920 is further configured to: identify dynamic target information and determine whether the content corresponding to the dynamic target information conforms to the construction standards based on the identification result, including: when the dynamic target information is identified as construction personnel information, identifying the wearing information of one or more protective gears of the construction personnel; if the identification result is that the personnel are not wearing protective gear or are not fully wearing protective gear, it is determined that the construction standards are not met; if the identification result is that the personnel are wearing protective gear, it is determined that the construction standards are met.

[0086] In some embodiments of this application, based on the aforementioned scheme, the strategy unit 930 is further configured to: formulate a rectification strategy based on the associated dynamic target information that is related to the dynamic target information when the construction standards are not met, generate a rectification plan, and send the rectification plan to a preset terminal, including: calculating the optimal prompting path for the corresponding construction personnel based on the associated dynamic target information that is related to the dynamic target information; and transmitting the prompting information for wearing protective gear through the terminal on the optimal prompting path to prompt the construction personnel who are not wearing or are not fully wearing protective gear.

[0087] In some embodiments of this application, based on the aforementioned scheme, the strategy unit 930 is further configured to: further include: when calculating the optimal prompt path for the corresponding construction worker, determine whether there is a communication barrier between the communication endpoint of the optimal prompt path and the construction worker; if there is, correct the optimal prompt path according to a preset correction rule.

[0088] In some embodiments of this application, based on the aforementioned scheme, the identification unit 920 is further configured to: identify dynamic target information and determine whether the construction content corresponding to the dynamic target information meets the construction standards based on the identification result; and further includes: when the dynamic target information is identified as environmental protection information of the construction site, identifying the dust concentration information of the construction site; if the identification result is not within the preset dust emission concentration range, it is determined that it does not meet the construction standards; if the identification result is within the preset dust emission concentration range, it is determined that it meets the construction standards.

[0089] In some embodiments of this application, based on the aforementioned scheme, the strategy unit 930 is further configured to: formulate a rectification strategy based on the associated dynamic target information that is related to the dynamic target information when the construction standards are not met, so as to generate a rectification plan and send the rectification plan to a preset terminal, including: calculating the duration and opening degree of the spray device based on the associated dynamic target information that is related to the dynamic target information, wherein the spray device is used to reduce the dust concentration at the construction site. Control commands are generated based on the duration and degree of operation of the spray device, and then sent to a terminal for controlling the spray device.

[0090] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for unmanned inspection of construction sites that dynamically generates calibration targets, characterized in that: The method includes: Periodically acquire dynamic target information from the construction site; The dynamic target information is identified, and the identification result is used to determine whether the content corresponding to the dynamic target information meets the construction standards. If the construction standards are not met, a rectification strategy is formulated based on the associated dynamic target information that is related to the dynamic target information to generate a rectification plan, and the rectification plan is sent to a preset terminal.

2. The method for unmanned inspection of construction sites by dynamically generating calibration targets according to claim 1, characterized in that: The periodic acquisition of dynamic target information at the construction site includes: The construction site was divided into multiple target areas; Each target area is marked according to a preset security level, so that when dynamic target information of each target area is obtained, the patrol duration of each target area is allocated according to its respective security level. Based on the preset patrol trajectory points, dynamic target information for multiple target areas is acquired.

3. The method for unmanned inspection of construction sites by dynamically generating calibration targets according to claim 1, characterized in that: The step of identifying the dynamic target information and determining whether the content corresponding to the dynamic target information conforms to the construction standards based on the identification result includes: When the dynamic target information is identified as construction worker information, the wearing information of one or more protective gears of the construction worker is identified; If the identification result is that protective gear is not being worn or is not being worn completely, it is determined that the construction does not meet the standards. If the identification result shows that protective gear is being worn, it is determined that the construction standards are met.

4. The method for unmanned inspection of construction sites by dynamically generating calibration targets according to claim 3, characterized in that: In cases where construction standards are not met, a rectification strategy is formulated based on associated dynamic target information related to the dynamic target information to generate a rectification plan, and the rectification plan is sent to a preset terminal, including: Based on the associated dynamic target information that is related to the dynamic target information, the optimal prompt path for the corresponding construction personnel is calculated; The protective gear wearing reminder information is transmitted through the terminal on the optimal reminder path to remind construction workers who are not wearing or are not wearing protective gear completely.

5. The method for unmanned inspection of construction sites by dynamically generating calibration targets according to claim 4, characterized in that: Also includes: When calculating the optimal prompt path for the corresponding construction worker, it is determined whether there is a communication barrier between the communication endpoint of the optimal prompt path and the construction worker; if so, the optimal prompt path is corrected according to the preset correction rules.

6. The method for unmanned inspection of construction sites by dynamically generating calibration targets according to claim 1, characterized in that: The step of identifying the dynamic target information and determining whether the construction content corresponding to the dynamic target information meets the construction standards based on the identification result also includes: When the dynamic target information is identified as environmental information of the construction site, the dust concentration information of the construction site is identified. If the identification result is outside the preset dust emission concentration range, it is determined that the construction does not meet the standards. If the identification result is within the preset dust emission concentration range, it is determined to meet the construction standards.

7. The method for unmanned inspection of construction sites by dynamically generating calibration targets according to claim 6, characterized in that: In cases where construction standards are not met, a rectification strategy is formulated based on associated dynamic target information related to the dynamic target information to generate a rectification plan, and the rectification plan is sent to a preset terminal, including: Based on the associated dynamic target information that is related to the dynamic target information, the duration and opening degree of the spray device are calculated. The spray device is used to reduce the dust concentration at the construction site. Control commands are generated based on the duration and degree of operation of the spray device, and the control commands are sent to a terminal for controlling the spray device.

8. A construction site unmanned inspection system that dynamically generates calibration targets, characterized in that: include: The data acquisition unit is used to periodically acquire dynamic target information at the construction site; The identification unit is used to identify the dynamic target information and determine whether the content corresponding to the dynamic target information conforms to the construction standards based on the identification result. The strategy unit is used to formulate a rectification strategy based on the associated dynamic target information that is related to the dynamic target information when the construction standards are not met, so as to generate a rectification plan and send the rectification plan to a preset terminal.

Images