An underground garage operation and maintenance task management system and method based on an internet of things

Abstract

The application provides an underground garage operation and maintenance task management system and method based on the Internet of Things, and relates to the technical field of garage management. The system comprises: a blind area identification unit, which is used to construct an initial digital twin model of the garage based on the structure data of the underground garage, collect state data of the underground garage, and identify blind area regions of the underground garage in combination with a grid comparison algorithm; a path planning unit, which is used to generate an optimal inspection path of the management personnel by using a path planning algorithm based on an initial inspection point of the operation and maintenance personnel and a preset receiver in the underground garage according to the identified blind area regions, and determine the receiver to be connected in combination with the optimal inspection path; and an inspection state monitoring unit, which is used to execute an inspection operation and maintenance task based on the optimal inspection path, determine the inspection state of the management personnel according to the receiver to be connected, and record the inspection state, so as to realize the monitoring of the operation and maintenance task of the underground garage. The application greatly improves the operation and maintenance efficiency, data accuracy and management transparency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of garage management, in particular, to an underground garage operation and maintenance task management system and method based on Internet of Things. BACKGROUND

[0002] The underground garage is an important underground space for providing safe parking and management for vehicles, and the operation and maintenance management level thereof directly affects the safety of vehicles, the reliability of facilities and the efficiency of emergency response. In order to protect the safety of vehicles and personnel, ensure the stable operation of various facilities such as fire fighting, drainage, ventilation and lighting, and achieve rapid response in emergency, it is necessary to comprehensively and systematically manage the operation and maintenance of the underground garage.

[0003] At present, the operation and maintenance management of the underground garage is mainly based on manual patrol, static monitoring or periodic maintenance, which has significant problems such as incomplete coverage, low patrol efficiency, and untraceable data. Especially in large or complex underground garage scenes, due to factors such as building obstruction and limited view angle, blind areas are easily formed, which leads to the fact that dead angles cannot be discovered or handled in time. The deployment of traditional monitoring equipment lacks dynamic updating capability, and the patrol path of the operation and maintenance personnel is often determined by experience, which lacks system optimization and is easy to miss key areas or cause waste of human resources. In addition, the patrol results lack standardized recording means, making it difficult to grasp the patrol state in real time, and also difficult to realize automatic monitoring and historical tracing.

[0004] In view of the problems in the related art, no effective solution has been proposed so far. SUMMARY

[0005] Therefore, the present application provides an underground garage operation and maintenance task management system and method based on Internet of Things to solve the above-mentioned problems.

[0006] In order to solve the above problems, the specific technical solutions adopted by the present application are as follows:

[0007] According to an aspect of the present application, an underground garage operation and maintenance task management system based on Internet of Things is provided, which comprises:

[0008] A blind area identification unit is configured to construct an initial digital twin model of the garage based on the structure data of the underground garage, collect state data of the underground garage, and identify blind area regions of the underground garage in combination with a grid comparison algorithm;

[0009] A path planning unit is configured to generate an optimal patrol path of the management personnel based on the initial patrol point of the operation and maintenance personnel and the preset receiver in the underground garage by using a path planning algorithm according to the identified blind area regions, and determine the receiver to be connected in combination with the optimal patrol path;

[0010] The inspection state monitoring unit is used for performing an inspection operation and maintenance task based on an optimal inspection path, determining and recording an inspection state of a manager according to a receiver to be connected, so as to realize monitoring of an underground garage operation and maintenance task.

[0011] Preferably, the blind area identification unit comprises:

[0012] The initial model construction module is used for acquiring point cloud data of the underground garage by the laser radar and pre-processing, and constructing an initial digital twin model of the garage based on the pre-processed point cloud data.

[0013] The real-time model construction module is used for acquiring state data of the underground garage based on a preset camera array in the underground garage, and constructing a real-time digital twin model of the garage by using a stereo vision matching algorithm of parallax depth according to the state data.

[0014] The grid comparison module is used for respectively performing grid division on the initial digital twin model of the garage and the real-time digital twin model of the garage, and identifying a blind area of the underground garage by combining a grid comparison method.

[0015] Preferably, the real-time model construction module comprises:

[0016] The garage image collected by the camera array is subjected to color space conversion, and a structured light stripe region is extracted based on a brightness channel threshold value.

[0017] In the extracted structured light stripe region, a structured light line is extracted based on a multi-level threshold segmentation of a hue channel, and a center line of each structured light line is extracted.

[0018] According to a preset hierarchical order, stereo matching is performed based on the center line of each structured light line, three-dimensional coordinates of a matching point are determined by calculating a parallax of the matching point, point cloud data is generated according to the three-dimensional coordinates of the matching point, and a real-time digital twin model of the garage is constructed.

[0019] Preferably, the method for constructing a real-time digital twin model of an underground garage comprises:

[0020] Based on an illumination intensity in the underground garage, a hue channel threshold value of each structured light is set.

[0021] According to the hue channel threshold value of each structured light, a multi-level threshold segmentation method is adopted to separate a color structured light stripe mask in the hue channel.

[0022] The morphological processing is performed on each structured light stripe mask, the initial geometric center line of each structured light line is extracted, and the initial geometric center line is fitted by using a line segment fitting to obtain the final center line of each structured light line.

[0023] Preferably, the garage initial digital twin model and the garage real-time digital twin model are respectively meshed; and combined with the grid comparison method, the blind area region of the underground garage is identified, including:

[0024] The garage initial digital twin model and the garage real-time digital twin model are aligned in coordinates, and the garage initial digital twin model and the garage real-time digital twin model are respectively uniformly meshed in a voxel grid;

[0025] The garage real-time digital twin model after the voxel grid division is mapped into the garage initial digital twin model after the voxel grid division;

[0026] Based on the mapped garage initial digital twin model, the overlap degree of each grid is calculated, and the grid with an overlap degree lower than a preset threshold is screened out;

[0027] The screened grid is marked, and the blind area region of the underground garage is determined based on the marked grid.

[0028] Preferably, the path planning unit comprises:

[0029] The coverage relationship establishment module is configured to determine the center points of all blind area regions by using a spatial clustering algorithm based on the identified blind area region, and establish a coverage relationship between the receivers and the blind area center points;

[0030] The key node screening module is configured to screen out a minimum receiver set capable of covering all blind area center points as key nodes according to the coverage relationship;

[0031] The inspection path generation module is configured to generate a weighted optimal inspection path by using an improved ant colony algorithm in combination with the patrol initial point and the key nodes;

[0032] The receiver determination module is configured to determine a set of receivers to be connected according to a sequence of key nodes passed by the optimal inspection path.

[0033] Preferably, the coverage relationship establishment module is configured to determine the center points of all blind area regions by using a weighted centroid method based on the identified blind area region, and establish a coverage relationship between the receivers and the blind area center points, including:

[0034] The center point coordinates of each blind area region are calculated by using a weighted centroid method according to the point cloud data of the blind area region;

[0035] According to the position and signal characteristics of the receiver, a receiver coverage model is established, and a spatial index is used to accelerate the query and establish the coverage relationship between the receiver and the center point of the blind area.

[0036] Preferably, the receiver coverage model is established according to the position and signal characteristics of the receiver, and the coverage relationship between the receiver and the center point of the blind area is established by using a spatial index to accelerate the query.

[0037] The three-dimensional coordinates and signal coverage ranges of all receivers are obtained, and a spatial representation of the receiver coverage model is constructed.

[0038] The three-dimensional coordinates of all blind area center points are loaded into a preset spatial index structure, and a data index for fast neighbor query is constructed.

[0039] The coverage model of each receiver is queried using the spatial index, and the blind area center points intersecting the coverage range of the receiver are filtered out, and the coverage relationship between the receiver and the blind area center point is generated according to the query result.

[0040] Preferably, the inspection state monitoring unit comprises:

[0041] The signal connection module is used to establish signal connection between the transmitter on the administrator and the receiver to be connected when the administrator performs the inspection operation task based on the optimal inspection path and inspects the signal coverage range of the receiver to be connected.

[0042] The inspection record module is used to record the connection time and receiver number of the transmitter and the receiver to be connected, and generate corresponding inspection record entries.

[0043] The inspection state judgment module is used to compare the inspection record entries with the set of receivers to be connected, to determine whether the current inspection of the administrator covers all blind areas, and to obtain the inspection result.

[0044] According to another aspect of the present application, a method for managing underground garage operation tasks based on Internet of Things is provided, comprising the following steps:

[0045] S1, constructing a garage initial digital twin model based on the structure data of the underground garage, collecting state data of the underground garage, and identifying blind area regions of the underground garage by combining a grid comparison algorithm;

[0046] S2, according to the identified blind area regions, based on the patrol starting point of the operation and management personnel and the preset receiver in the underground garage, using a path planning algorithm to generate an optimal inspection path for the administrator, and determining the receiver to be connected in combination with the optimal inspection path.

[0047] S3, based on the optimal inspection path, the inspection operation task is executed, the management personnel inspection state is determined according to the receiver to be connected, and the record is made, so as to realize the monitoring of the underground garage operation task.

[0048] The beneficial effects of the present application are:

[0049] 1. The present application realizes the full automation process of personnel inspection task closed-loop management from the detection of the sensing dead angle, can construct a digital twin model based on structure data and state information, accurately identify the blind area, reasonably arrange the inspection route through the optimal path planning algorithm, dynamically determine the task node, and finally record the management personnel operation in the inspection execution stage, ensure the full coverage and traceability of the task, greatly improve the operation efficiency, data accuracy and management transparency.

[0050] 2. The present application constructs a high-precision initial digital twin model through laser radar point cloud, can fully reflect the garage space structure, dynamically generates a real-time digital twin model based on parallax depth by combining stereo vision matching and multi-level threshold extraction method, and accurately identifies the blind area caused by shielding, structure change or light difference through voxel grid division and alignment comparison, not only realizes high-resolution identification of the blind area, but also maintains continuous monitoring of the invisible area in complex and dynamic environment.

[0051] 3. The present application generates a weighted optimal inspection path considering distance, energy consumption and priority by combining the patrol starting point of the management personnel and using the improved ant colony algorithm, and determines the receiver set to be connected according to the key nodes passed in the path, not only greatly compresses the inspection path length and time cost, but also guarantees the integrity of blind area coverage and the pertinence of receiver management, and provides efficient and reliable decision support for inspection scheduling and resource allocation in intelligent garage. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings. In the drawings:

[0053] Figure 1 is a principle block diagram of an underground garage operation task management system based on Internet of Things according to an embodiment of the present application;

[0054] Figure 2 is a flowchart of an underground garage operation task management method based on Internet of Things according to an embodiment of the present application.

[0055] Fig. 1 is a schematic diagram of a system according to an embodiment of the present application.

[0056] 1, blind area identification unit; 101, initial model construction module; 102, real-time model construction module; 103, grid comparison module; 2, path planning unit; 201, coverage relationship establishment module; 202, key node screening module; 203, inspection path generation module; 204, receiver determination module; 3, inspection state monitoring unit; 301, signal connection module; 302, inspection record module; 303, inspection state judgment module. DETAILED DESCRIPTION

[0057] In order for those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0058] According to an embodiment of the present application, an underground garage operation and maintenance task management system and method based on Internet of Things are provided.

[0059] The present application will be further described in conjunction with the drawings and specific embodiments. Figure 1 As shown in the drawings, according to an embodiment of the present application, an underground garage operation and maintenance task management system based on Internet of Things is provided, which comprises:

[0060] A blind area identification unit 1 is used to construct an initial digital twin model of the garage based on the structural data of the underground garage, collect state data of the underground garage, and identify blind area regions of the underground garage in combination with a grid comparison algorithm;

[0061] As a preferred embodiment, the blind area identification unit 1 comprises:

[0062] An initial model construction module 101 is used to obtain point cloud data of the underground garage by laser radar and perform preprocessing, and construct an initial digital twin model of the garage based on the preprocessed point cloud data;

[0063] It should be noted that the preprocessing of the point cloud data includes denoising and filtering processing; denoising is to remove random noise, error points and external interference in the point cloud data; filtering is to further process abnormal points and non-target regions in the point cloud data, and is commonly used to remove ground points, local details, and enhance the performance of main target objects.

[0064] The real-time model construction module 102 is configured to acquire state data of the underground garage based on a preset camera array of the underground garage, and construct a real-time digital twin model of the garage based on the state data and a stereo vision matching algorithm of parallax depth.

[0065] As a preferred embodiment, the real-time model construction module 102 is configured to acquire state data of the underground garage based on a preset camera array of the underground garage, and construct a real-time digital twin model of the garage based on the state data and a stereo vision matching algorithm of parallax depth.

[0066] The garage image collected by the camera array is subjected to color space conversion, and a structured light stripe region is extracted based on a luminance channel threshold.

[0067] It should be noted that the camera array adopts a binocular camera. Specifically, a set of binocular cameras (baseline length 0.5 m) can be installed every 10 m on the top of the underground garage to ensure that the field of view covers all parking spaces and passages.

[0068] The color space conversion usually refers to conversion from RGB to HSV (hue H, saturation S, and luminance V) or YUV space. When extracting the structured light stripe region, the luminance (V or Y channel) is the most stable reference channel, which can effectively separate the background and the structured light with high contrast. Then, an adaptive or fixed threshold is set based on the luminance channel to extract the structured light irradiation region with high luminance. The region usually exhibits regular and uniform high-brightness stripes.

[0069] In the extracted structured light stripe region, the structured light lines are extracted based on multi-level threshold segmentation of the hue channel, and the center lines of the structured light lines are extracted.

[0070] As a preferred embodiment, the structured light lines are extracted based on multi-level threshold segmentation of the hue channel in the extracted structured light stripe region, and the center lines of the structured light lines are extracted.

[0071] The hue channel threshold of each structured light is set based on the illumination intensity in the underground garage.

[0072] Specifically, the illumination intensity (usually 50-500 lux) of the underground garage changes the color temperature distribution of the image collected by the camera. Under low illumination, the increase in sensor noise causes the hue channel value to be compressed (for example, the hue value of the blue stripe may be shifted from the standard 180° to 160°); under high illumination, the hue value of the overexposed region diffuses (for example, the green stripe may be diffused from 120° to 100°-140°).

[0073] Wherein, the fixed threshold value will cause stripe missing or false detection in the light mutation area (such as the garage entrance). The dynamic threshold value can ensure the stable extraction under different light conditions by adjusting the hue segmentation range automatically through real-time sensing of the ambient light intensity. When setting the dynamic threshold value, the linear / non-linear relationship between the light intensity E and the hue threshold value T needs to be established. For example:

[0074] Blue stripe: T low = 180°-0.1E, T high = 180°+0.05E;

[0075] Green stripe: T low = 120°-0.08E, T high = 120°+0.03E;

[0076] T low represents the minimum value of the hue threshold value, T high represents the minimum value of the hue threshold value, and finally the light sensors are evenly distributed on the top of the garage (such as one every 20m), and the multi-point average value is used as the current light intensity input to reduce the interference of local shadows.

[0077] According to the hue channel threshold value of each structured light, a multi-level threshold segmentation method is used to separate each color structured light stripe mask in the hue channel;

[0078] It should be noted that when using the multi-level threshold segmentation method to separate each color structured light stripe mask in the hue channel, the hue channel needs to be divided into multiple subintervals, and the threshold value is set for different color structured light (such as blue and green double color). For example:

[0079] Blue stripe: main interval 160°-200°, auxiliary interval 150°-210° (for low contrast area).

[0080] Green stripe: main interval 100°-140°, auxiliary interval 90°-150°.

[0081] For the area with uneven light (such as near the column), a sliding window (such as 11x11 pixels) is used to calculate the local hue average, and the threshold value range is dynamically corrected to avoid excessive segmentation or under-segmentation caused by the global threshold value.

[0082] For each type of structured light stripe mask, morphological processing is performed to extract the initial geometric center line of each structured light line, and the initial geometric center line is fitted by using line segment fitting to obtain the final center line of each structured light line.

[0083] It should be noted that the morphological processing of each type of structured light stripe mask includes thinning processing and breakpoint repair processing. The thinning processing is used to compress the wider structured light stripe into a one-pixel-wide thin line to obtain an image expression closer to the actual light beam path. The breakpoint repair processing is used to bridge the interruptions in the stripe caused by noise, occlusion or imaging distortion, so that the extracted light rays are coherent and continuous. Through a thinning algorithm (such as centerline extraction or image skeletonization), the initial geometric skeleton of the structured light stripe in the image can be obtained, thereby providing a structural basis for subsequent geometric fitting.

[0084] Based on the center line of each structured light line, stereo matching is performed according to a preset hierarchical order, the three-dimensional coordinates of the matching points are determined by calculating the disparity of the matching points, the point cloud data is generated according to the three-dimensional coordinates of the matching points, and the real-time digital twin model of the garage is constructed.

[0085] It should be noted that the camera array is arranged using a binocular camera, so each set of binocular cameras has a fixed left and right viewing angle level. In the extracted structured light center line, matching needs to be performed in the corresponding order between the left and right image pairs to ensure spatial geometric consistency.

[0086] In the binocular image, the center point of the same structured light line will have a pixel position offset in the left and right views, which is called disparity. The disparity is inversely proportional to the depth, that is, the greater the disparity, the closer the object. Therefore, when calculating the disparity of the matching points, a point on the center line of a certain structured light in the left image needs to be found, and then a matching point on the corresponding light line in the right image is searched (which can be based on neighborhood similarity, structured light line features, etc.), and then the pixel offset between the two points is calculated, that is, the disparity value is obtained.

[0087] In addition, when calculating the three-dimensional coordinates according to the disparity, the calculation formula is:

[0088] Z = (f B) / d;

[0089] In the formula, Z represents the depth coordinate (the distance from the camera); f represents the focal length of the camera; B represents the baseline length between the binocular cameras; and d represents the disparity of the matching points.

[0090] Using Z and the pixel coordinates in the image, the corresponding spatial coordinates can be further calculated, that is, the two-dimensional points in the image are converted into three-dimensional space points through stereo geometric projection.

[0091] The grid comparison module 103 is used for grid division of the initial digital twin model of the garage and the real-time digital twin model of the garage, respectively; and in combination with the grid comparison method, the blind area of the underground garage is identified.

[0092] As a preferred embodiment, the garage initial digital twin model and the garage real-time digital twin model are respectively meshed; and combined with the grid comparison method, the blind area region of the underground garage is identified, which includes:

[0093] The garage initial digital twin model and the garage real-time digital twin model are aligned in coordinates, and the garage initial digital twin model and the garage real-time digital twin model are respectively uniformly meshed in voxels;

[0094] It should be noted that in the two digital twin models, the initial model is the original and designed state of the garage, and the real-time model is the current actual state of the garage. In order to compare, it is necessary to first ensure that the two are aligned in space, which is usually done through rigid transformation (for example, translation, rotation) to ensure that they share a unified coordinate system.

[0095] Once the models are aligned, the two models need to be meshed in voxels respectively; voxel is the basic unit of three-dimensional space, similar to pixel in two-dimensional image; meshing divides the three-dimensional space of the garage into a number of small voxel units, each unit representing a certain area or volume of the garage.

[0096] The garage real-time digital twin model after voxel meshing is mapped into the garage initial digital twin model after voxel meshing;

[0097] Specifically, when mapping the grid of the real-time model to the grid of the initial model, the voxel position of the real-time data needs to be mapped to the corresponding voxel area of the initial garage model through the conversion relationship after coordinate alignment; so that each voxel of the real-time model can be associated with the corresponding position of the initial model, thereby providing a basis for subsequent overlap calculation and blind area identification.

[0098] Based on the mapped garage initial digital twin model, the overlap of each grid is calculated respectively, and the grid with an overlap lower than a preset threshold is screened out;

[0099] The screened grid is marked, and the blind area region of the underground garage is determined based on the marked grid.

[0100] Specifically, the overlap of each voxel grid refers to the proportion of the intersection area (or volume) of the corresponding voxel area in the real-time model and the initial model to the total area (or volume). Then by calculating the overlap of each grid, the coincidence degree of a certain area of the real-time model with the initial model can be evaluated. In comparison with the preset threshold, according to the comparison result, the area is determined as the blind area standard. For example, if the overlap is less than a certain set value (such as 30%), the grid area is considered as a blind area.

[0101] The path planning unit 2 is configured to generate an optimal inspection path for the maintenance personnel based on the identified blind area region, the initial inspection point of the maintenance personnel, and the preset receivers in the underground garage, and determine the receivers to be connected based on the optimal inspection path.

[0102] As a preferred embodiment, the path planning unit 2 comprises:

[0103] The coverage relationship establishing module 201 is configured to determine the center point of each blind area region based on the identified blind area region, and establish the coverage relationship between the receivers and the center points of the blind area regions by using a spatial clustering algorithm.

[0104] As a preferred embodiment, the method of determining the center point of each blind area region based on the identified blind area region and establishing the coverage relationship between the receivers and the center points of the blind area regions comprises:

[0105] According to the point cloud data of the blind area region, the center point coordinates of each blind area region are calculated by using the weighted centroid method.

[0106] It should be noted that before the center point coordinates of each blind area region are calculated by using the weighted centroid method, different weights are assigned to different points according to the density distribution or the reflection intensity distribution of the point cloud in the blind area region. Specifically, first, the local neighborhood point density of each point in the blind area region is calculated (for example, the density is estimated based on the number of points within a fixed radius or based on the k-nearest neighbor distance) ; then, the points with higher density usually represent the parts of the region that are observed multiple times or the more stable parts of the object surface, and thus are assigned higher weights; the weight can be set as the local density value or the normalized value of the point.

[0107] After the weights are assigned, the weighted centroid calculation is performed, and the calculation formula is as follows:

[0108]

[0109] In the formula, (x i ,y i ,z i ) represents the coordinates of the i-th point; w i represents the weight assigned based on the density; N represents the total number of points in the blind area region; and (x c ,y c ,z c ) represents the coordinates of the center point.

[0110] According to the position and signal characteristics of the receivers, a receiver coverage model is established, and a spatial index is used to accelerate the query, so as to establish the coverage relationship between the receivers and the center points of the blind area regions.

[0111] As a preferred embodiment, the establishing a receiver coverage model according to the position and signal characteristics of the receiver and accelerating the query through spatial indexing to establish the coverage relationship between the receiver and the center points of the blind area comprises:

[0112] The three-dimensional coordinates and signal coverage ranges of all the receivers are acquired, and a spatial representation of the receiver coverage model is constructed;

[0113] It should be noted that each receiver has a fixed three-dimensional spatial position and signal coverage parameters defined by its physical characteristics, such as coverage radius (spherical signal model, suitable for omnidirectional receivers); directivity angle and direction vector (constructing a conical or fan-shaped model, suitable for directional receivers); then combining the above parameters, the coverage range model of the receiver can be simulated in three-dimensional space using standard geometric bodies (such as spheres, cones, polyhedrons).

[0114] The three-dimensional coordinates of all the center points of the blind area are loaded into a preset spatial indexing structure, and a data index for fast proximity query is constructed;

[0115] It should be noted that when loading the three-dimensional coordinates of all the center points of the blind area into the preset spatial indexing structure, the set of blind area center points calculated by the weighted centroid method needs to be first organized into a unified data format and imported into an efficient indexing structure of three-dimensional space (such as k-d tree, R-tree or Octree). These structures can support fast proximity query and range intersection operations, thereby significantly reducing the overhead of brute-force traversal of all blind area points in the receiver coverage relationship calculation.

[0116] The coverage model of each receiver is queried using spatial indexing (judging which points the receiver model overlaps), and the center points of the blind area intersected by the coverage range of the receiver are filtered out, and the coverage relationship between the receiver and the center points of the blind area is generated according to the query result.

[0117] Specifically, for the geometric coverage model of each receiver, an intersection query of spatial indexing is performed; the query result is a set of blind area center points that can be effectively covered by the receiver; the result is recorded as a set of "receiver-blind area center point" mapping relationships to form a coverage matrix, i.e. a Boolean structure of coverage state.

[0118] The key node screening module 202 is configured to screen out a minimum set of receivers capable of covering all the center points of the blind area according to the coverage relationship, and serve as key nodes.

[0119] It should be noted that, specifically, in the coverage relationship, some blind areas may be repeatedly covered by multiple receivers; it is necessary to use a linear programming / integer linear programming model to construct a ''minimum receiver set'' coverage problem: the constraint condition is that each blind area center point is covered by at least one receiver; the optimization goal is to minimize the number of receivers used. Finally, a heuristic algorithm (such as the greedy set cover algorithm) is used to approximately solve it.

[0120] The inspection path generation module 203 is configured to generate a weighted optimal inspection path by using an improved ant colony algorithm in combination with the patrol starting point and the key nodes.

[0121] Specifically, the improved ant colony algorithm is used to generate a weighted optimal inspection path, which means that based on the spatial distribution information of the patrol starting point and the minimum receiver set (i.e., the key nodes) selected, a weighted graph model containing factors such as distance, energy consumption, and priority is constructed, and an improved ant colony algorithm (ACO) is used to search for an inspection path that passes through all key nodes and has the lowest total cost.

[0122] The receiver determination module 204 is configured to determine the set of receivers to be connected according to the sequence of key nodes passed through by the optimal inspection path.

[0123] It should be noted that according to the sequence of key nodes passed through by the optimal inspection path, the set of receivers that need to be connected and inspected in sequence in the path can be directly determined, i.e., the set of receivers to be connected. Since these key nodes are selected by minimum coverage optimization, they can collectively cover all blind area regions.

[0124] The inspection state monitoring unit 3 is configured to perform an inspection operation and maintenance task based on the optimal inspection path, determine the inspection state of the administrator according to the set of receivers to be connected, and record the inspection state of the administrator, so as to monitor the underground garage operation and maintenance task.

[0125] As a preferred embodiment, the inspection state monitoring unit 3 comprises:

[0126] The signal connection module 301 is configured to establish a signal connection between the transmitter preset on the administrator and the set of receivers to be connected when the administrator performs an inspection operation and maintenance task based on the optimal inspection path and inspects to the signal coverage range of the set of receivers to be connected.

[0127] The inspection record module 302 is configured to record the connection time of the transmitter and the set of receivers to be connected and the receiver number, and generate a corresponding inspection record entry.

[0128] The inspection state judgment module 303 is configured to compare the inspection record entry with the set of receivers to be connected, judge whether the current inspection of the administrator inspects all blind areas, and obtain an inspection result.

[0129] Specifically, in the process of performing inspection and maintenance tasks based on the optimal inspection path, the transmitter carried by the management personnel is dynamically connected with the key receivers in the path through signal connection. When the management personnel enters the signal coverage range of a certain receiver, communication is automatically established to ensure touchless confirmation. Subsequently, the inspection record module records the timestamp and receiver number of each connection in real time to form standardized inspection log entries. Finally, the inspection status judgment module compares the generated records with the preset list of key receivers to determine whether the inspection task for all target areas has been completed, thereby outputting the result of whether the inspection meets the standard. This mechanism not only realizes the automatic inspection execution and data collection closed loop, but also provides traceable, quantifiable, and assessable execution basis for underground garage maintenance tasks, significantly improving management transparency and inspection compliance.

[0130] As shown in Figure 2 According to another embodiment of the present application, a method for managing underground garage maintenance tasks based on the Internet of Things is provided, comprising the following steps:

[0131] S1, constructing an initial digital twin model of the garage based on the structural data of the underground garage, collecting state data of the underground garage, and identifying blind area regions of the underground garage by combining a grid comparison algorithm;

[0132] S2, based on the identified blind area regions, the initial patrol point of the maintenance personnel, and the preset receivers in the underground garage, generating an optimal inspection path for the management personnel using a path planning algorithm, and determining the receivers to be connected in combination with the optimal inspection path;

[0133] S3, performing inspection and maintenance tasks based on the optimal inspection path, determining the inspection status of the management personnel according to the receivers to be connected and recording it to realize the monitoring of the underground garage maintenance tasks.

[0134] To sum up, by means of the technical scheme of the present application, the present application realizes a full-automatic process from sensing dead angle detection to personnel patrol inspection task closed-loop management, can construct a digital twin model based on structure data and state information, accurately identifies the coverage blind area, then reasonably arranges the inspection route through an optimal path planning algorithm, dynamically determines the task node, finally, in the inspection execution stage, combines signal connection and state comparison, records the management personnel's work situation in real time, ensures the full coverage and traceability of the task, greatly improves the operation and maintenance efficiency, data accuracy and management transparency. The present application constructs a high-precision initial digital twin model through laser radar point cloud, can comprehensively reflect the garage space structure, dynamically generates a real-time digital twin model based on parallax depth in combination with stereovision matching and multi-level threshold extraction method, then, through voxel grid division and alignment comparison, the spatial accuracy of the two models is matched, the blind area caused by shielding, structure change or light difference is accurately identified, not only high-resolution identification of the blind area is realized, but also the invisible area is continuously monitored in a complex and dynamic environment. The present application generates a weighted optimal inspection path considering distance, energy consumption and priority by combining the patrol starting point of the management personnel with the improved ant colony algorithm, and determines the set of receivers to be connected according to the key nodes passed in the path, not only greatly compresses the inspection path length and time cost, but also guarantees the integrity of blind area coverage and the pertinence of receiver management, provides efficient and reliable decision support for inspection scheduling and resource allocation in the intelligent garage.

[0135] Those skilled in the art will understand that embodiments of the present application can be provided as methods, systems or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) containing computer-usable program code.

[0136] The above specific embodiments further illustrate the purpose, technical scheme and beneficial effects of the present application, and it should be understood that the above description is only for specific embodiments of the present application and is not used to limit the protection scope of the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. An Internet of Things-based underground parking garage operation and maintenance task management system, characterized in that, include: The blind spot identification unit is used to build an initial digital twin model of the underground parking garage based on its structural data, collect the status data of the underground parking garage, and identify blind spot areas of the underground parking garage by combining grid comparison algorithms. The path planning unit is used to generate the optimal inspection path for management personnel based on the identified blind spot areas, the initial patrol point of the operation and maintenance personnel, and the preset receivers in the underground parking garage, and to determine the receivers to be connected based on the optimal inspection path. The inspection status monitoring unit is used to execute inspection and maintenance tasks based on the optimal inspection path. It determines and records the inspection status of the management personnel according to the receiver to be connected, so as to realize the monitoring of the operation and maintenance tasks of the underground parking garage. The path planning unit includes: a coverage relationship establishment module, used to determine the center points of all blind area regions based on the identified blind area regions using a spatial clustering algorithm, and establish the coverage relationship between receivers and the blind area center points; a key node selection module, used to select the smallest set of receivers that can cover all blind area center points based on the coverage relationship, and use them as key nodes; an inspection path generation module, used to generate a weighted optimal inspection path by combining the patrol initial point and key nodes using an improved ant colony algorithm; and a receiver determination module, used to determine the set of receivers to be connected based on the key node sequence traversed by the optimal inspection path. The process of determining the center points of all blind area regions based on the identified blind area regions using a weighted centroid method and establishing the coverage relationship between receivers and blind area center points includes: using the point cloud data of the blind area regions, and using a weighted centroid method... The method calculates the coordinates of the center point of each blind zone; based on the receiver's location and signal characteristics, a receiver coverage model is established, and spatial indexing is used to accelerate the query and establish the coverage relationship between the receiver and the blind zone center point. This includes: obtaining the three-dimensional coordinates and signal coverage range of all receivers and constructing a spatial representation of the receiver coverage model; loading the three-dimensional coordinates of all blind zone center points into a preset spatial index structure and constructing a data index for fast proximity query; using the spatial index to perform a spatial query on the coverage model of each receiver, filtering out the blind zone center points that intersect with the receiver's coverage range, and generating the coverage relationship between the receiver and the blind zone center point based on the query results.

2. The IoT-based underground parking garage operation and maintenance task management system according to claim 1, characterized in that, The blind spot identification unit includes: The initial model building module is used to acquire point cloud data of the underground parking garage using LiDAR and perform preprocessing. Based on the preprocessed point cloud data, an initial digital twin model of the parking garage is built. The real-time model building module is used to acquire the status data of the underground parking garage based on a pre-set camera array in the underground parking garage, and to build a real-time digital twin model of the parking garage based on the status data using a stereo vision matching algorithm with parallax depth. The grid comparison module is used to divide the initial digital twin model and the real-time digital twin model of the parking garage into grids respectively; and combined with the grid comparison method, it identifies blind spots in the underground parking garage.

3. The Internet of Things-based underground parking garage operation and maintenance task management system according to claim 2, characterized in that, The process of acquiring status data of the underground parking garage based on a pre-installed camera array, and constructing a real-time digital twin model of the parking garage using a parallax depth stereo vision matching algorithm based on the status data, includes: The garage images captured by the camera array are converted to a color space, and the structured light stripe region is extracted based on the brightness channel threshold. Within the extracted structured light stripe region, structured rays are extracted based on multi-level threshold segmentation of the hue channel, and the center line of each structured ray is extracted. Based on the centerline of each structured ray, stereo matching is performed in a preset hierarchical order. The three-dimensional coordinates of the matching points are determined by calculating the parallax of the matching points. Point cloud data is generated based on the three-dimensional coordinates of the matching points, and a real-time digital twin model of the garage is constructed.

4. The Internet of Things-based underground parking garage operation and maintenance task management system according to claim 3, characterized in that, The step of extracting structured light rays within the extracted structured light fringe region, based on multi-level threshold segmentation of the hue channel, and extracting the center line of each structured light ray includes: Based on the light intensity in the underground parking garage, the hue channel thresholds for each structured light are set; Based on the hue channel threshold of each structured light, a multi-level threshold segmentation method is used to separate the structured light stripe masks of each color in the hue channel. Morphological processing is performed on each type of structured light stripe mask to extract the initial geometric center line of each structured light ray. The initial geometric center line is then fitted using line segment fitting to obtain the final center line of each structured light ray.

5. The Internet of Things-based underground parking garage operation and maintenance task management system according to claim 2, characterized in that, The initial digital twin model and the real-time digital twin model of the garage are respectively divided into grids; In conjunction with grid comparison, blind spots in underground parking garages were identified, including: The initial digital twin model and the real-time digital twin model of the garage are aligned in coordinates, and a unified voxel mesh is generated for both models. Map the real-time digital twin model of the garage after voxel mesh division to the initial digital twin model of the garage after voxel mesh division; Based on the initial digital twin model of the mapped garage, the overlap of each grid is calculated, and grids with an overlap of less than a preset threshold are selected. The selected grids are marked, and blind spots in the underground parking garage are determined based on the marked grids.

6. The Internet of Things-based underground parking garage operation and maintenance task management system according to claim 1, characterized in that, The inspection status monitoring unit includes: The signal connection module is used to establish a signal connection between the transmitter on the administrator and the receiver to be connected when the administrator performs inspection and maintenance tasks based on the optimal inspection path and inspects to the signal coverage area of ​​the receiver to be connected. The inspection record module is used to record the connection time and receiver number between the transmitter and the receiver to be connected, and to generate corresponding inspection record entries. The inspection status judgment module is used to compare the inspection record entries with the set of receivers to be connected to determine whether the manager has inspected all blind spots during the current inspection and obtain the inspection results.

7. A method for managing the operation and maintenance tasks of an underground parking garage based on the Internet of Things (IoT), employing the IoT-based underground parking garage operation and maintenance task management system as described in any one of claims 1-6, characterized in that, Includes the following steps: S1. Construct an initial digital twin model of the underground parking garage based on its structural data, collect status data of the underground parking garage, and identify blind spots in the underground parking garage using a grid comparison algorithm. S2. Based on the identified blind spots, the patrol starting point of the maintenance personnel, and the pre-set receivers in the underground parking garage, the optimal patrol route for the personnel is generated using a path planning algorithm, and the receivers to be connected are determined in combination with the optimal patrol route. S3. Execute inspection and maintenance tasks based on the optimal inspection path. Determine and record the inspection status of management personnel according to the receiver to be connected, so as to realize the monitoring of underground parking garage operation and maintenance tasks.

Images