A patrol inspection system and method for garbage classification management personnel

Abstract

The invention discloses a patrol inspection system and method for garbage classification management personnel, belonging to the technical field of garbage classification; a patrol inspection method for garbage classification management personnel, determining all patrol inspection points and patrol inspection subjects within a management area, spatializing the management area based on the patrol inspection points, and marking a plurality of discrete patrol inspection points N on the management area; according to the spatialized management area, the management area can be divided into a plurality of area clusters by a clustering algorithm, and the patrol inspection route of the patrol inspection subject is planned by a greedy algorithm, and the patrol inspection subject completes the inspection of a patrol inspection point and uploads the corresponding patrol inspection record; until the inspection of all patrol inspection points is completed; the patrol inspection of the patrol inspection points can be realized by the collaborative inspection of unmanned aerial vehicles and manual personnel, and the routes of unmanned aerial vehicles and patrol inspection personnel are planned in real time, so as to improve the efficiency of the patrol inspection and facilitate the management of management personnel.

Description

Technical Field

[0001] The present invention belongs to the technical field of garbage classification, and more specifically, relates to an inspection system and method for garbage classification management personnel. Background Art

[0002] In the existing garbage classification management, usually only fixed inspection locations are set up, and the inspectors manually inspect and take photos and then upload the inspection results to the group. The management personnel need to check each inspection result one by one and compare it with the corresponding inspectors. Since the inspection route can only be determined in advance, temporary inspections of other locations may be required in actual situations. These inspection points that require temporary inspections are mobile inspection points. When mobile inspection points appear, the management personnel need to re-arrange the tasks of the inspection personnel, which not only affects the mutual synchronization transmission and processing efficiency of information, but also brings many inconveniences to the work of management personnel. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a patrol inspection system and method for garbage classification management personnel, which can realize patrol inspection of patrol points through the collaboration of drones and manual labor, and plan the routes of drones and patrol personnel in real time, thereby improving the efficiency of patrol inspections and facilitating management by management personnel.

[0004] The present invention provides a patrol inspection system and method for garbage classification management personnel, which includes steps S1-S5;

[0005] Step S1: Determine all inspection points and inspection subjects in the management area, spatialize the management area based on the inspection points, and mark several discrete inspection points N on the management area; the inspection subject includes at least one inspection personnel and at least one drone; assume that the inspection subject has a constant stay time at the inspection point; define the optimal total inspection time function F of the inspection subject;

[0006] Step S2: According to the spatialized management area, the management area can be divided into several regional clusters through clustering algorithm, and the geometric center of all inspection points in the regional cluster is used as the regional cluster candidate point; the regional cluster is selected as the drone initial cluster and the personnel initial cluster; the selection of the drone initial cluster and the personnel initial cluster meets the constraint condition A: the pair of the farthest distance between the regional cluster candidate points is used as the drone initial cluster and the personnel initial cluster respectively;

[0007] Step S3: Plan the inspection route of the inspection subject through the greedy algorithm. The inspection route planning satisfies the constraint condition B: There are no repeated inspection points K in the inspection route of the inspection subject. i , the inspection route traverses all inspection points {K1, ...K i , …K N}; The inspection point selected by the greedy algorithm is the next inspection point that can be reached in the shortest time from the current inspection subject's location;

[0008] Step S4: After the inspection subject completes the inspection of a patrol point and uploads the corresponding patrol record, if the stay time of any patrol subject at the patrol point is not equal to the assumed time, repeat steps S2 and S3 to recalibrate the planned patrol route;

[0009] Step S5: When the drone uploads the inspection point K i If the inspection record does not meet the requirements of the management personnel, the point K i Mark as a necessary inspection point and re-plan the inspection route so that at least one inspector's inspection route includes K i ; until all inspection points are inspected.

[0010] As a further improvement of the present invention, the specific contents of steps S1-S5 are as follows:

[0011] Step S1: The inspection subject includes an inspection person and a drone. The inspection person and the drone conduct inspections at the same time. Discrete inspection points are set with corresponding coordinates. The inspection points are defined as K based on the coordinates of the discrete inspection points. i , forming a point set {K1,…K i , …K N}; i is an integer from 1 to N; where the optimal total inspection time function of the inspection personnel is defined as F (h), and the point set corresponding to F (h) is recorded as {K h}; Define the optimal total inspection time function of the drone as F(m), and the point set corresponding to F(m) is recorded as {K m}; Point set {K h} and the point set {K m} have no intersection and their union is the point set {K N}.

[0012] Step S2: Preliminarily determine the location of the points; the total number of points ≥ 3; according to the plane position distance of the map, the point set {K1, ...K i , …K N} is divided into several clusters; the geometric center point of each cluster can be used as a candidate point for a regional cluster; when the number of inspection points in the drone initial cluster or the personnel initial cluster is ≤3, the geometric center of the inspection point in the regional cluster is directly selected as the candidate point for the regional cluster; when the number of inspection points in the drone initial cluster or the personnel initial cluster is >3, three inspection points are randomly selected and their geometric centers are used as candidate points for the regional cluster;

[0013] Step S3: Plan F(h) and F(m) separately according to the greedy algorithm; at any time of planning, K iWhen the corresponding personnel inspect the points, it is removed from the inspection point set of the drone's planned path; otherwise, K i When the drone inspects the site, it is removed from the inspection point set of the personnel's planned path; the greedy algorithm is to use the shortest time of the path planning as the radius for the current point of the inspection subject, and cover the inspection points through the time radius, where the shortest time refers to the fastest time from the current point to the next inspection point; when the shortest time is used as the radius to cover two inspection points at the same time, the route function is added according to step S2;

[0014] Step S4: After the inspection subject completes the inspection of a patrol point and uploads the corresponding patrol record, if the stay time of any patrol subject at the patrol point is not equal to the assumed time, repeat steps S2 and S3 to recalibrate the planned patrol route;

[0015] Step S5: When the drone uploads the inspection point K i If the inspection record does not meet the requirements of the management personnel, the point K i Mark it as a necessary inspection point, re-plan the inspection route, and change the inspection point K i From the point set {K m} move out, move in point set {K h}; so that at least one inspector's inspection route includes K i ; until all inspection points are inspected.

[0016] As a further improvement of the present invention, the specific content of step S3 is as follows, including steps S31 to S35:

[0017] Step S31: Obtain the inspection point set {K N}; Each inspection subject defines an empty queue Q to record the inspection points that have been inspected, and defines an empty queue R to record the route turning points;

[0018] Step S32: obtaining the current location of the inspection subject, denoted as S;

[0019] Step S33: Calculate the point-to-point set {K N}, select point S to {K N}, select the point O with the shortest travel time, add the point O to the queue Q, add the route turning point from point S to the point O to the queue R, and move the point O from {K N}remove;

[0020] Step S34: taking the point O selected in step S33 as the new origin S, performing step S33 again to select the next inspection point of the trip;

[0021] Step S35: recursively repeat steps S34 and S35 until {K N} becomes an empty set; at this time, the queue order of queue Q is the inspection order of each inspection point planned for this trip; the inspection route of this trip can be obtained by connecting the route turning points in queue R according to the queue order of queue Q.

[0022] As a further improvement of the present invention, the specific content of step S4 is as follows, including step S41-step S42;

[0023] Step S41: The inspection background collects the inspection results of the inspection personnel, and the inspection background stores all the inspection records of the inspection personnel to generate an inspection log, and updates the inspection point information on the virtual electronic map in real time according to the inspection results;

[0024] Step S42: The inspection personnel use the inspection background to query the inspection records submitted by themselves; the management personnel use the inspection background to query the inspection information of each inspection point or each inspection personnel.

[0025] As a further improvement of the present invention, step S41 is specifically as follows: searching based on the inspector, the manager enters the name of the inspector through the search box of the inspection background to view the inspection situation of the inspector, and the search results will display the inspection information of the person in the form of a list, and the inspection information includes the total number of inspections, the number of fixed-point inspections and the number of mobile inspections of the inspector counted in the list, including the start time and end time of each inspection;

[0026] Search by date. In the date search box of the inspection background, select the corresponding date to search. The search results will show the inspection status of the selected date.

[0027] Search by region and select the region under the organizational structure in the inspection background to view the inspection status of a certain region.

[0028] A patrol inspection system for garbage classification management personnel is implemented based on a patrol inspection method for garbage classification management personnel; it includes a patrol inspection point module, an allocation module and a patrol inspection background; the patrol inspection point module includes a number of patrol inspection points N, and the patrol inspection points are all located in a management area. After the patrol inspection subject arrives at the patrol inspection point, the patrol inspection result of the patrol inspection point is recorded and uploaded; the allocation module is used to spatialize the management area based on the patrol inspection point, and dynamically plan the patrol inspection route of the patrol inspection subject; the patrol inspection background is used to record and update the patrol inspection situation of each patrol inspection subject and patrol inspection point.

[0029] As a further improvement of the present invention, after arriving at the inspection point, the inspection personnel at least record and upload the inspection content including inspection instructions, inspection photos, inspection points, hygiene conditions, personnel conditions, classification conditions and packet loss conditions; the inspection personnel scan the QR code of the inspection point to enter the mini program to fill in the inspection content, and inspection photos only support real-time shooting and uploading.

[0030] As a further improvement of the present invention, the inspection background also includes a personnel unit; based on the inspection personnel, the search is performed, and the management personnel enters the name of the inspection personnel through the search box of the inspection background to view the inspection status of the inspection personnel. The search results will display the inspection information of the personnel in the form of a list. The inspection information includes the total number of inspections, fixed-point inspections and mobile inspections of the inspection personnel counted in the list, including the start time and end time of each inspection.

[0031] As a further improvement of the present invention, the inspection background also includes a date unit; in the date search box of the inspection background, a corresponding date is selected for search, and the search results will display the inspection status of the selected date.

[0032] As a further improvement of the present invention, the inspection background further includes a region unit; in the inspection background, a region under the organizational structure is selected to view the inspection status of a certain region.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] By setting up the inspection point module, allocation module and inspection background, all inspection points and inspection subjects in the management area are determined, the management area is spatialized based on the inspection points, and several discrete inspection points N are marked on the management area; according to the spatialized management area, the management area can be divided into several area clusters through the clustering algorithm, and the inspection route of the inspection subject is planned through the greedy algorithm. The inspection subject completes the inspection of a inspection point and uploads the corresponding inspection record; until the inspection of all inspection points is completed, the inspection of the inspection points can be realized through the collaboration of drones and manual work, and the routes of drones and inspection personnel are planned in real time, which improves the efficiency of inspection and facilitates management by management personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic diagram of an inspection route of the present invention;

[0036] Figure 2 is a schematic diagram of another inspection route of the present invention;

[0037] Figure 3 is a schematic diagram of another inspection route of the present invention;

[0038] Figure 4 It is a schematic diagram of two inspection points in the regional cluster of the present invention;

[0039] Figure 5 It is a schematic diagram of three inspection points in the regional cluster of the present invention;

[0040] Figure 6 A schematic diagram of a necessary personnel inspection point of the present invention;

[0041] Figure 7 It is a schematic diagram of the workflow of the present invention;

[0042] Figure 8 This is a schematic diagram of the inspection workflow. DETAILED DESCRIPTION

[0043] Specific embodiment 1: Please refer to Figure 1-Figure 8 A method for inspection by garbage classification management personnel, comprising steps S1-S5;

[0044] Step S1: Determine all inspection points and inspection subjects in the management area, spatialize the management area based on the inspection points, and mark several discrete inspection points N on the management area; the inspection subject includes at least one inspection personnel and at least one drone; it is assumed that the inspection subject has a constant stay time at the inspection point; define the optimal inspection total time function F of the inspection subject. It should be noted that in other feasible embodiments, the number of inspection personnel or drones can be ≥2. Among them, the positions of inspection points, inspection personnel and drones can all be obtained through positioning technologies such as GPS. The position of the inspection personnel is obtained based on the mobile device they carry, and the position of the drone is obtained based on its own positioning device.

[0045] The specific content of step S1 is as follows: In this embodiment, the inspection subject includes an inspection person and a drone. The inspection person and the drone inspect at the same time. The discrete inspection points are set with corresponding coordinates. The inspection points are defined as K based on the coordinates of the discrete inspection points. i , forming a point set {K1,…K i , …K N}; i is an integer from 1 to N; where the optimal total inspection time function of the inspection personnel is defined as F (h), and the point set corresponding to F (h) is recorded as {K h}; Define the optimal total inspection time function of the drone as F(m), and the point set corresponding to F(m) is recorded as {K m}; Point set {K h} and the point set {K m} have no intersection and their union is the point set {K N}.

[0046] Step S2: According to the spatialized management area, the management area can be divided into several regional clusters through a clustering algorithm, and the geometric centers of all inspection points in the regional cluster are used as candidate points of the regional cluster; the regional clusters are selected as the initial clusters of drones and the initial clusters of personnel; the selection of the initial clusters of drones and the initial clusters of personnel meets the constraint condition A: the pair of candidate points of the regional clusters with the farthest distance between them are respectively used as the initial clusters of drones and the initial clusters of personnel.

[0047] Among them, the regional cluster candidate points are determined by the clustering algorithm. First, the distances between all regional cluster candidate points are traversed and calculated respectively, and then they are compared one by one. Finally, the pair of regional cluster candidate points with the farthest distance between them are the regional cluster candidate points to be selected; so that the drone and the inspection personnel can start inspection from the farthest point respectively, maximizing the inspection efficiency.

[0048] For example, suppose there are currently three regional clusters, namely, regional cluster X, regional cluster Y, and regional cluster Z; there are and only candidate points of regional cluster X1 in regional cluster X, there are and only candidate points of regional cluster Y1 in regional cluster Y, and there are and only candidate points of regional cluster Z1 in regional cluster Z; firstly, the distance between candidate points of regional cluster X1 and candidate points of regional cluster Y1 is calculated, then the distance between candidate points of regional cluster X1 and candidate points of regional cluster Z1 is calculated, and finally the distance between candidate points of regional cluster Y1 and candidate points of regional cluster Z1 is calculated; a pair of candidate points of regional cluster with the farthest distance between them are selected as the initial cluster of drones and the initial cluster of personnel, respectively.

[0049] The specific contents of step S2 are as follows: preliminarily determine the location of the points; the total number of points ≥ 3; according to the plane position distance of the map, the point set {K1, ...K i , …K N} is divided into several clusters; the geometric center point of each cluster can be used as a candidate point for a regional cluster; when the number of inspection points in the drone initial cluster or the personnel initial cluster is ≤3, the geometric center of the inspection point in the regional cluster is directly selected as the candidate point for the regional cluster.

[0050] Among them, when the number of inspection points = 1, this inspection point is a candidate point for the regional cluster; Figure 4 As shown in , when the number of inspection points = 2, the midpoint between the two points is taken as the candidate point of the regional cluster; Figure 5 As shown in the figure, when the number of inspection points = 3, the centroid of the triangle formed by the three points is taken as the candidate point of the regional cluster ( Figure 5 The dashed line in the middle is the auxiliary line for constructing the centroid).

[0051] Preferably, the number of inspection points in a regional cluster should be limited to 3 or less. When the number of inspection points is greater than 3, the complexity of calculating the route will be greatly increased.

[0052] When the number of inspection points in the initial cluster of drones or personnel is greater than 3, three inspection points are randomly selected and their geometric centers are used as candidate points for the regional cluster.

[0053] Step S3: Plan the inspection route of the inspection subject through the greedy algorithm. The inspection route planning satisfies constraint condition B: there are no repeated inspection points Ki in the inspection route of the inspection subject, and the inspection route traverses all inspection points {K1, ...Ki, ...KN}; the inspection point selected by the greedy algorithm is the next inspection point that can be reached in the shortest time from the current location of the inspection subject.

[0054] The specific content of step S3 is: plan F(h) and F(m) respectively according to the greedy algorithm; at any time of planning, K i When the corresponding personnel inspect the points, it is removed from the inspection point set of the drone's planned path; otherwise, K i When the drone is inspecting the site, it is removed from the inspection point set of the personnel's planned path;

[0055] The greedy algorithm takes the shortest time of the path planning as the radius of the current point of the inspection subject, and covers the inspection points through the time radius, where the shortest time refers to the fastest time from the current point to the next inspection point; when the shortest time is used as the radius to cover two inspection points at the same time, the route function is added according to step S2; that is, the subsequent inspection routes of the two inspection points are calculated separately, and the one that takes less time is selected.

[0056] In this embodiment, the initial cluster area of ​​the drone is taken as an example; Figure 1 As shown in the figure, point 1 is selected as the starting point, and the estimated duration of the planned route is 33 minutes; Figure 2 As shown in the figure, point 2 is selected as the starting point, and the estimated duration of the planned route is 35 minutes; Figure 3 As shown in the figure, point 3 is selected as the starting point, and the estimated duration of the planned route is 34 minutes; therefore, in the current situation, the drone chooses Figure 1 It should be noted that the route selected in this embodiment is a local optimal solution based on the inspection points that have been inspected and the inspection points that have not been completed in the current situation. Similarly, the working principle of the route planning of the inspectors will not be repeated here (the time taken by the inspectors to inspect the route is not marked in the figure).

[0057] It should be noted that the calculation formula for the shortest time is: , where t is the shortest time, s is the distance between the current inspection point and the target inspection point, is the average speed of the inspection personnel or drones; since the flight of drones is less affected by obstacles, the flight distance of drones can be approximated as the straight-line distance between two points, and the inspection personnel use normal transportation methods to travel between various inspection points, including walking, bicycles, and electric vehicles. Their travel distance can be obtained through existing conventional technical means such as GPS. Among them, let the normal flight speed of the drone be v, is 0.8~0.95v. When there are few obstacles between the two inspection points, The value of is close to 0.95v. When there are many obstacles between the two inspection points, The value is close to 0.8v.

[0058] Specifically, step S3 also includes steps S31 to S35: Step S31: Obtaining a set of inspection points in the management area {K N}; Each inspection subject defines an empty queue Q to record the inspection points that have been inspected, and defines an empty queue R to record the route turning points;

[0059] Step S32: obtaining the current location of the inspection subject, denoted as S;

[0060] Step S33: Calculate the point-to-point set {K N}, select point S to {K N}, select the point O with the shortest travel time, add the point O to the queue Q, add the route turning point from point S to the point O to the queue R, and move the point O from {K N}remove;

[0061] Step S34: taking the point O selected in step S33 as the new origin S, performing step S33 again to select the next inspection point of the trip;

[0062] Step S35: recursively repeat steps S34 and S35 until {K N} becomes an empty set; at this time, the queue order of queue Q is the inspection order of each inspection point planned for this trip; the inspection route of this trip can be obtained by connecting the route turning points in queue R according to the queue order of queue Q.

[0063] Step S4: After the inspection subject completes the inspection of a patrol point and uploads the corresponding patrol record, if the stay time of any patrol subject at the patrol point is not equal to the assumed time, repeat steps S2 and S3 to recalibrate the planned patrol route;

[0064] Preferably, after arriving at the inspection point, the inspection personnel at least record and upload the inspection content including inspection instructions, inspection photos, inspection points, hygiene conditions, personnel conditions, classification conditions and packet loss conditions; the inspection personnel scan the QR code of the inspection point to enter the mini program to fill in the inspection content, and inspection photos only support real-time shooting and uploading.

[0065] Specifically, step S4 also includes step S41-step S42; step S41: the inspection background collects the inspection results of the inspection personnel, and the inspection background stores all the inspection records of the inspection personnel to generate an inspection log, and updates the inspection point information on the virtual electronic map in real time according to the inspection results;

[0066] Specifically, step S41 is: searching based on the inspector, the manager enters the name of the inspector through the search box of the inspection background to view the inspection status of the inspector, and the search results will display the inspection information of the inspector in the form of a list, and the inspection information includes the total number of inspections, fixed-point inspections and mobile inspections of the inspector counted in the list, including the start time and end time of each inspection;

[0067] Search by date. In the date search box of the inspection background, select the corresponding date to search. The search results will show the inspection status of the selected date.

[0068] Search by region and select the region under the organizational structure in the inspection background to view the inspection status of a certain region.

[0069] Step S42: The inspection personnel use the inspection background to query the inspection records submitted by themselves; the management personnel use the inspection background to query the inspection information of each inspection point or each inspection personnel.

[0070] Step S5: When the drone uploads the inspection point K i If the inspection record does not meet the requirements of the management personnel, the point K i Mark it as a necessary inspection point, re-plan the inspection route, and change the inspection point K i From the point set {K m} move out, move in point set {K h}; so that at least one inspector's inspection route includes K i ; until all inspection points are inspected.

[0071] like Figure 6 As shown, the black dots in the figure are marked as points that must be inspected by personnel. Although they have been inspected once by a drone, since the inspection record does not meet the requirements of the management personnel, they need to be inspected again by the inspection personnel, and their route planning still follows steps S1 to S4.

[0072] It should be noted that in this embodiment, the drone completes the inspection record by taking and uploading photos of the inspection point. Due to the influence of actual factors such as light, weather, and object obstruction, the photos taken may not fully reflect the inspection content of the sanitation, personnel, classification, and packet loss in the inspection point. In this case, the drone is regarded as an inspection point K that does not meet the requirements of the management personnel. i Inspection records.

[0073] A patrol inspection system for garbage classification management personnel is implemented based on a patrol inspection system for garbage classification management personnel; it includes a patrol inspection point module, an allocation module and a patrol inspection background; the patrol inspection point module includes a number of patrol inspection points N, and the patrol inspection points are all located in the management area. After the patrol inspection subject arrives at the patrol inspection point, the patrol inspection result of the patrol inspection point is recorded and uploaded; the allocation module is used to spatialize the management area based on the patrol inspection point, and dynamically plan the patrol inspection route of the patrol inspection subject; the patrol inspection background is used to record and update the patrol inspection situation of each patrol inspection subject and patrol inspection point.

[0074] Preferably, after arriving at the inspection point, the inspection personnel at least record and upload the inspection content including inspection instructions, inspection photos, inspection points, hygiene conditions, personnel conditions, classification conditions and packet loss conditions; the inspection personnel scan the QR code of the inspection point to enter the mini program to fill in the inspection content, and inspection photos only support real-time shooting and uploading.

[0075] Preferably, the inspection background also includes a personnel unit; based on the inspection personnel, the search is performed, and the management personnel enter the inspection personnel's name through the search box of the inspection background to view the inspection status of the inspection personnel. The search results will display the inspection information of the personnel in the form of a list. The inspection information includes the total number of inspections, fixed-point inspections and mobile inspections of the inspection personnel counted in the list, including the start time and end time of each inspection.

[0076] Preferably, the inspection background also includes a date unit; in the date search box of the inspection background, select a corresponding date to search, and the search results will display the inspection status of the selected date.

[0077] Preferably, the inspection background also includes a regional unit; in the inspection background, a region under the organizational structure is selected to view the inspection situation of a certain region. The inspection background collects the inspection results of the inspectors, stores all the inspection records of the inspectors, generates inspection logs, and updates the inspection point information on the virtual electronic map in real time according to the inspection results. The inspection logs can be viewed on the inspection background, which is conducive to the management personnel to quickly view the inspection content and efficiently arrange personnel.

[0078] In this embodiment, the types of inspection points N include fixed inspection points A (including A1, A2, A3...) and mobile inspection points B (including B1, B2, B3...). Fixed inspection points: The location of fixed inspection points is fixed and is generally used as a long-term inspection location. Mobile inspection points: The location of mobile inspection points is uncertain. When a certain location needs to be temporarily inspected, it is arranged by the management personnel. The type of each inspection point N is set by the management personnel. The information of fixed inspection points and mobile inspection points is marked on a virtual electronic map constructed by the terrain of the inspection area.

[0079] It should also be noted that the inspection point N is divided into recyclables collection points, hazardous waste collection points, kitchen waste collection points and other waste collection points according to the type of garbage recycled; after the garbage is collected, it is transported to the corresponding location for recycling and treatment by special vehicles.

[0080] like Figure 8 As shown, in this embodiment, the problem feedback personnel include resident users, managers and / or inspectors who use a certain inspection point. The inspection workflow is as follows: first, the person who discovers the problem reports the corresponding problem to the mini program of the inspection center. The resident user can scan the QR code of the problematic inspection point or the equipment within the point. The uploaded identity of the resident user is an anonymous user. The manager or inspector can directly select the inspection point on the mobile phone, enter the mini program of the inspection center after selection or scanning, and fill out the feedback form, which includes the inspection point and the problems it has; then the inspection center automatically notifies the front-line personnel of the information, and the notification methods include text messages and phone calls; finally, the front-line personnel arrive at the target inspection point, solve the relevant problems, and upload the solution to the inspection center for feedback.

Claims

1. A method for inspection by garbage classification management personnel, characterized in that: Comprising steps S1-S5; Step S1: determine all inspection points and inspection subjects in the management area, spatialize the management area based on the inspection points, and mark a number of discrete inspection points N on the management area; the inspection subject includes at least one inspection personnel and at least one drone; Assume that the inspection subject's stay time at the inspection point is constant; define the inspection subject's optimal total inspection time function F; When the inspection subject includes an inspector and a drone, the inspector and the drone inspect at the same time, and the discrete inspection points are set with corresponding coordinates. The inspection points are defined as K based on the coordinates of the discrete inspection points. i , forming a point set {K1,…K i , …K N }; i is an integer from 1 to N; where the optimal total inspection time function of the inspection personnel is defined as F(h), and the point set corresponding to F(h) is denoted as {K h }; Define the optimal total inspection time function of the drone as F(m), and the point set corresponding to F(m) is recorded as {K m }; Point set {K h } and the point set {K m } have no intersection and their union is the point set {K N }; Step S2: According to the spatialized management area, the management area is divided into several regional clusters by using a clustering algorithm, and the geometric centers of all inspection points in the regional cluster are used as candidate points of the regional cluster; Select regional clusters as the initial clusters of drones and personnel; the selection of the initial clusters of drones and personnel satisfies constraint condition A: the pair of candidate points with the longest distance between regional clusters is used as the initial clusters of drones and personnel respectively; Preliminarily determine the location of the points; the total number of points is ≥ 3; according to the plane position distance of the map, the point set {K1, ...K i , …K N } is divided into several clusters; the geometric center point of each cluster is used as a candidate point for a regional cluster; when the number of inspection points in the drone initial cluster or the personnel initial cluster is ≤3, the geometric center of the inspection point in the regional cluster is directly selected as the candidate point for the regional cluster; when the number of inspection points in the drone initial cluster or the personnel initial cluster is >3, three inspection points are randomly selected and their geometric centers are used as candidate points for the regional cluster; Step S3: Plan the inspection route of the inspection subject through the greedy algorithm. The inspection route planning satisfies the constraint condition B: There are no repeated inspection points K in the inspection route of the inspection subject. i , the inspection route traverses all inspection points {K1, ...K i , …K N }; The inspection point selected by the greedy algorithm is the next inspection point that can be reached in the shortest time from the current inspection subject's location; According to the greedy algorithm, F(h) and F(m) are planned separately; at any time of planning, K i When the corresponding personnel inspect the points, it is removed from the inspection point set of the drone's planned path; otherwise, K i When the drone is inspecting the site, it is removed from the inspection point set of the personnel's planned path; The greedy algorithm is to use the shortest time of the path planning as the radius of the current point of the inspection subject, and cover the inspection points through the time radius, where the shortest time refers to the fastest time from the current point to the next inspection point; when the shortest time is used as the radius to cover two inspection points at the same time, the route function is added according to step S2; Step S4: After the inspection subject completes the inspection of a patrol point and uploads the corresponding patrol record, if the stay time of any patrol subject at the patrol point is not equal to the assumed time, repeat steps S2 and S3 to recalibrate the planned patrol route; Step S5: When the drone uploads the inspection point K i If the inspection record does not meet the requirements of the management personnel, the point K i Mark it as a necessary inspection point, re-plan the inspection route, and change the inspection point K i From the point set {K m } move out, move in point set {K h }; so that at least one inspector's inspection route includes K i ; until all inspection points are inspected.

2. A method for inspection of garbage classification management personnel according to claim 1, characterized in that: The specific content of step S3 is as follows, including steps S31 to S35: Step S31: Obtain the inspection point set {K N }; Each inspection subject defines an empty queue Q to record the inspection points that have been inspected, and defines an empty queue R to record the route turning points; Step S32: obtaining the current location of the inspection subject, denoted as S; Step S33: Calculate the point-to-point set {K N }, select point S to {K N }, select the point O with the shortest travel time, add the point O to the queue Q, add the route turning point from point S to the point O to the queue R, and move the point O from {K N }remove; Step S34: taking the point O selected in step S33 as the new origin S, performing step S33 again to select the next inspection point of the trip; Step S35: recursively repeat steps S34 and S35 until {K N } becomes an empty set; at this time, the queue order of queue Q is the inspection order of each inspection point planned for this trip; the inspection route of this trip can be obtained by connecting the route turning points in queue R according to the queue order of queue Q.

3. A method for inspection of garbage classification management personnel according to claim 1, characterized in that: The specific content of step S4 is as follows, including step S41-step S42; Step S41: The inspection background collects the inspection results of the inspection personnel, and the inspection background stores all the inspection records of the inspection personnel to generate an inspection log, and updates the inspection point information on the virtual electronic map in real time according to the inspection results; Step S42: The inspection personnel use the inspection background to query the inspection records submitted by themselves; Management personnel use the inspection background to query the inspection information of each inspection point or each inspection personnel.

4. A method for inspection by garbage classification management personnel according to claim 3, characterized in that: Step S41 is specifically as follows: searching based on the inspector, the manager enters the inspector's name through the search box in the inspection background to view the inspector's inspection status, and the search results will display the inspector's inspection information in the form of a list, the inspection information includes the total number of inspections, fixed-point inspections, and mobile inspections of the inspector counted in the list, including the start time and end time of each inspection; Search by date. In the date search box of the inspection background, select the corresponding date to search. The search results will show the inspection status of the selected date. Search by region and select the region under the organizational structure in the inspection background to view the inspection status of a certain region.

5. A patrol inspection system for garbage classification management personnel, characterized by: It is realized based on a patrol inspection method used by garbage classification management personnel as described in any one of claims 1 to 4; Includes inspection point module, allocation module and inspection background; The inspection point module includes several inspection points N, all of which are located in the management area. After the inspection subject arrives at the inspection point, the inspection results of the inspection point are recorded and uploaded; the allocation module is used to spatialize the management area based on the inspection point and dynamically plan the inspection route of the inspection subject; The inspection background is used to record and update the inspection status of each inspection subject and inspection point.

6. A patrol inspection system for garbage classification management personnel according to claim 5, characterized in that: After arriving at the inspection point, the inspection personnel shall record and upload at least the inspection description, inspection photos, inspection points, sanitation conditions, personnel conditions, classification conditions and lost packages; Inspection personnel scan the QR code of the inspection point to enter the mini program and fill in the inspection content. Inspection photos can only be taken and uploaded in real time.

7. A patrol inspection system for garbage classification management personnel according to claim 5, characterized in that: The inspection background also includes a personnel unit; by searching based on the inspection personnel, the management personnel can enter the inspection personnel's name in the search box of the inspection background to view the inspection status of the inspection personnel. The search results will display the inspection information of the personnel in the form of a list. The inspection information includes the total number of inspections, fixed-point inspections and mobile inspections of the inspection personnel counted in the list, including the start time and end time of each inspection.

8. The inspection system for garbage classification management personnel according to claim 5 is characterized by: The inspection background also includes a date unit; in the date search box of the inspection background, select the corresponding date to search, and the search results will display the inspection status of the selected date.

9. A patrol inspection system for garbage classification management personnel according to claim 5, characterized in that: The inspection background also includes regional units; in the inspection background, select the region under the organizational structure to view the inspection status of a certain area.

Images