A smart city garbage cleaning route planning method and an internet of things system

Abstract

This specification provides a smart city waste collection route planning method and an Internet of Things (IoT) system. The method is implemented through a smart city waste collection route planning IoT system, which includes a user platform, a service platform, a management platform, a sensor network platform, and an object platform. The method is executed by the management platform and includes: acquiring monitoring information from at least one road within a road network area and identifying waste accumulation on at least one road; determining at least one target waste collection point based on the waste accumulation; and determining a waste collection route based on the at least one target waste collection point. This method can be implemented using a smart city waste collection route planning device. The method can also be executed after being read from computer instructions stored on a computer-readable storage medium.

Description

Technical Field

[0001] This specification relates to the field of waste disposal technology, and in particular to a smart city waste disposal route planning method and Internet of Things system. Background Technology

[0002] Urban waste collection is a crucial aspect of urban governance. Cities are relatively small and densely populated, generating a massive amount of waste daily. During urban waste collection, some roads may be frequently cleaned, while others may remain uncleaned for extended periods, resulting in significant waste accumulation.

[0003] Therefore, it is hoped that a smart city garbage cleaning route planning method and Internet of Things system can be proposed to plan garbage cleaning routes, improve the cleanliness of urban roads, reduce repeated cleaning, and effectively save manpower and material resources. Summary of the Invention

[0004] One embodiment of this specification provides a smart city garbage collection route planning method, the method comprising: acquiring monitoring information on at least one road within a road network area, identifying garbage accumulation on at least one road; determining at least one target garbage collection point based on the garbage accumulation; and determining a garbage collection route based on the at least one target garbage collection point.

[0005] One embodiment of this specification provides an IoT system for smart city garbage collection route planning. The system includes: a user platform, a service platform, a management platform, a sensor network platform, and an object platform. The service platform sends garbage collection routes to the user platform. The object platform acquires monitoring information from at least one road within the road network area and transmits it to the management platform via the sensor network platform. The management platform is used to: acquire monitoring information from at least one road within the road network area and identify garbage accumulation on at least one road; determine at least one target garbage collection point based on the garbage accumulation; determine a garbage collection route based on the at least one target garbage collection point; and generate remote control commands based on the garbage collection route and send them to the sensor network platform. The sensor network platform sends the remote control commands to the object platform to enable the object platform to perform cleaning operations.

[0006] One embodiment of this specification provides a smart city garbage collection route planning device, the device including at least one processor and at least one memory; the at least one memory is used to store computer instructions; the at least one processor is used to execute at least some of the computer instructions to implement a smart city garbage collection route planning method.

[0007] One embodiment of this specification provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes any of the smart city waste cleaning path planning methods described in the above embodiments. Attached Figure Description

[0008] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0009] Figure 1 This is an exemplary schematic diagram of an Internet of Things (IoT) system for smart city waste collection route planning, as shown in some embodiments of this specification.

[0010] Figure 2 This is an exemplary flowchart of a smart city waste collection route planning method according to some embodiments of this specification;

[0011] Figure 3 This is an exemplary flowchart illustrating the determination of at least one target waste collection point according to some embodiments of this specification;

[0012] Figure 4 These are exemplary schematic diagrams of traffic flow prediction models according to some embodiments of this specification;

[0013] Figure 5 This is an exemplary flowchart illustrating the determination of a garbage collection route according to some embodiments of this specification. Detailed Implementation

[0014] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0015] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0016] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0017] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0018] Figure 1 This is an exemplary schematic diagram of an IoT system for smart city waste collection route planning, according to some embodiments of this specification. In some embodiments, the smart city waste collection route planning IoT system 100 includes a user platform 110, a service platform 120, a management platform 130, a sensor network platform 140, and an object platform 150.

[0019] In some embodiments, information processing in the Internet of Things (IoT) can be divided into a processing flow for sensing information and a processing flow for control information. Control information can be generated based on sensing information. Specifically, the processing of sensing information involves the user platform 110 acquiring the sensing information and transmitting it to the management platform 130. Control information is then sent from the management platform 130 to the user platform 110 to achieve corresponding control.

[0020] User platform 110 is a platform that can be used to interact with users. In some embodiments, user platform 110 can be configured as a terminal device. For example, a terminal device may include a mobile device, a tablet computer, or any combination thereof. In some embodiments, user platform 110 can be used to provide users with route plans for garbage collection in various road network areas of the city. In some embodiments, user platform 110 can interact bidirectionally with service platform 120. User platform 110 can send a garbage collection route plan query command to service platform 120 and receive garbage collection route plans uploaded by service platform 120.

[0021] Service platform 120 is a platform that can be used to receive and transmit data and / or information. In some embodiments, service platform 120 is configured with multiple sub-platforms according to the division of urban road network areas, each sub-platform corresponding to at least one urban road network area. In some embodiments, the service sub-platforms of each urban road network area can independently receive instructions from user platform 110 and independently provide feedback on garbage collection route planning, etc., for the urban road network area corresponding to the service sub-platform. In some embodiments, the sub-platforms corresponding to each urban road network area in service platform 120 can independently and bidirectionally interact with the sub-platforms of the corresponding urban road network area in management platform 130, sending garbage collection route planning query instructions to management sub-platforms and receiving garbage collection route plans uploaded by management sub-platforms.

[0022] The management platform 130 can refer to a platform that coordinates and integrates the connections and collaborations between various functional platforms, gathers all the information of the Internet of Things (IoT), and provides sensing, management, and control functions for the IoT operating system. In some embodiments, the management platform 130 can be used to execute a smart city garbage collection route planning method, respond to user query requests, process monitoring information from at least one road within the road network area uploaded by the sensor network platform 140, and determine the garbage collection route.

[0023] In some embodiments, the management platform 130 may be configured with at least one sub-platform according to the division of urban road network areas, wherein each management sub-platform may correspond one-to-one with a service sub-platform of the corresponding urban road network area. In some embodiments, each management sub-platform may acquire and process the monitoring information of the sensor network sub-platform of the corresponding road network area, and send the processed garbage collection route plan to the service sub-platform of the corresponding area. In some embodiments, each sub-platform of the management platform 130 may independently and bidirectionally interact with each sub-platform of the corresponding road network area in the sensor network platform 140, receive and process the monitoring information of each road network area uploaded by the sensor network sub-platform, and issue a command to acquire monitoring information to the sensor network sub-platform.

[0024] In some embodiments of this specification, by processing monitoring information of different road network areas through a management sub-platform, the data processing pressure of the entire management platform can be reduced, while the determination of route planning for garbage collection in each road network area of ​​the city can be more targeted and managed independently.

[0025] The sensor network platform 140 can be a functional platform for managing sensor communication. The sensor network platform 140 can be configured as a communication network and gateway, implementing functions such as network management, protocol management, command management, and data parsing. In some embodiments, the sensor network platform 140 has at least one sub-platform divided according to urban road network areas, wherein each sensor network sub-platform corresponds one-to-one with a sub-platform in the management platform 130, and the communication network and gateway can be configured independently for each sensor network sub-platform. In some embodiments, each sub-platform of the sensor network platform 140 can bidirectionally interact with the sub-platform in the corresponding urban road network area of ​​the object platform 150, receiving monitoring information uploaded by the object sub-platform in the corresponding area and issuing instructions to obtain monitoring information to the object sub-platform in the corresponding area.

[0026] The object platform 150 is a functional platform capable of generating sensing information. In some embodiments, the object platform 150 is configured as a monitoring device (such as a camera), with a unique identifier, and can be deployed in communities across different urban road network areas for management. In some embodiments, the object platform 150 is divided into at least one sub-platform based on urban road network areas, with each sub-platform corresponding one-to-one with a sub-platform in the sensor network platform 140. In some embodiments, each sub-platform in the object platform 150, configured as a monitoring device with a unique identifier, can acquire monitoring information for that urban road network area and upload it to the corresponding regional sensor network sub-platform.

[0027] like Figure 1 As shown, in the smart city waste cleaning IoT system 100, the "service sub-platform - management sub-platform - sensor network sub-platform - object sub-platform" of each city area are independent branches, which independently and in parallel process monitoring information and independently feed it back to the user platform 110.

[0028] It should be noted that the above description of the IoT system and its components is for convenience only and should not be construed as limiting this specification to the embodiments described. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the various components or construct subsystems connected to other components without departing from these principles. For example, the management platform 130 may be integrated into a single component. Alternatively, the components may share a single storage device, or each component may have its own separate storage device. Such variations are all within the scope of this specification.

[0029] Figure 2 This is an exemplary flowchart illustrating a smart city waste collection route planning method according to some embodiments of this specification. Figure 2As shown, process 200 includes the following steps. In some embodiments, process 200 may be executed by a management platform.

[0030] Step 210: Obtain monitoring information on at least one road within the road network area and identify garbage accumulation on at least one road.

[0031] A road network area refers to a region consisting of interconnected and interwoven road networks of various types. For example, a road network area can be composed of highways, while an urban road network area can be composed of various roads within a city. A city can be divided into multiple road network areas based on actual needs.

[0032] Monitoring information refers to surveillance and control information on roads. For example, monitoring information may include information on litter, pedestrians, and vehicles on the road.

[0033] In some embodiments, the management platform can acquire monitoring information based on monitoring devices. For example, the monitoring device may be a camera.

[0034] Garbage accumulation refers to the situation where garbage accumulates in piles on different roads. For example, garbage accumulation can be represented by the amount of garbage (e.g., less, moderate, excessive), the degree of accumulation (e.g., level 1, level 2, level 3), and the number of piles (1, 3, 5, etc.). A larger amount of garbage and / or a higher degree of accumulation and / or a larger number of piles indicates a more severe garbage accumulation situation on that road.

[0035] In some embodiments, the management platform can identify garbage accumulation on at least one road based on a first preset condition. The first preset condition may be a pre-set condition limiting garbage accumulation on the road. For example, the first preset condition may be the extent of garbage accumulation.

[0036] For example, a waste accumulation area of ​​no more than 1 square meter is classified as less and / or Level 1, a waste accumulation area of ​​more than 1 square meter but no more than 2 square meters is classified as moderate and / or Level 2, and a waste accumulation area of ​​more than 2 square meters is classified as excessive and / or Level 3.

[0037] For example, the first preset condition could be the height of the garbage pile. For instance, a garbage pile height of no more than 0.2 meters is classified as less and / or Level 1, a garbage pile height of more than 0.2 meters but no more than 0.5 meters is classified as moderate and / or Level 2, and a garbage pile height of more than 0.5 meters is classified as excessive and / or Level 3, etc.

[0038] For example, the first preset condition could be the amount of garbage piled up. For instance, a garbage pile of 1 piece is considered "less" and / or Level 1, a garbage pile of more than 1 piece but not more than 3 pieces is considered "moderate" and / or Level 2, and a garbage pile of more than 3 pieces is considered "excessive" and / or Level 3, etc.

[0039] For example, if the monitoring information on Road 1 shows that the garbage accumulation area at location A on Road 1 exceeds 2 square meters, or the garbage accumulation height exceeds 0.5 meters, or the number of garbage piles exceeds 3, the management platform can identify the garbage accumulation situation on Road 1 at location A as excessive and / or level 3.

[0040] For example, if the monitoring information on Road 2 shows that the garbage accumulation area at location B on Road 2 does not exceed 1 square meter, or the garbage accumulation height does not exceed 0.2 meters, or the number of garbage piles is 1, the management platform can identify the garbage accumulation situation on Road 2 at location B as less and / or Level 1.

[0041] Step 220: Based on the garbage accumulation situation, determine at least one target garbage collection point.

[0042] A target garbage collection point refers to a road where garbage needs to be collected. For example, a target garbage collection point could be a road with a garbage accumulation level of 2 or 3.

[0043] In some embodiments, the management platform can determine the target garbage collection point based on a first preset threshold. The first preset threshold refers to a pre-set range of garbage accumulation. For example, if the first preset threshold is 2 square meters, when the garbage accumulation (e.g., the amount of garbage) on road 1 at location A is greater than the first preset threshold of 2 square meters, the management platform can determine road 1 as the target garbage collection point.

[0044] In some embodiments, the management platform can preset a garbage accumulation lookup table to determine which roads are target garbage collection points. The garbage accumulation lookup table includes different rules for determining target garbage collection points for different roads. For example, the garbage accumulation lookup table specifies that if the garbage accumulation on road 1 at location A exceeds the normal level, road 1 is a target garbage collection point. For instance, when the garbage accumulation on road 1 exceeds the normal level, the management platform can identify road 1 as the target garbage collection point.

[0045] For more information on identifying target garbage collection points, please refer to [link / reference]. Figure 3 And its related descriptions.

[0046] Step 230: Determine the garbage collection route based on at least one target garbage collection point.

[0047] A garbage collection route refers to the route planning for cleaning up garbage at all target garbage collection points. For example, a garbage collection route can be the shortest route that passes through and cleans up garbage at all target garbage collection points in a sequential order.

[0048] In some embodiments, the management platform can determine a garbage collection route based on a cleaning route determination model, using at least one target garbage collection point. The cleaning route determination model can be a machine learning model. In some embodiments, the input to the cleaning route determination model can be at least one target garbage collection point, and the output can be the garbage collection route.

[0049] The parameters of the cleaning route determination model can be obtained through training. In some embodiments, the cleaning route determination model can be trained using multiple sets of cleaning route training samples labeled with cleaning route labels. For example, multiple sets of cleaning route training samples labeled with cleaning route labels can be input into the initial cleaning route determination model. A loss function is constructed using the cleaning route labels and the results of the initial cleaning route determination model. The parameters of the cleaning route determination model are then iteratively updated based on the loss function. When the loss function of the initial cleaning route determination model meets the set requirements, the model training is complete, and the trained cleaning route determination model is obtained. These set requirements may include loss function convergence, the number of iterations reaching a threshold, etc.

[0050] In some embodiments, each set of sweeping route training samples may include historical target garbage collection points on each road. Each set of sweeping route labels may be the actual garbage collection route corresponding to each set of sweeping route training samples. In some embodiments, the sweeping route training samples may be obtained based on historical target garbage collection points, and the sweeping route labels may be obtained through manual annotation.

[0051] For more information on determining garbage collection routes, please see [link / reference]. Figure 5 And its related descriptions.

[0052] By acquiring monitoring information on roads within a road network area through some embodiments of this specification, and identifying the garbage accumulation situation on the roads, the garbage accumulation situation can be obtained in real time. Based on the garbage accumulation situation, target garbage collection points can be obtained, and garbage collection routes can be determined. The garbage collection routes can be adjusted in real time according to the garbage accumulation situation, thereby improving road cleanliness, reducing repeated cleaning, and effectively saving manpower and material resources.

[0053] In some embodiments, the management platform can generate remote control commands based on the garbage collection route and send them to the sensor network platform.

[0054] Remote control commands refer to control instructions generated remotely by a computer. In some embodiments, an IoT system can be planned based on smart city garbage collection routes to automatically generate remote control commands. For example, if the garbage collection route is ABCD, remote control command 1 is automatically generated; if the garbage collection route is ACDB, remote control command 2 is automatically generated, and so on.

[0055] In some embodiments, the management platform can send remote control commands to the object platform based on the sensor network platform, so that the object platform can perform cleaning operations.

[0056] A cleaning operation refers to the actions related to cleaning up and removing garbage. For example, a cleaning operation could be sweeping away garbage, washing roads, etc. In some embodiments, the object platform can execute cleaning operations based on remote control commands. For example, remote control command 1 executing a cleaning operation could mean sweeping away garbage first and then washing the road, while remote control command 2 executing a cleaning operation could mean washing the road first and then sweeping away garbage, etc.

[0057] The remote control commands generated based on the garbage collection route, as described in some embodiments of this specification, are used to execute cleaning operations. This allows for targeted cleaning operations according to the garbage collection route, thereby improving garbage collection efficiency.

[0058] Figure 3 This is an exemplary flowchart illustrating the determination of at least one target waste collection point according to some embodiments of this specification. Figure 3 As shown, process 300 includes the following steps. In some embodiments, process 300 may be executed by a management platform.

[0059] Step 310: Obtain monitoring information on at least one road within the road network area and determine traffic flow information on at least one road.

[0060] Traffic flow information refers to the number of objects moving through a road per unit of time. For example, traffic flow information could be the number of vehicles or pedestrians crossing a road per unit of time. Traffic flow information can include pedestrian traffic. It can also include vehicle traffic, etc. For more information on vehicle traffic, see [link to relevant documentation]. Figure 4 And its related descriptions.

[0061] In some embodiments, the management platform can determine traffic flow information on roads based on monitoring information within the road network area using algorithms. For example, the management platform can use algorithms such as contour recognition, dynamic video tracking, and stereo vision to detect and identify objects of different shapes in monitoring information (e.g., video images) and obtain traffic flow information on roads in real time.

[0062] Pedestrian flow refers to the number of people passing through a road per unit of time. For example, 780 people passed through Road 3 between 8:00 and 9:00 AM. The pedestrian flow on Road 3 during this time period is 13 people per minute. In some embodiments, the management platform can determine pedestrian flow on roads based on monitoring information from roads within the road network area, using video algorithms. In some embodiments, the management platform can determine pedestrian flow on roads based on wireless access points. For example, using WiFi probes to identify the device information of passersby, thereby determining pedestrian flow on roads.

[0063] In some embodiments, traffic information may also include estimated pedestrian traffic at future times on each of at least one road.

[0064] Estimated pedestrian traffic refers to the predicted number of pedestrians crossing a road at a future time. A future time can be a point in time some distance from the current time. For example, a future time could be half an hour, an hour, or other times away from the current time. In some embodiments, the management platform can determine the estimated pedestrian traffic based on historical pedestrian traffic data for multiple historical times on each road using a pedestrian traffic prediction model. The pedestrian traffic volume will vary for different roads or even for different future times on the same road.

[0065] Historical pedestrian traffic refers to the number of people passing through a road at historical times. A historical time can refer to a time corresponding to a future time. For example, if the future time is 9:00, the historical time could be 9:00 corresponding to a past period of time (e.g., a week, half a month, a month, etc.). Another example is if the current time is 8:00, and the future time is 9:00, the historical time could be 6:00, 6:10, 6:20, ..., 7:50, 8:00, etc., on the same day. For more information on determining historical pedestrian traffic on roads, please refer to the above description of determining pedestrian traffic.

[0066] Pedestrian flow prediction models can be used to predict the estimated pedestrian flow on each road at future times. These models can be machine learning models. For example, they can be convolutional neural network (CNN) models, long short-term memory (LSTM) models, etc.

[0067] In some embodiments, the input to the pedestrian flow prediction model can be the historical pedestrian flow at multiple historical moments corresponding to future moments on a certain road, and the output can be the estimated pedestrian flow at future moments on that road.

[0068] The parameters of the pedestrian flow prediction model can be obtained through training. In some embodiments, the pedestrian flow prediction model can be trained using multiple sets of pedestrian flow training samples labeled with pedestrian flow characteristics. For example, multiple sets of pedestrian flow training samples labeled with pedestrian flow characteristics can be input into the initial pedestrian flow prediction model. A loss function is constructed using the pedestrian flow labels and the results of the initial pedestrian flow prediction model. The parameters of the pedestrian flow prediction model are then iteratively updated based on the loss function. When the loss function of the initial pedestrian flow prediction model meets the set requirements, the model training is complete, and a trained pedestrian flow prediction model is obtained. These set requirements may include loss function convergence, the number of iterations reaching a threshold, etc.

[0069] In some embodiments, each set of pedestrian flow training samples may include historical pedestrian flow data for each road at multiple historical moments corresponding to future moments. Each set of pedestrian flow labels may be the actual pedestrian flow data for the future moment corresponding to each set of pedestrian flow training samples. In some embodiments, pedestrian flow training samples may be obtained based on historical pedestrian flow data, and pedestrian flow labels may be obtained through manual annotation.

[0070] Based on historical pedestrian flow data from multiple historical moments on each road, as described in some embodiments of this specification, a pedestrian flow prediction model can be used to determine the estimated pedestrian flow, thus achieving intelligent pedestrian flow prediction.

[0071] Step 320: Based on the garbage accumulation and flow information, determine at least one target garbage collection point.

[0072] In some embodiments, the management platform can set a second preset threshold and a second preset condition. When the amount of garbage accumulation on a road exceeds the second preset threshold and the traffic flow information meets the second preset condition, the road is determined to be a target garbage collection point. The second preset threshold refers to another range of pre-set values ​​for garbage accumulation. The second preset condition can be a pre-set condition restricting traffic flow information on the road, where traffic flow information can include pedestrian flow and / or vehicle flow. For example, the second preset threshold is 2 square meters, and the second preset condition is pedestrian flow greater than 200 people / hour and / or vehicle flow greater than 200 vehicles / hour. When the amount of garbage accumulation on the road (e.g., the amount of garbage) exceeds the second preset threshold of 2 square meters, and / or the pedestrian flow exceeds the second preset condition of 200 people / hour and / or the vehicle flow exceeds 200 vehicles / hour, the management platform can determine the road as a target garbage collection point.

[0073] For example, the second preset threshold is 2 square meters, and the second preset condition is that the estimated pedestrian flow at future time t is greater than 300 people / hour and / or the estimated vehicle flow is greater than 300 vehicles / hour. When the amount of garbage accumulated on the road (e.g., the amount of garbage) exceeds the second preset threshold of 2 square meters, and the estimated pedestrian flow at future time t exceeds the second preset condition of 300 people / hour and / or the estimated vehicle flow is greater than 300 vehicles / hour, the management platform can determine that road as a target garbage collection point. For more information on vehicle flow, please refer to [link / reference needed]. Figure 4 Related descriptions.

[0074] By acquiring monitoring information on roads within a road network area through some embodiments described in this specification, and determining traffic flow information on the roads, the target garbage collection point can be determined in real time based on the actual traffic flow by combining garbage accumulation information with traffic flow information.

[0075] Figure 4 This is an exemplary schematic diagram of a traffic flow prediction model according to some embodiments of this specification.

[0076] In some embodiments, traffic information may also include vehicle traffic.

[0077] Traffic flow refers to the number of vehicles passing through a road within a unit of time period. For example, the number of vehicles passing through Road 6 between 8:00 and 9:00 AM is 300. The traffic flow of Road 6 during this time period is 5 vehicles per minute.

[0078] In some embodiments, the management platform can determine traffic flow on roads based on monitoring information from roads within the road network area using video algorithms. In some embodiments, the management platform can determine traffic flow on roads based on induction coils combined with monitoring information from roads within the road network area. For example, induction coils are buried underground in the road, and digital cameras are mounted on crossbars for monitoring; when vehicles are traveling on the road, traffic flow can be acquired in real time.

[0079] In some embodiments, traffic information may further include estimated traffic flow at future times for each road in at least one road. In some embodiments, the management platform may determine the estimated traffic flow based on historical traffic flow at multiple historical times for each road using a traffic flow prediction model, wherein the traffic flow prediction model is a machine learning model.

[0080] Estimated traffic volume 430 refers to the predicted traffic volume on the road at a future time. A future time can be a point in time some distance from the current time. In some embodiments, the management platform can determine the estimated traffic volume using a traffic volume prediction model based on historical traffic volumes at multiple historical times for each road. The traffic volume at different future times will vary for different roads or even on the same road.

[0081] Historical traffic flow 410 refers to the traffic flow on the road at a historical time. A historical time can refer to a time corresponding to a future time. For example, a future time might be 9:00, while a historical time might be 9:00 corresponding to a past period (e.g., a week, half a month, a month, etc.). For more information on determining historical traffic flow on roads, please refer to the above description of determining traffic flow.

[0082] Traffic flow prediction model 420 can be used to predict the estimated traffic flow at future times on each road. The traffic flow prediction model can be a machine learning model. For example, it can be a convolutional neural network (CNN) model, a long short-term memory (LSTM) model, etc.

[0083] In some embodiments, the input to the traffic flow prediction model can be the historical traffic flow of a road at multiple historical moments corresponding to future moments, and the output can be the estimated traffic flow of the road at future moments.

[0084] The parameters of the traffic flow prediction model can be obtained through training. In some embodiments, the traffic flow prediction model can be trained using multiple sets of traffic flow training samples labeled with traffic flow information. For example, multiple sets of traffic flow training samples labeled with traffic flow information can be input into the initial traffic flow prediction model. A loss function is constructed using the traffic flow labels and the results of the initial traffic flow prediction model. The parameters of the traffic flow prediction model are then iteratively updated based on the loss function. When the loss function of the initial traffic flow prediction model meets the set requirements, the model training is complete, and a trained traffic flow prediction model is obtained. These set requirements may include loss function convergence, the number of iterations reaching a threshold, etc.

[0085] In some embodiments, each set of traffic flow training samples may include historical traffic flow data for each road at multiple historical moments corresponding to future moments. Each set of traffic flow labels may be the actual traffic flow at the future moment corresponding to each set of traffic flow training samples. In some embodiments, traffic flow training samples may be obtained based on historical traffic flow data, and traffic flow labels may be obtained through manual annotation.

[0086] Based on historical traffic flow data from multiple historical moments on each road, as described in some embodiments of this specification, a traffic flow prediction model can be used to determine the estimated traffic flow, thus achieving intelligent traffic flow prediction.

[0087] Figure 5 This is an exemplary flowchart illustrating the determination of a garbage collection route according to some embodiments of this specification. Figure 5As shown, process 500 includes the following steps. In some embodiments, process 500 may be executed by a management platform.

[0088] In some embodiments, the management platform can determine the planned route from the i-th target garbage collection point back to the starting point as the garbage collection route.

[0089] The i-th target garbage collection point refers to any garbage collection point that needs to be processed. For example, the i-th target garbage collection point could be the initial garbage collection point on the garbage collection route, a garbage collection point in the middle of the route, or a garbage collection point at the end of the route. The value of i can be a natural number, such as 1, 2, 3, etc. The maximum value of i can be the number of target garbage collection points.

[0090] The starting point refers to the location where the cleaning equipment departs. In some embodiments, the management platform can determine the starting point location manually.

[0091] The optimal solution refers to the best solution selected from multiple garbage collection route options based on the principle of comparative selection. For example, the optimal solution could be the garbage collection route with the lowest cost.

[0092] Step 510: Determine whether the preset set meets the preset conditions.

[0093] In some embodiments, the management platform can determine whether a preset set meets preset conditions. Based on different determination results, the preferred solution is determined in different ways.

[0094] The preset set refers to the set of target garbage collection points that need to be processed. Specifically, the preset set excludes the starting point of the target garbage collection points. The preset set (excluding the starting point) can be represented by S. For example, S = {c1, ..., c...} n}, where c1 represents the first target garbage collection point, c n Let c represent the nth target garbage collection point (1≤i≤n), and the starting point can be represented as c0.

[0095] In some embodiments, the management platform may determine a preset set based on multiple target waste collection points.

[0096] Preset conditions refer to the pre-defined conditions for starting from the i-th target waste collection point and returning to the starting point. For example, preset conditions could be that the preset set is an empty set. When the preset set meets the preset conditions, it means the preset set is empty and there are no target waste collection points that need to be processed. For example, the preset set meets the preset conditions. At that time, starting from the i-th target garbage collection point c iYou can return directly to the starting point c0 without passing through any target garbage collection points.

[0097] In some embodiments, the management platform may be based on preset conditions determined manually.

[0098] In some embodiments, the management platform can determine whether the preset set meets the preset conditions by judging whether the preset set is an empty set.

[0099] Step 520: In response to the preset set satisfying the preset conditions, based on the first cost of returning from the i-th target garbage collection point to the starting point, determine the optimal solution and its planning cost for starting from the i-th target garbage collection point and returning to the starting point.

[0100] Cost refers to the expense incurred in traveling from one target trash collection point to another. Cost can include the cost represented by the distance traveled from one target trash collection point to another. The first cost refers to the cost incurred in returning to the starting point from the i-th target trash collection point when a preset set satisfies preset conditions. For example, the first cost could be the cost at target trash collection point c. i The cost represented by the distance directly to the starting point c0. Planning cost refers to the cost incurred by the optimal waste collection route.

[0101] In some embodiments, the management platform may determine the first cost based on the cost of traveling from the i-th target waste collection point to the starting point.

[0102] In some embodiments, in response to a preset set satisfying a preset condition, the planning cost of the preferred route from the i-th target garbage collection point back to the starting point can be expressed by formula (1):

[0103] Wherein, P(c ior0 S) refers to the distance from the i-th target garbage collection point c i Or, the planning cost of the optimal route from the starting point to all target garbage collection points in the preset set S and then back to the starting point c0. D(c ior0 c0) refers to the i-th target garbage collection point c i Alternatively, the cost of traveling from the starting point to the starting point c0 (i.e., the first cost). The management platform can obtain the distance from different target garbage collection points to the starting point through the distance matrix, and determine the corresponding first cost based on this distance. The distance matrix can represent the distance between the starting point and different target garbage collection points, as well as the distances between different target garbage collection points. The distance matrix can be a pre-set matrix.

[0104] In some embodiments, the first cost is also related to the estimated time to reach the target waste collection point. In some embodiments, the first cost is also related to the estimated rate of waste growth after collection at the target waste collection point. Details regarding the estimated time to reach the target waste collection point and the estimated rate of waste growth after collection at the target waste collection point are provided below.

[0105] Step 530: In response to the preset set not meeting the preset conditions, based on the comparison of multiple second costs, determine the optimal solution and its planning cost for starting from the i-th target garbage collection point and returning to the starting point.

[0106] The second cost refers to the cost incurred when returning to the starting point from the i-th target garbage collection point if the preset set does not meet the preset conditions. For example, the preset conditions are not an empty set. When returning to the starting point from the i-th target garbage collection point, the second cost can be based on the i-th target garbage collection point c. i Heading to the transit destination garbage collection point (e.g., the j-th transit destination garbage collection point c) j The cost represented by the distance from i to the starting point c0, where i ≠ j.

[0107] In some embodiments, the management platform can determine multiple second costs based on multiple reference schemes that lead from the i-th target waste collection point through multiple intermediate target waste collection points and back to the starting point. The multiple intermediate target waste collection points can have various configurations (e.g., different order of passage), and each configuration corresponds to a reference scheme and a second cost.

[0108] A transit target garbage collection point refers to one or more intermediate target garbage collection points along the route from the i-th target garbage collection point back to the starting point. For example, a transit target garbage collection point can be a target garbage collection point other than the i-th target garbage collection point and the starting point (e.g., the j-th transit target garbage collection point c). j etc., where i ≠ j).

[0109] In some embodiments, the management platform can determine the transit target garbage collection point based on the i-th target garbage collection point and the starting point using a transit target garbage collection point determination model. The transit target garbage collection point determination model can be a machine learning model. For example, the transit target garbage collection point can be determined based on a historical i-th target garbage collection point and a historical starting point, using a machine learning model. In some embodiments, the input to the transit target garbage collection point determination model can be different target garbage collection points and starting points, and the output can be the transit target garbage collection point.

[0110] The cost of traveling from the i-th target waste collection point to multiple transit target waste collection points can be represented by the cost of the distance traveled from the i-th target waste collection point to the multiple transit target waste collection points. For example, the cost of traveling from the i-th target waste collection point to the j-th transit target waste collection point can be the cost of traveling from the i-th target waste collection point c... i Heading to the j-th transit point, garbage collection point c j Costs, etc., represented by distance.

[0111] Multiple reference routes refer to reference routes for garbage collection from the i-th target garbage collection point to multiple intermediate target garbage collection points and from multiple intermediate target garbage collection points back to the starting point.

[0112] The second cost of the reference scheme can be expressed as D(c) ior0 ,c j )+P(c j ,Sc j ), where D(c ior0 ,c j ) refers to the i-th target garbage collection point c i Or, starting from c0, proceed to the transfer destination, garbage collection point c. j The cost, transit destination garbage collection point c j It is included in the predefined set S. Specifically, D(c ior0 ,c j The distance can be determined based on an n×n distance matrix, which includes elements representing the i-th target garbage collection point c. i Or the starting point c0 and the j-th target garbage collection point c j The distance between them. P(c) j ,SC j ) refers to the garbage collection point c from the j-th transfer destination. j The minimum cost to return to the starting point c0. The values ​​of i and j differ in different reference schemes, and their second cost D(c) varies. ior0 ,c j )+P(c j ,Sc j The value of ) will also be different.

[0113] In some embodiments, in response to a preset set not meeting a preset condition (such as... (At time), the preferred solution could be to start from the i-th target garbage collection point c. i Find the shortest garbage collection route that goes to multiple transit garbage collection points and then returns to the starting point c0. The planning cost can be the cost of the i-th transit garbage collection point c0. i The cost represented by the shortest distance from multiple transit destinations to the starting point c0.

[0114] In some embodiments, the management platform may determine the preferred route and its planning cost for starting from the i-th target waste collection point and returning to the starting point based on a comparison of multiple second costs.

[0115] In some embodiments, in response to a preset set not meeting a preset condition (such as... When the planning cost is (2), it can be expressed by formula (2):

[0116] in, This refers to the minimum of the multiple second costs corresponding to multiple reference schemes when j takes different values. For example, when the starting point is c0, and S = {c1, c2, c3}, formula (2) can be expressed as:

[0117] Among them, P(c1,S-c1), P(c2,S-c2), and P(c3,S-c3) can be further decomposed according to the above formula, and the process continues recursively until all values ​​can be directly obtained (based on formula (1)).

[0118] In some embodiments, the second cost is also related to the estimated time to reach the target waste collection point. In some embodiments, the second cost is also related to the estimated rate of waste growth after collection. Details regarding the estimated time to reach the target waste collection point and the estimated rate of waste growth after collection are provided below.

[0119] The planned route from the preferred scheme of starting from the i-th target garbage collection point and returning to the starting point, as described in some embodiments of this specification, can be determined as the garbage collection route. The optimal garbage collection route can be dynamically planned according to the target garbage collection point. The value of the garbage collection route can be improved based on multiple reference schemes and the actual situation of the target garbage collection point, thus saving manpower and resources.

[0120] In some embodiments, the first cost and / or the second cost are also related to the estimated time to reach the target garbage collection point. The estimated time to reach the target garbage collection point is related to the real-time traffic flow to one of the at least one target garbage collection point and the estimated traffic flow at a future time.

[0121] The estimated time to reach the target garbage collection point refers to the predicted time from the current location to the target garbage collection point. For example, the estimated time to reach the target garbage collection point could be 10 minutes, 30 minutes, etc.

[0122] In some embodiments, the estimated time to reach a target garbage collection point is related to the real-time traffic flow to one of the at least one target garbage collection point and the estimated traffic flow at a future time. For example, a garbage truck may currently choose to go to target garbage collection point A or target garbage collection point B. The management platform determines, based on real-time traffic flow, that the route to target garbage collection point A is congested at the current time, while the route to target garbage collection point B is relatively uncongested. Therefore, a longer estimated time to reach target garbage collection point A results in a higher initial cost, while a shorter estimated time to reach target garbage collection point B results in a lower initial cost.

[0123] For example, the management platform determines, based on estimated traffic flow in the future, that the road to target garbage collection point A will become clear in 10 minutes, while the road to target garbage collection point B will be congested. Therefore, the estimated time to reach target garbage collection point A in 10 minutes is shorter, resulting in a lower initial cost; conversely, the estimated time to reach target garbage collection point B in 0 minutes is longer, resulting in a higher initial cost. Thus, traveling to target garbage collection point A first, and then to target garbage collection point B, in 10 minutes will avoid the traffic congestion.

[0124] In some embodiments, the first cost and / or second cost related to the estimated time to reach the target garbage collection point can dynamically change over time. For example, at the current time t0, the garbage truck is currently located at the target garbage collection point A. The estimated time t1 for traveling from target garbage collection point A to target garbage collection point B is predicted. Assuming that cleaning at target garbage collection point B takes 1 hour, when the garbage truck finishes cleaning at target garbage collection point B and travels from target garbage collection point B to target garbage collection point C, the time at this point is t0 + t1 + 1 h. If it is necessary to predict the estimated time t2 for traveling from target garbage collection point B to target garbage collection point C, that is, to determine the traffic flow from target garbage collection point B to target garbage collection point C at time t0 + t1 + 1 h, the estimated time t2 changes with the estimated time t1.

[0125] The first and / or second costs described in some embodiments of this specification are also related to the estimated time to reach the target garbage collection point. The garbage collection sequence can be better determined based on the estimated time, thereby reducing the time spent on congested roads and making it more convenient and time-saving to determine the garbage collection route by taking into account actual road conditions.

[0126] In some embodiments, the first cost and / or the second cost are also related to the estimated rate of garbage growth after reaching the target garbage collection point.

[0127] The estimated rate of increase in waste after cleaning refers to the estimated rate at which waste will increase at a cleaning point in the future. For example, the estimated rate of increase in waste at cleaning point A after cleaning could be 100 kg / 2h.

[0128] In some embodiments, the estimated rate of waste growth after cleaning can be obtained by processing and predicting the rate of waste growth and the waste accumulation situation before cleaning based on a growth rate prediction model.

[0129] In some embodiments, the first cost and / or second cost of the estimated garbage growth rate after reaching the target garbage collection point can dynamically change over time. For example, at 9:00 AM, the management platform can choose to go to either target garbage collection point A or target garbage collection point B first. If it goes to target garbage collection point A first, the estimated garbage growth rate after reaching target garbage collection point A (e.g., after 10:00 AM) can be determined. Based on the time of arrival at target garbage collection point B (e.g., arriving at 10:30 AM) and the time required to complete the cleaning (e.g., 1.5 hours), the estimated garbage growth rate after reaching target garbage collection point B (e.g., after 12:00 PM) can be determined.

[0130] The management platform can determine the first cost / second cost based on the estimated garbage growth rate after cleaning at two target garbage collection points. The platform can also determine the estimated garbage growth rate after cleaning target garbage collection point B first, followed by target garbage collection point A, and determine the corresponding first cost / second cost based on the above method. Since the cleaning time for each target garbage collection point differs in the two methods, the estimated garbage growth rate after cleaning also differs, thus affecting the first cost / second cost differently.

[0131] Growth rate prediction models can be used to predict the rate of garbage growth at future points after road sweeping. Growth rate prediction models can be machine learning models.

[0132] In some embodiments, the input to the growth rate prediction model can be the waste growth rate and waste accumulation status before cleaning, and the output can be the estimated waste growth rate after cleaning. The estimated time corresponding to the estimated waste growth rate after cleaning is the estimated future time after cleaning. For example, if the current time is 08:00, and the estimated time after cleaning is 09:00, then the estimated waste growth rate after cleaning is the waste growth rate corresponding to the future time 09:00.

[0133] The parameters of the growth rate prediction model can be obtained through training. In some embodiments, the growth rate prediction model can be trained using multiple sets of predicted rate training samples labeled with the predicted rate. For example, multiple sets of predicted rate training samples labeled with the predicted rate can be input into the initial growth rate prediction model. A loss function is constructed using the predicted rate labels and the results of the initial growth rate prediction model. The parameters of the growth rate prediction model are then iteratively updated based on the loss function. When the loss function of the initial growth rate prediction model meets the set requirements, the model training is complete, and a trained growth rate prediction model is obtained. These set requirements may include loss function convergence, the number of iterations reaching a threshold, etc.

[0134] In some embodiments, each set of estimated rate training samples may include the waste growth rate before cleaning and the waste accumulation status before cleaning. Each estimated rate label may be the actual waste growth rate after cleaning at a future time corresponding to each set of estimated rate training samples. In some embodiments, the estimated rate training samples may be obtained based on historical waste growth rates and waste accumulation status before cleaning, and the estimated rate labels may be obtained through manual annotation.

[0135] The growth rate prediction model described in some embodiments of this specification processes and predicts the garbage growth rate and garbage accumulation situation before cleaning to obtain the estimated garbage growth rate after cleaning. For dirty road environments, timely cleaning can be carried out to avoid garbage accumulation. Combining the estimated garbage growth rate after cleaning as a relevant quantity at a cost can make the determination of garbage cleaning routes more accurate.

[0136] In some embodiments, the inputs to the growth rate prediction model also include the pedestrian traffic at the target garbage collection point and the estimated pedestrian traffic at the target garbage collection point at future times.

[0137] The pedestrian flow at a target garbage collection point refers to the number of people passing through the target garbage collection point per unit of time. For example, if 3,000 people pass through target garbage collection point A between 8:00 AM and 12:00 PM, the pedestrian flow at target garbage collection point A during that time period is 750 people / hour or 13 people / minute. For more information on determining pedestrian flow, please see [link to relevant documentation]. Figure 2 And its related descriptions.

[0138] The estimated future pedestrian flow at a target waste collection point refers to the pre-estimated number of people who may pass through the target waste collection point at a future time. In some embodiments, the management platform can determine the estimated pedestrian flow based on historical pedestrian flow data for multiple historical times at each target waste collection point using a pedestrian flow prediction model. For more information on estimated pedestrian flow and pedestrian flow prediction models, please refer to [link to relevant documentation]. Figure 3 And its related descriptions.

[0139] In some embodiments, each set of estimated rate training samples also includes the pedestrian flow at the target garbage collection point and the estimated pedestrian flow at the target garbage collection point at future times.

[0140] The growth rate prediction model described in some embodiments of this specification uses inputs including pedestrian traffic at the target garbage collection point and estimated pedestrian traffic at the target garbage collection point at future times. This allows the growth rate prediction model to more accurately determine the estimated garbage growth rate at the target garbage collection point after cleaning. Furthermore, timely cleaning of roadside garbage can prevent garbage accumulation due to the herd effect among pedestrians, effectively maintaining road cleanliness.

[0141] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0142] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0143] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0144] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0145] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0146] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0147] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A smart city garbage cleaning route planning method, characterized in that, The method is executed through the management platform of a smart city waste collection route planning IoT system, and the method includes: Acquire monitoring information on at least one road within the road network area, and identify the garbage accumulation situation on the at least one road; Based on the monitoring information, traffic flow information on the at least one road is determined. The traffic flow information includes pedestrian flow, vehicle flow, and estimated pedestrian flow at future times on each of the at least one road. The estimated pedestrian flow at future times on each of the at least one road is determined by a pedestrian flow prediction model, which is a machine learning model. The input of the pedestrian flow prediction model includes historical pedestrian flow at multiple historical times corresponding to the future times on each road, and the output is the estimated pedestrian flow. Based on the garbage accumulation situation and the flow information, at least one target garbage collection point is determined; Based on the at least one target garbage collection point, a garbage collection route is determined; the determination of the garbage collection route based on the at least one target garbage collection point includes: The planned route from the preferred scheme of starting from the i-th target garbage collection point and returning to the starting point is determined as the garbage collection route, wherein determining the preferred scheme of starting from the i-th target garbage collection point and returning to the starting point includes: In response to the preset set satisfying the preset conditions, based on the first cost of returning from the i-th target garbage collection point to the starting point, the preferred scheme and its planning cost of starting from the i-th target garbage collection point and returning to the starting point are determined, wherein the preset set is the set of target garbage collection points excluding the starting point; In response to the preset set not satisfying the preset condition, based on the comparison of multiple second costs, the preferred scheme and its planning cost for starting from the i-th target garbage collection point and returning to the starting point are determined. The multiple second costs are determined based on the cost of traveling from the i-th target garbage collection point to multiple transfer target garbage collection points and multiple reference schemes for starting from multiple transfer target garbage collection points and returning to the starting point. The first cost and / or the second cost are related to the real-time traffic flow to the target garbage collection point, the estimated traffic flow at future times, and the estimated garbage growth rate after cleaning. The estimated garbage growth rate after cleaning is related to the pedestrian flow at the target garbage collection point and the estimated pedestrian flow at the target garbage collection point at future times.

2. The method of claim 1, wherein, The smart city garbage collection route planning IoT system also includes: a user platform, a service platform, a sensor network platform, and an object platform; The service platform is used to send the garbage collection route to the user platform; The object platform is used to acquire monitoring information on at least one road within the road network area and transmit it to the management platform through the sensor network platform; The method further includes: Based on the garbage cleaning route, a remote control command is generated and sent to the sensor network platform. The remote control command is then sent to the object platform via the sensor network platform, so that the object platform can perform a cleaning operation.

3. A smart city garbage cleaning route planning Internet of Things system, characterized in that, include: User platform, service platform, management platform, sensor network platform, and object platform; The service platform is used to send the garbage collection route to the user platform; The object platform is used to acquire monitoring information on at least one road within the road network area and transmit it to the management platform through the sensor network platform; The management platform is used for: Acquire monitoring information on at least one road within the road network area, and identify the garbage accumulation situation on the at least one road; Based on the monitoring information, traffic flow information on the at least one road is determined. The traffic flow information includes pedestrian flow, vehicle flow, and estimated pedestrian flow at future times on each of the at least one road. The estimated pedestrian flow at future times on each of the at least one road is determined by a pedestrian flow prediction model, which is a machine learning model. The input of the pedestrian flow prediction model includes historical pedestrian flow at multiple historical times corresponding to the future times on each road, and the output is the estimated pedestrian flow. Based on the garbage accumulation situation and the flow information, at least one target garbage collection point is determined; Based on the at least one target garbage collection point, determine the garbage collection route; Based on the garbage collection route, a remote control command is generated and sent to the sensor network platform; The sensor network platform sends the remote control command to the object platform so that the object platform can perform a cleaning operation. The management platform is further used for: The planned route from the preferred scheme of starting from the i-th target garbage collection point and returning to the starting point is determined as the garbage collection route, wherein determining the preferred scheme of starting from the i-th target garbage collection point and returning to the starting point includes: In response to the preset set satisfying the preset conditions, based on the first cost of returning from the i-th target garbage collection point to the starting point, the preferred scheme and its planning cost of starting from the i-th target garbage collection point and returning to the starting point are determined, wherein the preset set is the set of target garbage collection points excluding the starting point; In response to the preset set not satisfying the preset condition, based on the comparison of multiple second costs, the preferred scheme and its planning cost for starting from the i-th target garbage collection point and returning to the starting point are determined. The multiple second costs are determined based on the cost of traveling from the i-th target garbage collection point to multiple transfer target garbage collection points and multiple reference schemes for starting from multiple transfer target garbage collection points and returning to the starting point. The first cost and / or the second cost are related to the real-time traffic flow to the target garbage collection point, the estimated traffic flow at future times, and the estimated garbage growth rate after cleaning. The estimated garbage growth rate after cleaning is related to the pedestrian flow at the target garbage collection point and the estimated pedestrian flow at the target garbage collection point at future times.

4. A smart city garbage cleaning route planning device, characterized in that, The device includes at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is configured to execute at least a portion of the computer instructions to implement the smart city waste cleaning route planning method as described in any one of claims 1 to 2.

5. A computer-readable storage medium storing computer instructions, wherein when a computer reads the computer instructions in the storage medium, the computer executes the smart city garbage cleaning route planning method as described in any one of claims 1 to 2.

Images