CN117963375A - Mutual-aid sharing multipurpose garbage can processing system - Google Patents

Abstract

The invention relates to a mutual-aid sharing multipurpose garbage can processing system. The system comprises an intelligent garbage can, a mobile application program, a cloud data management center and a transport vehicle. The intelligent garbage bin accurately classifies the input garbage through integrated image recognition, microwave detection, weight and volume sensing technologies, automatically shunts the garbage into a preset container and simultaneously transmits detailed information of the garbage to a cloud. The mobile application program provides garbage can positioning, garbage throwing record and reservation recycling service of oversized articles for users, and user experience and participation are enhanced. The cloud data management center is responsible for processing data from the intelligent garbage can and the application program, carrying out garbage storage information statistics, and optimizing a dispatching plan of the transport vehicle based on the information and the reservation recovery requirement of a user. And the transport vehicle receives a scheduling instruction issued by the cloud and executes garbage collection and reservation recovery tasks. The system realizes the high efficiency and the intellectualization of garbage disposal and promotes the sharing and the cyclic utilization of resources.

Description

Mutual-aid sharing multipurpose garbage can processing system

Technical Field

The invention relates to the technical field of environmental protection, in particular to a mutual-assistance sharing multipurpose garbage can processing system.

Background

In the current urban garbage disposal system, accurate garbage classification is not only a necessary condition for improving the resource recovery efficiency, but also a key factor for realizing the environment sustainability. Despite the increasing importance of garbage classification, existing techniques and methods face significant challenges in practice, particularly in ensuring garbage classification accuracy. Most current systems rely on manual sorting and basic mechanical sorting techniques, which have various limitations.

First, manual classification is limited by the accuracy of manual operations. Secondly, although the simple mechanical sorting system improves the sorting speed, the identification capability of the mechanical sorting system is limited, and garbage with similar appearance and different materials, such as different types of plastics and composite materials, cannot be accurately distinguished. This technical limitation results in a high error rate, which in turn affects the subsequent resource recovery and reuse process.

The defects of the traditional methods not only reduce the overall efficiency of garbage classification, but also cause resource waste and environmental pollution because a lot of recyclable resources are misplaced to a landfill or an incineration plant due to inaccurate classification. Misclassified waste disposal adds additional economic burden, including increased disposal costs and loss of resource value, as well as increased greenhouse gas emissions and hazardous material emissions, pose a threat to environmental sustainability.

Therefore, it is necessary to develop a new garbage classification system.

Disclosure of Invention

The application provides a mutual-assistance sharing multi-purpose garbage can treatment system, which is used for improving the accuracy and efficiency of garbage treatment.

The application provides a mutual-aid sharing multipurpose garbage can processing system, which comprises:

The intelligent garbage can is provided with a unified throwing port and is used for processing the thrown garbage by using an image processing technology, a microwave technology, a weight sensor and a volume scanner to obtain classification information, weight, volume and throwing time of the thrown garbage; according to the classification information, the input garbage is shunted into corresponding containers preset in the intelligent garbage can through an internal mechanical structure of the intelligent garbage can; the classification information, weight, volume and throwing time of the throwing garbage are sent to a cloud data management center;

The mobile application program is interacted with the cloud data management center and is used for displaying the position of the nearest intelligent garbage can to the user and recording the type, weight and volume information and the throwing time of garbage thrown by the user; allowing a user to conduct reservation recovery application of the oversized article, and sending the reservation recovery application to a cloud data management center, wherein the reservation recovery application comprises the type, weight, volume, reservation recovery position and reservation recovery time of the oversized article;

The cloud data management center is used for receiving classification information, weight, volume and throwing time of throwing garbage from the intelligent garbage bin; counting the classification information, weight, volume and throwing time of the throwing garbage to obtain garbage storage information of the intelligent garbage can; receiving a reservation recovery application sent by a mobile application program; determining a dispatching instruction of the transport vehicle according to the garbage storage information and the reservation recovery application, wherein the dispatching instruction comprises an optimal route of the transport vehicle; transmitting the scheduling instruction to a transport vehicle;

and the transport vehicle is used for receiving a scheduling instruction of the cloud data management center, and carrying out garbage collection of the intelligent garbage can and reservation recovery of oversized articles according to the scheduling instruction.

Furthermore, the intelligent garbage can adopts a trained hybrid neural network to analyze data acquired by an image recognition technology, a microwave technology, a weight sensor and a volume scanner so as to obtain classification information of input garbage; the hybrid neural network comprises a multi-scale image recognition branch, a microwave-material sensing branch, a weight-volume characteristic coding branch, a time sequence-putting mode branch and a multi-mode data fusion and decision layer;

The multi-scale image recognition branch adopts a multi-scale convolutional neural network architecture and is used for processing image data of garbage to obtain an image feature vector; the microwave-material sensing branch adopts a framework based on a graph attention network, the acquired microwave signal data are regarded as nodes in the graph, and the relationship among the nodes is learned by using an attention mechanism to obtain a microwave reflection characteristic vector; the weight-volume characteristic coding branch uses a self-encoder network to perform characteristic extraction on collected garbage weight and volume data to obtain a physical attribute characteristic vector; the time sequence-putting mode branch adopts a time sequence processing model based on a transformer model, and the acquired putting time data is processed to obtain a putting mode feature vector; the multi-mode data fusion and decision layer processes the image feature vector, the microwave reflection feature vector, the physical attribute feature vector and the throwing mode feature vector by using an attention mechanism and a full connection layer to obtain the classification information of garbage.

Furthermore, the training step of the hybrid neural network is performed by adopting a staged method, and specifically comprises the following steps:

in an independent training stage, respectively and independently training a multi-scale image recognition branch, a microwave-material sensing branch, a weight-volume characteristic coding branch and a time sequence-put mode branch, so as to ensure that each branch can effectively process the corresponding data type and extract the related characteristics;

In the end-to-end training phase, the whole hybrid neural network performs end-to-end training as a whole.

Still further, in a separate training phase, the multi-scale image recognition branches employ cross entropy loss functions as loss functions; the microwave-material perception branch adopts a cross entropy loss function as a loss function; the weight-volume characteristic coding branch adopts a mean square error loss function as a loss function; the time series-put mode branch employs a cross entropy loss function as the loss function.

Further, the mobile application program allows a user to submit a reservation recycling application of the oversized article by filling in an electronic form or uploading a photo, and receives recycling confirmation and reservation time schedule of the cloud data management center in real time.

Still further, the mobile application actively pushes nearby upcoming garbage collection activity information or alerts the user to participate in periodic reclamation activities according to the user's location and preference settings.

Furthermore, the cloud data management center performs data analysis on the received garbage classification information, weight, volume and throwing time so as to predict the garbage generation trend and the emptying period of the intelligent garbage can, thereby optimizing the whole garbage collection and treatment flow.

Further, the cloud data management center adopts an optimization algorithm to determine an optimal path of the transport vehicle; the optimization algorithm is shown in the following equation 1:

Wherein S _opt represents an optimal score representing the selected path; n represents the total number of considered paths; d _i denotes the total distance of the ith path; v _i denotes the average vehicle speed of the i-th path; t _i denotes the estimated travel time of the i-th path; w _i denotes the congestion index of the i-th path; c _i represents the carbon emission cost of the ith path; e _i represents the urgency score for the ith path; omega ₁、ω₂、ω₃、ω₄ and omega ₅ are weight coefficients; α ₁、α₂、α₃、α₄ and α ₅ are tuning parameters; τ is a time sensitivity threshold for adjusting the weight of the urgent task.

Still further, the transport vehicle is equipped with GPS positioning and mobile communication technology to report vehicle location, status and garbage collection progress to a cloud data management center in real time to facilitate real-time monitoring and adjustment of the transport plan.

The application has the following beneficial technical effects:

(1) Improving the accuracy of garbage classification: by combining an image recognition technology, a microwave technology, a weight sensor and a volume scanner, the system can efficiently and accurately recognize and classify the input garbage. Compared with the traditional method relying on manual sorting or simple mechanical sorting, the comprehensive technical scheme greatly improves the accuracy of garbage sorting and lays a solid foundation for recycling and reutilizing resources.

(2) Optimizing garbage collection and transportation processes: the cloud data management center in the system can receive and process data from the intelligent garbage can in real time, and the data comprise garbage type, weight, volume and throwing time information. These data are used to optimize the scheduling instructions of the transport vehicle, including optimal routes and schedules, thereby improving the efficiency of garbage collection and transportation, reducing energy consumption and operating costs.

(3) Enhancing user engagement and experience: through mobile application, the user can easily find the nearest intelligent garbage bin position, can also track own garbage throwing record, and even carry out reservation recovery application of oversized articles. The interaction mode encourages the user to actively participate in garbage classification and recycling processes, and meanwhile, the environmental awareness and satisfaction degree of the user are improved.

(4) Facilitating resource sharing and recycling: the system promotes the efficient utilization of garbage recycle through accurate garbage classification and effective resource allocation, reduces the proportion of landfill and incineration, is beneficial to the sharing and recycling of resources, and has important significance for environmental protection and sustainable development.

Drawings

Fig. 1 is a schematic diagram of a processing system for a mutual-sharing multi-purpose garbage can according to a first embodiment of the present application.

Detailed Description

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. The present application may be embodied in many other forms than those herein described, and those skilled in the art will readily appreciate that the present application may be similarly embodied without departing from the spirit or essential characteristics thereof, and therefore the present application is not limited to the specific embodiments disclosed below.

The first embodiment of the application provides a mutual-assistance sharing multi-purpose garbage can processing system. Referring to fig. 1, a schematic diagram of a first embodiment of the present application is shown. A first embodiment of the present application is described in detail below with reference to fig. 1.

The mutual-assistance sharing multi-purpose garbage can processing system comprises an intelligent garbage can 101, a mobile application program 102, a cloud data management center 103 and a transport vehicle 104.

The intelligent garbage can 101 is provided with a unified throwing port and is used for processing the thrown garbage by using an image processing technology, a microwave technology, a weight sensor and a volume scanner to obtain classification information, weight, volume and throwing time of the thrown garbage; according to the classification information, the input garbage is shunted into corresponding containers preset in the intelligent garbage can through an internal mechanical structure of the intelligent garbage can; the classification information, weight, volume and throwing time of the throwing garbage are sent to a cloud data management center;

The intelligent garbage can 101 is a core component of the mutually-assisted sharing multi-purpose garbage can processing system, is exquisite in design and comprehensive in function, and aims to improve the accuracy and efficiency of garbage classification through high-tech means. The intelligent garbage can is provided with a unified putting port, so that a user can put garbage conveniently. The intelligent garbage can comprehensively analyze the input garbage through an integrated image processing technology, a microwave technology, a weight sensor and a volume scanner, so that the garbage type, the weight, the volume and the input time can be accurately identified.

Image processing technology enables intelligent garbage cans to judge the kinds of garbage, such as plastic, paper, metal or harmful garbage, by analyzing the appearance characteristics of the garbage. The microwave technology is further used for assisting in classification by analyzing the material characteristics of the garbage, and is particularly important for garbage with similar appearance but different materials. The weight sensor and the volume scanner respectively measure the weight and the volume of the garbage, and provide necessary data support for subsequent treatment.

The mechanical structure inside the intelligent garbage can automatically shunts garbage into different containers preset in the can according to the classification information obtained through various sensors and technologies. The process not only reduces the manual classification requirement, but also greatly improves the classification accuracy and efficiency.

And the communication module arranged in the intelligent garbage can sends the data such as the classification information, the weight, the volume, the throwing time and the like to the cloud data management center in real time. The data are not only used for optimizing garbage disposal and resource recovery, but also provide precious data support for urban garbage management.

To ensure that the smart trash can be implemented, the following are some key implementation details:

Image recognition technology: and capturing the garbage image by adopting a high-resolution camera, and analyzing by matching with a deep learning algorithm to identify the garbage type.

Microwave technology: the material composition of the refuse is determined by emitting microwaves to the refuse and analyzing the variation of the reflected signal.

Weight sensor and volume scanner: the weight and volume measurements were made using pressure and infrared sensors, respectively.

Internal mechanical structure: the waste is diverted to designated receptacles by means of motorized slide rails, sorting arms, or other mechanical means based on the sensor data.

The intelligent garbage can fully considers the usability, accuracy and data communication requirements, and ensures the effective application of the intelligent garbage can in a modern urban garbage management system.

Furthermore, the intelligent garbage can adopts a trained hybrid neural network to analyze data acquired by an image recognition technology, a microwave technology, a weight sensor and a volume scanner so as to obtain classification information of input garbage; the hybrid neural network comprises a multi-scale image recognition branch, a microwave-material sensing branch, a weight-volume characteristic coding branch, a time sequence-putting mode branch and a multi-mode data fusion and decision layer;

The multi-scale image recognition branch adopts a multi-scale convolutional neural network architecture and is used for processing image data of garbage to obtain an image feature vector; the microwave-material sensing branch adopts a framework based on a graph attention network, the acquired microwave signal data are regarded as nodes in the graph, and the relationship among the nodes is learned by using an attention mechanism to obtain a microwave reflection characteristic vector; the weight-volume characteristic coding branch uses a self-encoder network to perform characteristic extraction on collected garbage weight and volume data to obtain a physical attribute characteristic vector; the time sequence-putting mode branch adopts a time sequence processing model based on a transformer model, and the acquired putting time data is processed to obtain a putting mode feature vector; the multi-mode data fusion and decision layer processes the image feature vector, the microwave reflection feature vector, the physical attribute feature vector and the throwing mode feature vector by using an attention mechanism and a full connection layer to obtain the classification information of garbage.

In a mutual-aid sharing multi-purpose garbage can processing system, one of the core technologies of the intelligent garbage can is that the intelligent garbage can comprehensively analyze data from various sensors through a hybrid neural network so as to accurately acquire the classification information of garbage. This process involves highly complex data processing and intelligent decision making, and a detailed description of the implementation of each part of this hybrid neural network follows.

The multi-scale image recognition branch utilizes a multi-scale convolutional neural network (MS-CNN) architecture to extract image features through rolling and pooling operations of different scales aiming at captured garbage images. To accommodate different sizes and shapes of junk items, the architecture employs a Spatial Pyramid Pooling (SPP) layer, allowing the network to process arbitrary sized images without rescaling. In practice, the network may be designed by selecting the appropriate convolution kernel size, step size, and activation function, and by stacking multiple convolution layers and pooling layers. Training such a network requires a large amount of annotated spam image data to ensure that the network can learn enough image features for subsequent spam classification.

The Python reference implementation code of the multi-scale image recognition branch is as follows:

The microwave-material sensing branch adopts a structure based on a graph attention network (GAT) and specially processes microwave reflection signals reflecting garbage material information. In this architecture, the microwave signals are considered nodes in the graph, and the relationships between the nodes are defined by the interaction of the microwave signals. The attention mechanism is used for giving different weights to different nodes and highlighting the signal characteristics which are most important for garbage material classification. Implementing the branch requires constructing a graph network capable of processing node (signal) data and learning how to effectively apply the attention mechanism to extract the material features through training data.

The Python reference implementation code of the microwave-material aware branch is as follows:

The weight-volume feature encoding branch is implemented with a self-encoder network, which performs advanced feature extraction on the weight and volume data of the garbage. The input data is compressed from the encoder to a low-dimensional representation by an encoder, and then the decoder attempts to reconstruct the input data, by which process the compressed representation of the data is learned. In the garbage classification task, only the encoded low-dimensional feature representation is required to be used as the physical attribute feature vector. This requires the design of a suitable network structure and training through the weight and volume data of the large amount of garbage.

The Python reference implementation code of the weight-volume feature encoding branch is as follows:

The time sequence-throwing mode branch adopts a time sequence processing model based on a transformer model to process the data of garbage throwing time. The transformer model learns long-distance dependency in the time sequence through a self-attention mechanism, so that the throwing mode is identified. Implementing this branch requires appropriate preprocessing of the delivery time data and designing the transformer network structure, such as selecting an appropriate number of attention headers and multi-layer network depth, in order to capture and learn patterns in the time data.

The Python reference implementation code of the time series-put mode branch is as follows:

And in a multi-mode data fusion and decision layer, the hybrid neural network synthesizes and fuses the output characteristic vectors of all branches. The layer uses an attention mechanism to strengthen key features and restrain irrelevant information, and then processes the fused features through a full-connection layer to obtain a final garbage classification decision. Implementing this layer requires developing a network architecture that can handle multiple types of features and effectively fuse these features, while selecting appropriate penalty functions and optimizers to train the network for accurate classification decisions.

The Python reference implementation code of the multi-mode data fusion and decision layer is as follows:

model # construction

Num_ classes =5# assuming 5 garbage classifications

hybrid_model＝create_hybrid_model(image_shape,microwave_input_shape,weight_volume_shape,time_series_shape,num_classes)

hybrid_model.summary()

Note that the above code is a schematic frame. In practical applications, the network structure, the layer number and the parameters need to be adjusted according to specific data sets, task demands and hardware resources. In addition, the implementation of the steps of attention mechanism, model training, tuning, verification and the like are also key parts of the successful implementation of the hybrid neural network.

The training process of the hybrid neural network, including how to prepare data, configure the training environment, perform training steps, and how to verify and evaluate model performance, will be described in detail below.

1. Data preparation:

First, the necessary data sets are collected, which should include image data of the refuse, microwave signal data related to the material of the refuse, weight and volume data of the refuse, and time series data of the refuse delivery. These data need to be appropriately labeled to indicate the garbage classification to which each data point corresponds.

The image data needs to be standardized so that it has a uniform size and color range.

The microwave signal data may need to be pre-processed and converted into a format suitable for the attention network processing of the graph.

The weight and volume data should be normalized to facilitate network learning.

The time series data may need to be converted into a fixed length series to accommodate the input requirements of the transformer model.

2. Configuring a training environment:

a suitable deep learning framework and hardware environment are selected to train the model. TensorFlow or PyTorch are both good choices that provide the rich libraries and APIs needed to build complex models. Depending on the complexity of the model and the size of the data volume, it may be desirable to use computing resources with high performance GPUs to accelerate the training process.

3. The training steps are executed:

A staged approach may be taken when training the hybrid neural network. First, the model of each branch is trained separately, ensuring that each branch can efficiently process its corresponding data type and extract relevant features. This stage can use standard supervised learning methods, choosing the appropriate penalty function and optimizer to optimize the network weights.

For multi-scale image recognition branches, the cross entropy loss function may be used to train the image classification task.

The microwave-texture-aware branch, the weight-volume feature encoding branch and the time-series-put-mode branch also need to select an appropriate loss function according to their task characteristics.

In hybrid neural networks, the selection of an appropriate loss function for training of the microwave-material sensing branch, the weight-volume feature coding branch, and the time series-delivery mode branch is critical because it directly affects the efficiency and final performance of model learning. The task characteristics of each branch are different, so that a corresponding loss function needs to be selected according to the characteristics.

For the microwave-material sensing branch, the task is to identify the material of the garbage according to the reflection characteristics of the microwave signal. This can be seen as a classification problem, where each material type is a class. Thus, for this branch, a Cross entropy loss function (Cross-Entropy Loss) may be used, which is a common choice in dealing with multiple classes of classification problems.

Weight-volume feature encoding branching uses a self-encoder to extract advanced features of the weight and volume data of the garbage. The goal of the self-encoder is to minimize the difference between the input data and the reconstructed data, so a mean square error loss function (Mean Squared Error, MSE) can be used to quantify this difference.

The task of the time series-delivery mode branch is to learn the mode of garbage delivery time. This task can be seen as a sequence prediction problem, where the model needs to predict the next value or class of the time sequence. If the task is to predict future delivery pattern categories, the cross entropy loss function may be used as well.

In performing training, a comprehensive loss function needs to be set for the entire hybrid neural network, which typically involves a weighted sum of the individual branch loss functions. The weighting method needs to be adjusted according to the importance of each task and its contribution to the accuracy of the final garbage classification.

Through careful selection and adjustment of the loss function, the hybrid neural network can be effectively guided to learn the capability of extracting useful information from various sensor data, and garbage classification can be accurately performed.

After each branch is independently trained and achieves satisfactory performance, the branches are integrated together, and final garbage classification is performed through a multi-mode data fusion and decision layer. At this stage, the entire network will be trained end-to-end as a whole, possibly requiring adjustments to learning rates and other super parameters to optimize the performance of the overall model.

4. Verification and evaluation:

the performance of the model is tested using the validation dataset, and the evaluation index may include classification accuracy, recall, and F1 score, among others. The performance of the model on garbage classification tasks can be known through the indexes. In addition, further analysis, such as misclassification analysis, of the model may be performed to identify weaknesses of the model and improve it.

5. Continuous optimization:

Depending on the evaluation, it may be necessary to go back to the steps of data preparation and model training for iterative optimization. This may include increasing the amount of data, adjusting the model structure, or changing training strategies, etc. Continuous optimization is key to improving model performance.

Through the detailed description of the steps, the training process of the hybrid neural network can be implemented, so that efficient and accurate garbage classification is realized in a mutually-assisted shared multi-purpose garbage can processing system.

The mobile application 102 is interacted with the cloud data management center and is used for displaying the position of the nearest intelligent garbage can to the user and recording the type, weight and volume information and the putting time of garbage put by the user; allowing a user to conduct reservation recovery application of the oversized article, and sending the reservation recovery application to a cloud data management center, wherein the reservation recovery application comprises the type, weight, volume, reservation recovery position and reservation recovery time of the oversized article.

The mobile application 102 is a key interface to a mutually-assisted shared multi-purpose garbage can processing system, providing a platform for users to interact with the system. The application program aims to improve the enthusiasm of users for participating in garbage classification, and provides convenient reservation service for recycling large garbage.

The application program realizes multiple functions through high-efficiency interaction with the cloud data management center. Firstly, the method can show the nearest intelligent garbage can position to the user, and the user can conveniently find the nearest putting point by utilizing the geographic position information of the user and positioning and displaying the nearby intelligent garbage can through a map service API (application program interface).

Secondly, the application program provides an interface which allows the user to record and check the type, weight, volume information and throwing time of the garbage thrown by the user. The function is realized by reading data sent to the cloud data management center by the intelligent garbage can, and when a user puts garbage, relevant information identified by the intelligent garbage can is synchronized to an application interface of the user through the cloud.

In addition, the application program is provided with a function of reserving recycling service, and a user can submit a reservation recycling application of the oversized article through the function. The user needs to fill in the relevant information of reservation recovery in the application, including the type, weight, volume, reservation recovery position and reservation recovery time of the article. The information is sent by the application to a cloud data management center, which will schedule the transportation vehicle for recycling based on the information. The reserved recycling location may be near the intelligent dustbin nearest to the user.

The application is also designed with a User Interface (UI) that provides an intuitive and easy-to-use operating experience. The user can easily browse and operate, such as searching the garbage can, checking the classification guidance, submitting the reservation recovery application and the like. In order to realize the functions, the back-end development of the application program needs to adopt a stable server technology, and the front-end development needs to pay attention to user interaction design, so that the response speed of the application and the real-time performance of data updating are ensured.

In terms of technical implementation, mobile application programs can be developed by adopting a cross-platform framework, so that different operating systems such as iOS and Android can be ensured to be covered. By using modern application development frameworks such as REACT NATIVE and Flutter, development efficiency can be effectively improved, and meanwhile, performance and user experience of applications are guaranteed.

In summary, the mobile application 102 provides a convenient and intuitive platform for users to interact with the mutually-assisted shared multi-purpose garbage can processing system through the smart phone, including functions of searching garbage cans, recording garbage placement information, reserving large garbage recycling and the like.

Further, the mobile application program allows a user to submit a reservation recycling application of the oversized article by filling in an electronic form or uploading a photo, and receives recycling confirmation and reservation time schedule of the cloud data management center in real time.

In the mutual-assistance sharing multi-purpose garbage can processing system provided by the embodiment, the mobile application plays a vital role, and not only serves as an interface for a user to interact with the system, but also provides an efficient and user-friendly way for processing reservation recycling of oversized articles.

The mobile application program needs to design an intuitive and easy-to-use user interface through which a user can easily submit reservation recovery applications for oversized items. This process can be accomplished in two main ways: fill in the electronic form and upload the photo.

1. Filling in an electronic form: the application program needs to provide a spreadsheet for the user to enter detailed information about the oversized item, including but not limited to the type, size, weight, and desired recovery time and location of the item. This form should be designed as compact as possible to facilitate user filling.

2. Uploading a photo: in view of the fact that some users may have difficulty accurately describing their oversized items, the application should also allow the user to directly upload photographs of the items. The application may further instruct the user how to take the photos (e.g., to ensure that the size and texture of the item are clearly visible) so that the cloud data management center can more accurately evaluate and process the applications.

Once the user submits the reservation recovery application, the mobile application needs to send the information to the cloud data management center in real time. The cloud data management center is responsible for processing these applications, including assessing the recycling feasibility of the items, scheduling recycling times, recycling vehicles, and the like. After these processes are completed, the cloud data management center sends the recovery confirmation information and the specific reservation schedule back to the mobile application.

The mobile application should be able to receive feedback from the cloud data management center and notify the user in real time. This includes retrieving the acknowledgment information and specific reservation schedules. The application program can inform the user through push notifications, ensuring that the user can obtain the important information in time.

To ensure that the user can easily receive and view this information, the application should also provide a special interface or messaging center in which all reservation recovery applications and their status, including confirmed, pending, completed, etc. are listed.

Still further, the mobile application actively pushes nearby upcoming garbage collection activity information or alerts the user to participate in periodic reclamation activities according to the user's location and preference settings.

In a mutual-sharing multi-purpose garbage can processing system, the design of mobile application programs comprises an important function: based on the geographic location and personal preference settings of the user, relevant garbage collection activity information is actively pushed to the user or the user is reminded to participate in regular recycling activities. The implementation of this function involves location services, storage and processing of user preference settings, information push mechanisms, and real-time data exchange with a cloud data management center. The following details how this function is implemented.

First, the mobile application needs to obtain the user's permissions to access the geographic location information of its device. This is typically done by a pop-up request when the user first starts the application. Once licensed, the application may run in the background, updating the user's current location periodically or as needed.

Second, the application allows the user to specify personal preferences in the settings menu, including the type of garbage collection of interest (e.g., recyclables, organic waste, etc.), the time preference to receive notification (e.g., weekdays, weekends, or any particular period of time), the maximum distance to which to go, etc. These preference settings are stored on the local device and can be updated at any time.

Next, the application needs to communicate with the cloud data management center to obtain, in real-time, the information of the upcoming garbage collection activities, including the type of activity, time, place, and any activity-specific requirements or indications. The cloud data management center is responsible for processing data from the intelligent garbage bin, and organizing and scheduling garbage collection activities.

Once the application receives the garbage collection activity information that matches the user's location and meets the user's preferences, it sends a reminder to the user via an in-application notification or Short Message Service (SMS). These notifications contain detailed information about the activity, such as time, place and manner of participation, and any necessary preparation.

To increase user engagement, the application may also provide an interactive map showing the user's current location, nearby smart trash cans, and the location of the collection activity that is to be performed. The user may explore different activities through the map and choose to participate.

In addition, the application may also recommend specific garbage collection activities to the user using algorithms based on the user's engagement history and preferences, further personalizing the user experience.

Finally, to ensure that users can conveniently engage in such activities, the application should provide a simple feedback mechanism that allows users to share their experience after engaging in the activity, including submitting problem reports or advice, thereby helping to improve future garbage collection activities.

Through implementation of the method, the mobile application program not only can enhance user experience and promote communities to participate in environmental protection activities, but also can provide precious user feedback and data for the mutual-assistance sharing multi-purpose garbage can processing system so as to optimize the overall performance and efficiency of the system.

The cloud data management center 103 is used for receiving classification information, weight, volume and throwing time of throwing garbage from the intelligent garbage can; counting the classification information, weight, volume and throwing time of the throwing garbage to obtain garbage storage information of the intelligent garbage can; receiving a reservation recovery application sent by a mobile application program; determining a dispatching instruction of the transport vehicle according to the garbage storage information and the reservation recovery application, wherein the dispatching instruction comprises the optimal route and time of the transport vehicle; and sending the scheduling instruction to the transport vehicle.

The cloud data management center 103 plays a key role in data processing and scheduling in the mutual sharing multi-purpose garbage can processing system. The center adopts advanced cloud computing technology, and can receive and process a large amount of data from the intelligent garbage can 101 in real time, including classification information, weight, volume and throwing time of garbage. In addition, it processes user data and requests from mobile application 102, such as reservation recovery applications for oversized items.

In order to realize the functions, the cloud data management center constructs an efficient data processing platform, and the platform comprises a data receiving module, a data processing and analyzing module, a scheduling decision module and a communication module. The data receiving module is responsible for receiving data from the intelligent garbage can and the mobile application, and ensures real-time transmission and accuracy of the data. And the data processing and analyzing module is used for carrying out statistics and analysis on the received data, generating garbage storage information of the intelligent garbage can and processing reservation recovery application of a user.

The scheduling decision module is the core of the center and calculates the optimal route of the transport vehicle by adopting an algorithm according to the garbage storage information of the intelligent garbage can and the reservation recovery requirement of the user. This process involves complex logic judgment and optimization algorithms to ensure the efficiency and economy of garbage collection and reclamation work. The communication module is responsible for sending scheduling instructions to the transport vehicle 104 while ensuring smooth communication between the various parts within the system.

The cloud data management center needs to be realized by depending on a stable and reliable cloud service platform, such as China cloud, hundred-degree cloud and the like, so that the high efficiency of data processing and the stable operation of a system are ensured. In addition, the adoption of the micro-service architecture can enhance the flexibility and the expandability of the system, so that each module can independently run and update, thereby adapting to the continuously changing requirements and technical progress.

The cloud data management center 103 serves as a brain of the system, can process data from the intelligent garbage can, optimize garbage collection and recovery processes, can improve user participation, improves efficiency and response speed of the whole system through intelligent scheduling, and contributes important strength to urban garbage management.

Further, the cloud data management center adopts an optimization algorithm to determine an optimal path of the transport vehicle; the optimization algorithm is shown in the following equation 1:

Wherein S _opt represents an optimal score representing the selected path; n represents the total number of considered paths; f _i is the scoring function for the ith path; d _i denotes the total distance of the ith path; v _i denotes the average vehicle speed of the i-th path; t _i denotes the estimated travel time of the i-th path; w _i denotes the congestion index of the i-th path; c _i represents the carbon emission cost of the ith path; e _i represents the urgency score for the ith path; omega ₁、ω₂、ω₃、ω₄ and omega ₅ are weight coefficients; α ₁、α₂、α₃、α₄ and α ₅ are tuning parameters; τ is a time sensitivity threshold for adjusting the weight of the urgent task.

The optimization algorithm is designed to determine an optimal path for the transportation vehicle to increase the efficiency and response speed of the mutually-shared multi-purpose garbage can processing system while taking into account environmental protection requirements. The following are detailed explanations and calculation methods for each term in the formula:

The total distance D _i represents the total travel distance from the start point to the end point on the ith path. In path planning, distance is a key factor affecting fuel consumption and travel time. The method is obtained by acquiring the shortest distance between two points through a map service API or combining and calculating GPS data and map data.

The average vehicle speed V _i represents the average running speed on the i-th path. Speed affects travel time and energy consumption, high speed may mean shorter travel time, but may also result in higher energy consumption and risk. It can be estimated from historical data or real-time traffic information, and can also calculate an average value through real-time data of vehicle running.

The estimated travel time T _i represents the total time required to estimate the completion of the path. Time is one of the direct indicators of path efficiency. The estimated travel time T _i may be estimated in consideration of real-time traffic conditions and road condition changes in combination with the total distance and the average vehicle speed.

The congestion index W _i reflects an index of the degree of congestion on the path. A high congestion index may mean longer travel time and higher uncertainty. The congestion index W _i may be derived by analysis from real-time data provided by the traffic management system or based on historical traffic flow data.

The carbon emission cost C _i represents the carbon emission cost generated during the course of the path traveling. Environmental impact factors are considered to encourage the selection of low carbon emission routes. The carbon emission cost C _i is calculated based on the fuel efficiency of the vehicle, the travel distance, and the carbon emission factor of different fuel types.

The urgency score E _i reflects the score of task urgency. Emergency tasks may require priority even though path costs are somewhat higher. The urgency score E _i is comprehensively assessed according to factors such as the type of task, the starting time, the expected completion time and the like.

The weight coefficients (omega ₁、ω₂、ω₃、ω₄ and omega ₅) and the adjustment parameters (alpha ₁、α₂、α₃、α₄、α₅) determine the relative importance and influence mode of various factors in the total score, so that the algorithm can adjust the importance degree of different factors according to actual conditions. These weight coefficients and tuning parameters may be dynamically tuned by data analysis, expert system advice, or training results of a machine learning model.

The time sensitivity threshold τ is used to adjust the threshold of urgent task weights, and when the urgency of a task exceeds this threshold, its priority in path selection is increased. The time sensitivity threshold τ may be obtained through expert knowledge or set based on analysis of business needs and historical task urgency.

In a mutually-assisted shared multi-purpose garbage can processing system, the function of a cloud data management center is important, and particularly in the task of optimizing a transport vehicle path. In order to ensure that the transportation vehicle can finish garbage collection and reservation recycling tasks of oversized articles in the most efficient and most environment-friendly mode, a cloud data management center adopts a refined optimization algorithm. The optimization algorithm not only considers a plurality of physical and environmental factors of the path, but also introduces special consideration to emergency tasks, and ensures that the system can flexibly respond to various conditions.

The core of the optimization algorithm is by calculating the composite score for each possible path, and then selecting the path with the lowest score (i.e., lowest cost) as the optimal path. This score includes not only the total distance of the path and the estimated travel time, but also integrates vehicle speed, congestion conditions, carbon emission costs, and mission urgency. The factors are weighted by specific weighting coefficients to reflect their relative importance in path selection.

Specifically, each component in the algorithm has its unique role and computational manner:

The total distance D _i and the average vehicle speed V _i directly affect the cost of transportation and time, weighted by a specific power function to adjust the impact on the total score.

The estimated travel time T _i and the congestion index W _i together reflect the actual traffic efficiency of the route. By multiplying the power function of the estimated travel time by the logarithm of the congestion index, the algorithm can more accurately evaluate the time cost.

The consideration of the carbon emission cost C _i represents an environmental concern of the system. By introducing in the form of an exponential function, it is ensured that a greater penalty is imposed on the path of higher carbon emission costs.

The urgency score E _i is introduced by an expression containing a logistic function, so that the weight of the tasks with different urgency degrees in the total score can be dynamically adjusted, and the quick response capability of the system to the urgent tasks is reflected.

Implementation of the algorithm requires configuration of high-performance computing resources at the cloud data management center to handle large amounts of real-time data, including traffic flow, weather conditions, vehicle status, user requests, and the like. Furthermore, the determination of the weight coefficients (ω ₁、ω₂、ω₃、ω₄、ω₅) and the power exponent (α ₁、α₂、α₃、α₄、α₅) should be based on an analysis of a large amount of historical data, as well as the advice of the expert system.

By the method, the cloud data management center can ensure that the transportation vehicle meets the service requirement, minimize the transportation cost and the environmental influence, and improve the overall efficiency and the sustainability of the system.

And the cloud data management center determines a dispatching instruction of the transport vehicle according to the garbage storage information and the reservation recovery application. First, a possible path is determined based on the garbage storage information of the intelligent garbage can and the reservation recycling application of the mobile application. The following are detailed steps of how the possible paths are determined from this information:

1. Collecting and analyzing data

Intelligent garbage can data: the system periodically collects the type, weight and volume information of the garbage from each intelligent garbage bin. These data reflect the fill status of each garbage can, helping to predict when garbage collection is needed.

Reservation recovery application: the system collects reservation recovery applications submitted by the user through the mobile application, including the type, weight, volume, reservation recovery location, and reservation time of the item.

2. Evaluating trash can status and reservation recycling requirements

Based on the collected data, the system evaluates the emergency emptying requirements of each intelligent dustbin and the urgency of the scheduled recycling tasks. This step involves comparing the real-time fill data of the trash can to a capacity threshold and evaluating the time requirements of the scheduled reclamation task.

3. Generating candidate paths

The system generates a series of candidate paths based on the location of the intelligent garbage can and the location of the reservation recovery task. Each candidate route represents one possible route for the transportation vehicle, including a route from the current location, through the intelligent trash can and the reservation recycling location in turn, to the base or processing center.

Through the steps, the mutual-assistance sharing multi-purpose garbage can processing system can determine the candidate path of the transport vehicle according to the garbage storage information of the intelligent garbage can and the reservation recovery requirement of the user.

For each candidate path determined, the total distance D _i, average vehicle speed V _i, estimated travel time T _i, congestion index W _i, and carbon emission cost C _i for each path are evaluated using the optimization algorithm described above, and the mission urgency score E _i is considered. Each factor is weighted by a specific weighting coefficient ω ₁、ω₂、ω₃、ω₄ and ω ₅ to reflect their relative importance in path selection.

By comparing the composite scores of all possible paths, the path with the lowest score is selected as the optimal path. The process involves comprehensively considering a plurality of factors such as distance, speed, time, congestion condition, environmental impact and the like, ensuring that the selected path can minimize environmental impact and transportation cost while ensuring task completion efficiency.

Once the optimal path is determined, the cloud data management center generates corresponding scheduling instructions including travel routes, expected departure and arrival times, etc. of the transport vehicles, and then transmits the instructions to the corresponding transport vehicles.

The cloud data management center can dynamically adjust the existing scheduling instructions according to real-time traffic information, emergencies and other factors so as to ensure that transportation tasks can be efficiently completed under various conditions.

Furthermore, the cloud data management center performs data analysis on the received garbage classification information, weight, volume and throwing time so as to predict the garbage generation trend and the emptying period of the intelligent garbage can, thereby optimizing the whole garbage collection and treatment flow.

In the mutual sharing multi-purpose garbage can processing system, the cloud data management center plays a core role, in particular to the aspects of processing and analyzing garbage classification information, weight, volume and throwing time. These data not only provide immediate garbage status information, but also enable deep analysis to reveal trends in garbage generation and the emptying cycle of the intelligent garbage can. This feature is described in detail below.

Firstly, the cloud data management center receives data from each intelligent garbage can in real time through a safe network connection. Such data includes, but is not limited to, the type, weight, volume, and exact time of each trash placement. In order to accurately capture and store such information, centers need to deploy efficient database systems that can handle large amounts of data input and support fast query and analysis operations.

Once the data is collected, the cloud data management center will conduct in-depth analysis of the information using data analysis techniques and machine learning algorithms. In particular, time series analysis techniques may be applied to identify patterns and periodic changes in garbage generation, such as increases in garbage generation during certain time periods (e.g., holidays or special events). In addition, by analyzing the weight and volume data of different types of garbage, the center can predict the generation trend of each type of garbage and the time for the intelligent garbage can to reach the full-load state.

With these analysis results, the cloud data management center can provide valuable insight into garbage collection and processing flows. For example, by predicting trends in trash production and emptying cycles of intelligent trash cans, a center may schedule trash collection tasks in advance to avoid premature trash can loading while optimizing collection routes to reduce transportation costs and time. In addition, this information can also be used to guide resource allocation decisions such as dispatching collection vehicles and personnel in areas where increased amounts of waste are expected.

To accomplish this, cloud data management centers need to develop and deploy complex data processing and analysis systems. This may include a data cleansing and preprocessing module for preparing the data for analysis; the data analysis and model training module is used for constructing and optimizing a prediction model; and the decision support module is used for generating specific operation suggestions and scheduling instructions based on the analysis result.

In addition, in order to continuously improve the accuracy of prediction and the efficiency of the system, the cloud data management center also needs to periodically review the comparison between the analysis result and the actual situation, and adjust and optimize the data model and the analysis flow.

By the method, the cloud data management center can effectively utilize the received garbage classification information, weight, volume and throwing time, scientific decision support is provided for the whole garbage collection and processing flow, optimal allocation of resources is realized, service efficiency is improved, and meanwhile, contribution is made to environmental protection.

And the transport vehicle 104 is used for receiving a scheduling instruction of the cloud data management center, and carrying out garbage collection of the intelligent garbage can and reservation recovery of oversized articles according to the scheduling instruction.

The transport vehicle 104 plays a vital role in the present mutual-aid shared multi-purpose garbage can processing system, and is not only responsible for garbage collection work according to the information of the intelligent garbage can, but also processes the reservation recycling task of oversized articles from users.

The transport vehicle 104 is designed with the requirements of efficiency and flexibility in mind. The vehicle is equipped with an advanced positioning system, such as a GPS, so as to ensure that navigation information issued by a cloud data management center can be accurately received, and garbage collection and reservation recovery of oversized articles are performed according to an optimal route. In addition, the vehicle is equipped with various sensors including a distance sensor, a speed sensor, etc., which help the vehicle maintain a safe distance while performing a mission, avoid a collision accident, and also ensure optimization of a driving route.

The internal mechanical structure of the vehicle is designed to accommodate the collection and sorting of different types of refuse. For example, for normal household garbage, the transport vehicle may be equipped with an automated loading system, while for large items reserved for recycling, a more flexible manual loading may be required. Therefore, the space inside the vehicle is reasonably planned to adapt to garbage articles with different sizes and shapes, and meanwhile, the safety and the efficiency of the transportation process are ensured.

Interaction with the cloud data management center is critical to the implementation of transport vehicle 104 functions. The vehicle receives scheduling instructions from the cloud in real time through a built-in communication module, such as 4G or 5G network equipment, wherein the scheduling instructions comprise target positions for garbage collection, specific information for reservation recovery and recommended driving routes. In the task execution process, the vehicle can feed back state information to the cloud end, including task completion conditions, current positions, residual capacity and the like, so that the system can update the working state of the vehicle in real time, and a subsequent scheduling plan is optimized.

The hardware and software systems of the transport vehicle need to be carefully designed and configured. The software system comprises an operating system of the vehicle-mounted computer, navigation software, data transmission and processing application and the like, and the software ensures that the vehicle can accurately understand and execute instructions from the cloud. The hardware system comprises a power system, a sensor, communication equipment and the like of the vehicle, and the hardware components need to select products with high performance and high reliability so as to adapt to various weather and road conditions and ensure the successful completion of tasks.

Still further, the transport vehicle is equipped with GPS positioning and mobile communication technology to report vehicle location, status and garbage collection progress to a cloud data management center in real time to facilitate real-time monitoring and adjustment of the transport plan.

In a mutual-sharing multi-purpose garbage can handling system, efficient and on-time operation of the transport vehicle is critical to maintaining the smoothness and reliability of the overall system. To achieve this goal, transport vehicles are designed to be equipped with advanced GPS positioning and mobile communication technologies. The combination of the technologies not only enables the cloud data management center to receive the accurate position and the running state of each vehicle in real time, but also allows the center to dynamically adjust the transportation plan according to the collected data, and ensures the high efficiency and timeliness of the garbage collection process.

The implementation details include:

GPS positioning technology: each transport vehicle is equipped with a high-precision GPS receiver that enables the geographic location of the vehicle to be determined in real-time. This location information is automatically sent to the cloud data management center at certain time intervals (e.g., once per minute) via the mobile communication device on board the vehicle. In this way, the center can track the position of each vehicle in real time, ensuring that the vehicle is traveling as planned.

2. Mobile communication technology: the transport vehicle is also equipped with a mobile communication module, such as a 4G LTE or 5G communication device, for transmitting the vehicle's operational data (including location, speed, and status) and garbage collection progress information to the cloud data management center in real time. This real-time data transmission ensures that the center can quickly respond to any emergency, such as traffic congestion or vehicle failure, and adjust the transportation route or plan in time.

3. Data analysis and transportation plan adjustment: and after receiving the real-time data from the transport vehicle, the cloud data management center inputs the information into an advanced data analysis system. The system uses algorithms to analyze vehicle operating conditions, garbage collection progress, and other relevant factors, such as traffic conditions and weather forecast, to predict possible delays or early completion. Based on these analysis results, the center may optimize the transportation plan, e.g., reschedule the transportation route, adjust the collection time, or redistribute tasks to other vehicles.

In order to achieve the above functions, it is necessary to install corresponding hardware devices in the transport vehicle and develop corresponding software programs. The hardware installation includes the installation and configuration of the GPS receiver and the mobile communication module. Software development then includes programming to automatically collect and send vehicle data, receive instructions from the center, and update the vehicle's transportation plan as necessary. In addition, the cloud data management center needs to develop a corresponding back-end system for receiving and processing vehicle data, and a dynamic adjustment algorithm for the transportation plan.

Through the implementation of the technology, the mutual-assistance sharing multi-purpose garbage can processing system can ensure that the garbage collection and recovery tasks of the transport vehicle can be completed efficiently and on time, and the overall operation efficiency and reliability of the system are greatly improved.

While the application has been described in terms of preferred embodiments, it is not intended to be limiting, but rather, it will be apparent to those skilled in the art that various changes and modifications can be made herein without departing from the spirit and scope of the application as defined by the appended claims.

Claims (9)

1. A mutually-assisted shared multi-purpose garbage can processing system, comprising:

The intelligent garbage can is provided with a unified throwing port and is used for processing the thrown garbage by using an image processing technology, a microwave technology, a weight sensor and a volume scanner to obtain classification information, weight, volume and throwing time of the thrown garbage; according to the classification information, the input garbage is shunted into corresponding containers preset in the intelligent garbage can through an internal mechanical structure of the intelligent garbage can; the classification information, weight, volume and throwing time of the throwing garbage are sent to a cloud data management center;

The mobile application program is interacted with the cloud data management center and is used for displaying the position of the nearest intelligent garbage can to the user and recording the type, weight and volume information and the throwing time of garbage thrown by the user; allowing a user to conduct reservation recovery application of the oversized article, and sending the reservation recovery application to a cloud data management center, wherein the reservation recovery application comprises the type, weight, volume, reservation recovery position and reservation recovery time of the oversized article;

The cloud data management center is used for receiving classification information, weight, volume and throwing time of throwing garbage from the intelligent garbage bin; counting the classification information, weight, volume and throwing time of the throwing garbage to obtain garbage storage information of the intelligent garbage can; receiving a reservation recovery application sent by a mobile application program; determining a dispatching instruction of the transport vehicle according to the garbage storage information and the reservation recovery application, wherein the dispatching instruction comprises an optimal route of the transport vehicle; transmitting the scheduling instruction to a transport vehicle;

and the transport vehicle is used for receiving a scheduling instruction of the cloud data management center, and carrying out garbage collection of the intelligent garbage can and reservation recovery of oversized articles according to the scheduling instruction.

2. The mutual-aid sharing multi-purpose garbage can processing system according to claim 1, wherein the intelligent garbage can adopts a trained hybrid neural network to analyze data collected by an image recognition technology, a microwave technology, a weight sensor and a volume scanner to obtain classification information of input garbage; the hybrid neural network comprises a multi-scale image recognition branch, a microwave-material sensing branch, a weight-volume characteristic coding branch, a time sequence-putting mode branch and a multi-mode data fusion and decision layer;

The multi-scale image recognition branch adopts a multi-scale convolutional neural network architecture and is used for processing image data of garbage to obtain an image feature vector; the microwave-material sensing branch adopts a framework based on a graph attention network, the acquired microwave signal data are regarded as nodes in the graph, and the relationship among the nodes is learned by using an attention mechanism to obtain a microwave reflection characteristic vector; the weight-volume characteristic coding branch uses a self-encoder network to perform characteristic extraction on collected garbage weight and volume data to obtain a physical attribute characteristic vector; the time sequence-putting mode branch adopts a time sequence processing model based on a transformer model, and the acquired putting time data is processed to obtain a putting mode feature vector; the multi-mode data fusion and decision layer processes the image feature vector, the microwave reflection feature vector, the physical attribute feature vector and the throwing mode feature vector by using an attention mechanism and a full connection layer to obtain the classification information of garbage.

3. The mutually-assisted shared multi-purpose garbage can processing system of claim 2, wherein the training step of the hybrid neural network is performed in a staged manner, comprising:

in an independent training stage, respectively and independently training a multi-scale image recognition branch, a microwave-material sensing branch, a weight-volume characteristic coding branch and a time sequence-put mode branch, so as to ensure that each branch can effectively process the corresponding data type and extract the related characteristics;

In the end-to-end training phase, the whole hybrid neural network performs end-to-end training as a whole.

4. A mutually-assisted shared multi-purpose garbage can processing system according to claim 3, wherein the multi-scale image recognition branches employ cross entropy loss functions as loss functions during a separate training phase; the microwave-material perception branch adopts a cross entropy loss function as a loss function; the weight-volume characteristic coding branch adopts a mean square error loss function as a loss function; the time series-put mode branch employs a cross entropy loss function as the loss function.

5. The system of claim 1, wherein the mobile application allows a user to submit a reservation recycling application for the oversized item by filling in a spreadsheet or uploading a photo, and to receive a recycling confirmation and reservation schedule from the cloud data management center in real time.

6. The mutually-assisted shared multi-purpose garbage can processing system of claim 1, wherein the mobile application actively pushes nearby upcoming garbage collection activity information or alerts users to participate in periodic reclamation activities according to user location and preference settings.

7. The mutually-assisted shared multi-purpose garbage can processing system according to claim 1, wherein the cloud data management center performs data analysis on the received garbage classification information, weight, volume and delivery time to predict a garbage generation trend and a garbage can emptying cycle, thereby optimizing the overall garbage collection and processing flow.

8. The mutually-assisted shared multi-purpose garbage can processing system according to claim 1, wherein the cloud data management center adopts an optimization algorithm to determine an optimal path of a transport vehicle; the optimization algorithm is shown in the following equation 1:

Wherein S _opt represents an optimal score representing the selected path; n represents the total number of considered paths; d _i denotes the total distance of the ith path; v _i denotes the average vehicle speed of the i-th path; t _i denotes the estimated travel time of the i-th path; w _i denotes the congestion index of the i-th path; c _i represents the carbon emission cost of the ith path; e _i represents the urgency score for the ith path; omega ₁、ω₂、ω₃、ω₄ and omega ₅ are weight coefficients; α ₁、α₂、α₃、α₄ and α ₅ are tuning parameters; τ is a time sensitivity threshold for adjusting the weight of the urgent task.

9. The mutually-sharing multipurpose garbage can processing system of claim 1, wherein the transport vehicle is equipped with GPS positioning and mobile communication technology, reporting vehicle location, status and garbage collection progress to a cloud data management center in real time for facilitating real-time monitoring and adjustment of transport plans.