Multi-mode interactive garbage classification guiding and recycling system

Abstract

The invention discloses a multi-mode interactive garbage classification guiding and recycling system, particularly relates to the field of garbage classification and recycling, is used for solving the problem that equipment capacity planning is not matched with space-time distribution of user delivery behaviors, and aims to provide a multi-mode interactive garbage classification guiding and recycling system through space-time joint modeling, combination of a user moving track and equipment load data and capture of behavior migration characteristics. Global collaboration of equipment full-load prediction and path optimization is realized; based on multi-modal interaction data, a user position and a garbage category are perceived in real time, heterogeneous indexes and path topology constraints are fused, and a shunting decision giving consideration to equipment capacity, regional load balance and user path preference is dynamically generated; meanwhile, load prediction and path screening rules are continuously optimized through user behavior feedback, and a closed-loop link of behavior recognition, load prediction, a shunting strategy and system iteration is formed; classification guiding accuracy, equipment load balance and user delivery convenience are improved, equipment overload and resource waste are effectively relieved, and the operation efficiency of urban garbage classification recycling is improved.

Description

Technical Field

[0001] The present invention relates to the field of garbage classification and recycling, and more specifically, to a multi-mode interactive garbage classification guidance and recycling system. Background Art

[0002] At present, smart urban waste sorting recycling bins are mostly concentrated in fixed scenes such as office areas and residential areas, but there is a deep gap between equipment capacity planning and the spatiotemporal distribution of user delivery behavior. Taking a typical office area as an example, a large number of users discard disposable lunch boxes during the lunch period, causing the recycling bins to be rapidly filled, while similar equipment in adjacent commercial areas is in a low-load state for a long time due to sparse traffic; the peak of kitchen waste delivery in residential areas at night often causes equipment overflow, but the equipment is idle before the next morning's transportation, causing resources to run idle. Although the existing system can predict the capacity of a single device based on historical data, it cannot capture the migration characteristics of user behavior in the spatiotemporal dimension, such as the dynamic offset of commuting routes and temporary delivery tides caused by regional activities, resulting in inaccurate equipment load prediction and failure of guidance strategies.

[0003] The essence of the problem is that the periodic volatility of user delivery behavior in time and the path dependence in space form a high-dimensional nonlinear coupling relationship with the rate of equipment capacity consumption. The traditional static scheduling model cannot simultaneously interpret two types of contradictory characteristics: on the one hand, the strong spatiotemporal inertia of user delivery paths causes local sudden accumulation of equipment loads; on the other hand, global load balancing requires dynamic diversion across regions, but is limited by the natural resistance of users to path detours. The system needs to build a spatiotemporal joint optimization model under the constraints of user behavior. The deeper contradiction lies in the modeling conflict between the time period regularity in historical data and the spatial load mutation triggered by real-time events, which makes it difficult to coordinate the prediction of equipment full load time and diversion path planning, and ultimately forms a negative cycle in which user behavior exacerbates equipment overload and overloaded equipment reversely distorts user behavior.

[0004] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a multi-modal interactive garbage classification guidance and recycling system. Aiming at the contradiction between the spatiotemporal distribution of user delivery behavior and the imbalance of equipment load, a spatiotemporal coupled closed-loop guidance mechanism is constructed to optimize garbage classification paths and resource scheduling. Based on the spatiotemporal modeling of user trajectories and equipment loads, the capacity consumption characteristics of scenes such as peak hours in office areas and tides in residential areas are quantified; multimodal data is used to perceive location and garbage categories in real time, and diversion decisions that take into account capacity, load differences and user preferences are generated; the model is continuously optimized through behavioral feedback to form a behavior recognition-load prediction-diversion generation-adaptive iteration closed loop. This method breaks through the limitations of traditional static scheduling, reduces the risk of overload and overflow while avoiding detours, reduces resource vacancy, and achieves multi-objective optimization of guidance accuracy, load balancing and delivery convenience to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: Multi-mode interactive garbage sorting guidance and recycling system, including: Trajectory clustering module, load prediction module, path optimization module and feedback adjustment module; The trajectory clustering module identifies high-frequency delivery paths through spatiotemporal clustering of user movement trajectories, and extracts dynamic prediction parameters of equipment full load time by combining spatiotemporal correlation modeling of equipment load data; The load prediction module quantifies the maximum detour range based on the user's current location and path preference, uses the spatiotemporal prediction results of the equipment capacity consumption rate, and dynamically calculates the equipment load urgency and path selection resistance to generate an indicator reflecting the matching degree of load and path, and dynamically selects a set of candidate devices that meet the remaining capacity requirements; The route optimization module uses the user's detour distance as a hard constraint, the remaining capacity of the equipment and the regional load difference as the optimization target, and solves the optimal delivery route through a multi-objective decision-making algorithm; The feedback adjustment module adjusts the prediction parameters of the equipment load in the spatiotemporal correlation model based on user behavior feedback data, and simultaneously optimizes the path screening threshold to achieve closed-loop optimization of prediction and scheduling.

[0007] In a preferred embodiment, the trajectory clustering module includes the following contents: High-frequency delivery paths are identified through spatiotemporal clustering of legitimate mobile trajectories actively provided by users, specifically including collecting trajectory point sets uploaded by users, clustering the trajectory point sets in the spatiotemporal dimension using a density-based spatiotemporal clustering algorithm, generating multiple spatiotemporal clusters, and screening out high-frequency delivery paths by counting the frequency of occurrence of trajectory points in each spatiotemporal cluster within a specific time period; at the same time, the spatiotemporal correlation modeling of equipment load data is combined, specifically including constructing a vector autoregression model to analyze the dynamic changes of equipment load in time and space, calculate the equipment load change rate, and then extract the dynamic prediction parameters of the equipment full load time.

[0008] In a preferred embodiment, the dynamic prediction parameters include a capacity consumption rate and a predicted full load time, and the predicted full load time is calculated by a ratio of the remaining capacity to the capacity consumption rate.

[0009] In a preferred embodiment, the load prediction module includes the following contents: The user's current location is obtained in real time, and the preference feature vector is extracted by analyzing the user's historical delivery behavior data. The maximum detour distance threshold acceptable to the user is calculated to determine the detour range; then the dynamic prediction parameters of the device are obtained from the trajectory clustering module, and the coupling coefficient is calculated using the device capacity consumption rate; finally, a set of candidate devices is dynamically generated through multiple screening of the user's maximum detour range, the device's remaining capacity greater than the remaining capacity threshold, and the coupling coefficient greater than or equal to the preset coupling coefficient threshold.

[0010] In a preferred embodiment, the coupling coefficient is obtained by adjusting the ratio of the spatiotemporal inertia attenuation index to the path topology resistance index and the regional equipment load difference.

[0011] In a preferred embodiment, the spatiotemporal inertia decay index is calculated by weighting the instantaneous change gradient of the equipment capacity consumption rate and the information entropy of the user's historical delivery time distribution by the inverse of the regional traffic density.

[0012] In a preferred embodiment, the path topology resistance index is obtained by multiplying the path network topology depth difference from the user's current location to the device location by the attenuation memory factor of the user's historical detour behavior.

[0013] In a preferred embodiment, the path optimization module includes the following contents: Taking the user detour distance as the hard constraint condition, and the device remaining capacity and regional load difference as the optimization objectives, the geographical distance from the user's current location to the candidate device location is first calculated to verify whether the hard constraint of the user's detour distance is met; then the device remaining capacity is defined as the full load capacity of the device minus the current load, and the regional load difference is defined as the sum of the deviation indexes of all device loads and the average load; then the improved non-dominated sorting method is used to construct the initial solution set, and the device remaining capacity and the regional load difference after delivery are evaluated for each candidate device one by one, and the Pareto frontier solution set is generated through non-dominated sorting; then the target preference factor is introduced to calculate the comprehensive score of the solutions in the Pareto frontier solution set, and the device with the highest comprehensive score is selected as the optimal device; finally, the Dijkstra algorithm is used to generate the shortest path from the user's current location to the optimal device location, thereby solving the optimal delivery path.

[0014] In a preferred embodiment, the feedback adjustment module includes the following contents: Firstly, multiple feedback data of users' actual delivery are collected and compared with the recommendation results generated by the path optimization module, and the equipment selection deviation, delivery time deviation and delivery volume deviation are analyzed. Then, by comparing the difference between the actual capacity consumption rate and the predicted capacity consumption rate, the capacity consumption rate of the equipment is updated by a nonlinear fusion method, and the full load time of the equipment is recalculated according to the updated capacity consumption rate. At the same time, according to the actual detour distance of the user, the remaining capacity of the equipment after delivery and the coupling coefficient of the actual path, the maximum detour distance threshold, the remaining capacity threshold and the coupling coefficient threshold are adjusted respectively. Finally, the updated prediction parameters and path screening threshold are fed back to the trajectory clustering module and the load prediction module.

[0015] Technical effects and advantages of the multi-mode interactive garbage classification guidance and recycling system of the present invention: By building a closed-loop optimized multi-modal interactive garbage sorting guidance and recycling system, we can effectively cope with the challenge of mismatch between equipment capacity planning and the spatiotemporal distribution of user delivery behavior in urban intelligent garbage sorting and recycling; relying on the spatiotemporal joint modeling of user movement trajectory and equipment load data, we can accurately capture the behavior migration characteristics and realize the global coordination of equipment full load prediction and path optimization; with the help of multimodal interactive data, we can perceive the user location and garbage category in real time, integrate heterogeneous indexes and path topology constraints for dynamic calculation, and generate diversion decisions that take into account equipment capacity, regional load balancing and user path preferences; at the same time, we can continuously optimize load prediction and path screening rules through user behavior feedback, forming a closed-loop link of behavior recognition, load prediction, diversion strategy and system iteration, thereby significantly improving the accuracy of classification guidance, equipment load balancing and user delivery convenience, effectively alleviating the problems of equipment overload and resource waste, and comprehensively improving the operational efficiency and user experience of urban garbage sorting and recycling. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a structural schematic diagram of the multi-mode interactive garbage classification guidance and recycling system of the present invention.

[0017] Figure 2 It is a schematic diagram of the sub-step flow of the path optimization module of the multi-mode interactive garbage sorting guidance and recycling system of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Embodiment 1: Figure 1 The invention provides a multi-mode interactive garbage classification guidance and recycling system, comprising: a trajectory clustering module, a load prediction module, a path optimization module and a feedback adjustment module.

[0020] The trajectory clustering module identifies high-frequency delivery paths through spatiotemporal clustering of user movement trajectories, and extracts dynamic prediction parameters of equipment full load time by combining spatiotemporal correlation modeling of equipment load data.

[0021] The load prediction module quantifies the maximum detour range based on the user's current location and path preference, uses the spatiotemporal prediction results of the equipment capacity consumption rate, and dynamically calculates the equipment load urgency and path selection resistance to generate an indicator reflecting the matching degree of load and path, and dynamically screens out a set of candidate devices that meet the remaining capacity requirements.

[0022] The path optimization module takes the user's detour distance as a hard constraint, the equipment's remaining capacity and the regional load difference as the optimization objectives, and solves the optimal delivery path through a multi-objective decision-making algorithm.

[0023] The feedback adjustment module adjusts the prediction parameters of the equipment load in the spatiotemporal correlation model based on user behavior feedback data, and simultaneously optimizes the path screening threshold to achieve closed-loop optimization of prediction and scheduling.

[0024] In the context of intelligent urban waste sorting and recycling, the problem of equipment load imbalance significantly restricts resource utilization efficiency and user delivery experience. There are significant differences in the spatiotemporal distribution of delivery behaviors between office areas and residential areas. For example, the concentrated delivery of disposable lunch boxes in office areas at noon causes the equipment to be fully loaded, while the peak of kitchen waste in residential areas at night is prone to overflow. This temporal and spatial unevenness makes it difficult for traditional capacity prediction based on historical data of a single device to adapt to the dynamic migration characteristics of user behavior. To this end, the present invention proposes to construct an accurate equipment full load time prediction mechanism through the spatiotemporal coupling analysis of user movement trajectories and equipment load data, so as to optimize garbage sorting paths and resource scheduling.

[0025] The trajectory clustering module includes the following: S1.1, spatiotemporal clustering of user movement trajectories: In the spatiotemporal clustering of user mobile trajectories, firstly, the legal mobile trajectory data actively uploaded by users are collected. These data are recorded as a set of trajectory points. Each trajectory point in the set contains the user's location information at a specific time, that is, the timestamp and geographic coordinates (longitude and latitude). Next, the density-based spatiotemporal clustering algorithm is used to process these trajectory points. Specifically, the extended version of the DBSCAN algorithm is used, which considers the time and space dimensions jointly, that is, the three dimensions of timestamp, longitude and latitude are analyzed simultaneously. In the clustering process, a neighborhood distance threshold and a minimum point number threshold are set. The neighborhood distance threshold is dynamically adjusted according to the density of user delivery behavior, and the minimum point number threshold is set as the minimum number of trajectory points that must be included in each cluster. Through the clustering algorithm, the trajectory points are divided into several spatiotemporal clusters, each of which represents a group of trajectory points that are close to each other in time and space. Subsequently, each spatiotemporal cluster is counted, and the number of occurrences of the trajectory point in a specific time period (such as every hour) is calculated, and it is defined as the occurrence frequency. The occurrence frequency is compared with a preset frequency threshold to screen out the spatiotemporal clusters with an occurrence frequency higher than the threshold. From these screened spatiotemporal clusters, users’ high-frequency delivery paths are further extracted, which reflect their delivery habits in a specific time period.

[0026] S1.2, Modeling the spatiotemporal correlation of equipment load data: In the modeling of the spatiotemporal correlation of device load data, the load data of the device is first collected. These data include the device ID, timestamp, and the current load of the device at the timestamp. Then, a vector autoregression model is constructed to analyze the dynamic relationship of the device load in time and space. Specifically, a load vector is defined to represent the set of loads of all devices at a specific moment. In this model, the change of the load vector is affected by both time and space factors. The time factor is represented by introducing a time lag term, that is, considering the impact of the load at several moments in the past on the current load; the spatial factor is represented by constructing a spatial adjacency matrix, which describes the spatial proximity relationship between devices. The coefficients in the model are estimated by the least squares method, specifically, the load vector is fitted with the weighted combination of its time lag term and the spatial adjacency matrix, and the value of each coefficient is calculated, thereby capturing the change law of the device load in time and space. Based on the prediction results of the model, the load change rate of each device is further calculated, which is defined as the increase in load per unit time, reflecting the growth trend of the device load.

[0027] S1.3, Dynamic prediction parameter extraction of equipment full load time: In the extraction of dynamic prediction parameters of equipment full load time, firstly, based on the load change rate obtained by the aforementioned vector autoregression model, the capacity consumption rate of each device is defined, which represents the increase in the load of the device per unit time. Then, the full load time of each device is predicted, assuming that the full load capacity of the device is a known value and the current load is the load at the current moment. The calculation process of the full load time is as follows: first, the difference between the current load and the full load capacity is calculated, that is, the full load capacity minus the current load is used to obtain the remaining capacity; then the remaining capacity is divided by the capacity consumption rate to obtain the time required to increase from the current load to the full load capacity; finally, this time is added to the current time to obtain the specific time point when the device is expected to reach full load. If the capacity consumption rate is greater than zero, it means that the load is increasing, and the full load time is calculated according to the above method; if the capacity consumption rate is less than or equal to zero, it means that there is no trend of load increase, and it is considered that the device will not be fully loaded in the short term, and the full load time is set to a very large value, indicating that the full load will not occur. Finally, dynamic prediction parameters are generated for each device, which include the capacity consumption rate and the predicted full load time, which fully record the dynamic change characteristics of the equipment load.

[0028] The trajectory clustering module accurately identifies the user's high-frequency delivery path by clustering the user's mobile trajectory in time and space, providing a behavioral basis for path optimization; through the spatiotemporal correlation modeling of equipment load data, the equipment capacity consumption rate and full load time are calculated to form dynamic prediction parameters. These parameters accurately reflect the spatiotemporal dynamic characteristics of equipment load, providing reliable data support for the generation of candidate equipment sets in the load prediction module, ensuring that subsequent scheduling can adapt to real-time changes in user behavior and equipment status.

[0029] In the trajectory clustering module, the dynamic prediction parameters of the equipment full load time are extracted by modeling the spatiotemporal clustering of user movement trajectories and the spatiotemporal correlation of equipment load data. These parameters characterize the dynamic change characteristics of equipment load and provide basic data support for the load prediction module. The goal of the load prediction module is to dynamically generate a set of candidate devices that meet the remaining capacity requirements based on the user's current location and path preference, combined with the spatiotemporal prediction results of the trajectory clustering module, to lay the foundation for solving the optimal delivery path in the subsequent path optimization module, and to achieve user delivery path optimization and equipment load balancing at the same time.

[0030] The load prediction module includes the following: S2.1, Analysis and modeling of user’s current location and path preference: In the analysis and modeling of the user's current location and path preference, the user's current location is first obtained through real-time positioning technology, which is expressed as longitude and latitude in the geographic coordinate system. Then, the user's historical delivery behavior data is used to analyze the path type that the user tends to choose, such as the shortest path, the fastest path, or the habitual path, and a preference feature vector is extracted by counting the characteristics of the user's historical delivery path. The dimension of this vector is determined according to the specific scenario, such as distance, time, or detour frequency. Then, the maximum detour distance acceptable to the user is calculated by combining the user's current location and the preference feature vector. The specific calculation method is: first calculate the average value of the user's historical detour distance; then calculate the modulus of the preference feature vector, that is, the square root of the sum of the squares of the values ​​of each dimension; then divide the average value of the historical detour distance by the natural logarithm of the modulus of the preference feature vector plus 1 to obtain the maximum detour distance. Finally, with the user's current location as the center and the maximum detour distance as the radius, the user's acceptable detour range is defined.

[0031] S2.2, Application of spatiotemporal prediction results of equipment capacity consumption rate: In the application of the spatiotemporal prediction results of the equipment capacity consumption rate, the dynamic prediction parameters of the equipment are first obtained from the trajectory clustering module, including the equipment capacity consumption rate and the predicted full load time. Then, the coupling coefficient is introduced to measure the matching degree between the load state of the equipment and the user's path selection. This coefficient is calculated by combining the spatiotemporal inertia attenuation index and the path topology resistance index.

[0032] Spatiotemporal inertia decay index: This index combines the instantaneous change gradient of the equipment capacity consumption rate and the information entropy of the user's historical delivery time distribution, and is weighted by the inverse of the regional traffic density. The specific calculation method is: first calculate the rate of change of the capacity consumption rate over time, that is, the time derivative; then calculate the information entropy of the user's historical delivery time distribution to measure the regularity of the time distribution; then multiply the time derivative by the information entropy, and then divide it by the inverse of the regional traffic density to obtain the spatiotemporal inertia decay index. The spatiotemporal inertia decay index quantifies the urgency of the equipment load in the time and space dimensions, reflecting the load growth trend of the equipment at a specific time and location and its priority in responding to demand. The larger the value, the higher the spatiotemporal urgency of the equipment load, which means that the equipment is currently facing strong load growth pressure and needs to prioritize delivery guidance or resource scheduling to alleviate the situation.

[0033] Path topology resistance index: This index is based on the path network topology depth difference from the user's current location to the device location, that is, the shortest path length in the path network, and superimposed with the attenuation memory factor of the user's historical detour behavior. The specific calculation method is: first determine the topology depth difference; then calculate the attenuation memory factor, which is obtained by taking an exponential function of the product of a negative attenuation coefficient and the time interval of the historical detour behavior; finally, multiply the topology depth difference by the result of the exponential function to obtain the path topology resistance index. The path topology resistance index measures the potential resistance of users when choosing a certain path, and reflects the degree of deviation between path selection and user habits. The larger the value, the greater the resistance to users choosing the path, that is, the higher the degree to which the path deviates from the user's habitual trajectory, resulting in a lower possibility that users will choose the path in actual delivery.

[0034] Coupling coefficient: This coefficient is based on the spatiotemporal inertia attenuation index and is adjusted by the inverse tangent function of the path topology resistance index and the regional equipment load difference. The specific calculation method is: first calculate the ratio of the spatiotemporal inertia attenuation index to the path topology resistance index, and map the ratio to between 0 and 1 through the hyperbolic tangent function; then calculate the inverse tangent function of the path topology resistance index, and use the regional equipment load difference (that is, the entropy value of the remaining capacity of the equipment in the region) as the power adjustment; finally, multiply the result of the hyperbolic tangent function by the adjustment factor to obtain the coupling coefficient. The coupling coefficient measures the degree of match between the equipment load status and the user's path selection. This coefficient comprehensively considers the urgency of the equipment load and the accessibility of the user's path, reflecting the coordination between the two. The larger the value, the higher the match between the equipment load status and the user's path selection, which means that the device is both within the user's acceptable path range and has a load status suitable for delivery, and is an ideal choice for user delivery.

[0035] S2.3, Dynamic Generation of Candidate Device Sets: In the dynamic generation of candidate device sets, the remaining capacity constraint of the device is first set, that is, after the full load capacity of the device is subtracted from the current load, the remaining capacity must be greater than or equal to the preset remaining capacity threshold. Then, based on the user's maximum detour range, remaining capacity requirements and coupling coefficient, the candidate device set is filtered. The filtering conditions include the following three aspects: Geographic constraint: Calculates the geographic distance between the device location and the user's current location. The distance is determined by the spherical distance calculation method and is required to be less than or equal to the maximum detour distance threshold acceptable to the user.

[0036] Capacity constraint: Ensure that the remaining capacity of the device is greater than or equal to the preset remaining capacity threshold to ensure that the device still has available capacity when the user arrives.

[0037] Coupling constraint: The coupling coefficient of the device is required to be greater than or equal to a preset coupling coefficient threshold, such as 0.5, to ensure that the matching degree between the device and the user path meets the standard.

[0038] Finally, by combining the above geographical constraints, capacity constraints and coupling constraints, a candidate device set is dynamically generated, which contains all devices that meet the conditions.

[0039] The load prediction module accurately defines the maximum detour range by analyzing the user's current location and path preference feature vector. Based on the dynamic prediction parameters of the equipment capacity consumption rate and predicted full load time provided by the trajectory clustering module, combined with the dynamic calculation of the equipment load urgency and path selection resistance, a coupling coefficient reflecting the matching degree of load and path is generated. Finally, through multiple screening of geographical distance, remaining capacity and matching degree indicators, a set of candidate equipment that meets the remaining capacity requirements is dynamically formed, which provides a high-quality equipment selection basis for the multi-objective path optimization of the path optimization module, ensuring the efficiency of the user's delivery path and the balance of equipment load.

[0040] In the load prediction module, the maximum detour range is defined by the user's current location and path preference feature vector, and a candidate device set is generated using the spatiotemporal prediction results of the device capacity consumption rate, where each device meets the constraints of geographical distance, remaining capacity, and coupling coefficient. These results provide a selectable device set for the path optimization module, aiming to solve the collaborative problem of user delivery path optimization and device load balancing. The path optimization module uses the user's detour distance as a hard constraint, and the device's remaining capacity and regional load differences as optimization targets. It solves the optimal delivery path through a multi-objective decision algorithm to ensure a balance between user convenience and device resource utilization efficiency.

[0041] like Figure 2 As shown, the path optimization module includes the following: S3.1, User detour distance: The user detour distance refers to the geographical distance from the user's current location to the candidate device location, which is determined by the spherical distance calculation method. The calculation process first needs to obtain the longitude and latitude coordinates of the user's current location, as well as the longitude and latitude coordinates of the candidate device location. Next, the Haversine formula is used to calculate the spherical distance between the two points. This method takes into account the curvature of the earth and converts the longitude and latitude coordinates into a straight-line distance in kilometers or meters to obtain the distance value between the user and the candidate device. Finally, the hard constraint requires that this distance must be less than or equal to the maximum detour distance threshold that the user can accept, to ensure that the user will not feel inconvenienced when going to the delivery due to excessive path deviation.

[0042] S3.2, equipment remaining capacity optimization: The optimization of device remaining capacity aims to select devices with larger remaining capacity to ensure that there is still more available space on the device after delivery. The calculation process first determines the full load capacity of each device, that is, the maximum load that the device can accommodate. Then, the current load of each device is obtained, and the current load is subtracted from the full load capacity to obtain the remaining capacity of the device. The optimization goal is to give priority to devices with larger remaining capacity so that the available space of the device will not be exhausted too quickly after the user delivers, thereby improving the utilization efficiency of the device.

[0043] S3.3, regional load difference optimization: Regional load difference optimization is achieved by measuring the uniformity of the load distribution of all devices in the region, with the goal of minimizing the degree of dispersion of the load distribution after user delivery. The calculation process first counts the sum of the loads of all devices in the region and divides it by the number of devices to obtain the average load of the devices in the region. Next, for each device, calculate the ratio of its load to the average load, then subtract 1 from the ratio and take its absolute value to obtain the degree of deviation of the device. In order to amplify the impact of the deviation effect, the absolute value is raised to a power greater than 1, such as 2, to generate the deviation index of the device. Subsequently, the deviation indices of all devices are added to obtain the load dispersion of the region. The optimization goal is to minimize the change in regional load dispersion as much as possible by selecting appropriate devices after user delivery.

[0044] S3.4, initial solution set construction of multi-objective decision-making algorithm: The multi-objective decision-making algorithm is based on the improved non-dominated sorting method. First, the initial solution set is constructed. The initial solution set consists of a set of candidate devices, each of which corresponds to a candidate path from the user's current location to the device's location. The candidate device set is part of the input parameters and contains all devices that meet the user's hard constraint of detour distance, which serves as the basis for subsequent optimization goal evaluation.

[0045] S3.5, Objective evaluation process of multi-objective decision-making algorithm: The target evaluation calculates the values ​​of two optimization targets for each candidate device. First, for the remaining capacity of the device, the result is obtained directly based on the full load capacity of the device minus the current load. Secondly, for the regional load dispersion, the load of the device is updated after the user adds the delivery volume to the candidate device, and then the average load of all devices in the area is recalculated. For each device, calculate the ratio of its updated load to the new average load, subtract 1 and take the absolute value, and then take a power greater than 1, such as 2, to obtain the deviation index. Add the deviation indexes of all devices to obtain the regional load dispersion of the candidate device after the user delivery. Through this process, two target values ​​​​are generated for each candidate device: the remaining capacity of the device and the regional load dispersion after delivery.

[0046] S3.6, non-dominated sorting process of multi-objective decision-making algorithm: Non-dominated sorting is used to select the optimal solution set from the candidate devices and form the Pareto frontier. The sorting process compares all candidate devices and evaluates them based on two objectives: the remaining capacity of the device and the dispersion of the regional load after delivery. For any two devices, if the remaining capacity of one device is greater than that of the other device, and its dispersion of the regional load after delivery is less than or equal to that of the other device, the former is determined to dominate the latter. By comparing one by one, all candidate devices are divided into multiple levels, and the non-dominated devices form the Pareto frontier solution set, which represents a solution that cannot be surpassed by other devices at the same time in terms of two objectives.

[0047] S3.7, solution set and comprehensive score calculation of multi-objective decision-making algorithm: In order to select the only optimal device from the Pareto frontier solution set, the target preference factor is introduced and the comprehensive score is calculated. The target preference factor is a value between 0 and 1, which is used to balance the importance of the remaining capacity of the device and the regional load dispersion. When calculating the comprehensive score, the remaining capacity of the device is first normalized. The remaining capacity of each device is divided by the maximum full load capacity of all devices to obtain a normalized value, which is then multiplied by the target preference factor. Secondly, the regional load dispersion after delivery is normalized. The minimum and maximum values ​​of the regional load dispersion of all candidate devices after delivery are first calculated. The difference between the load dispersion of the current device and the minimum value is subtracted from 1 and divided by the difference between the maximum value and the minimum value to obtain a normalized value, which is then multiplied by 1 minus the target preference factor. Finally, the normalized result of the remaining capacity of the device is added to the normalized result of the regional load dispersion to obtain the comprehensive score of the device. The device with the highest comprehensive score is selected as the optimal device.

[0048] S3.8, optimal path generation of multi-objective decision-making algorithm: After determining the optimal device, the shortest path from the user's current location to the device's location is calculated as the optimal delivery path. The calculation process uses the Dijkstra algorithm, which first takes the user's current location as the starting point and the optimal device location as the end point, and builds a path network based on geographic distance. The algorithm starts from the starting point and gradually explores all possible paths, records the cumulative distance to each node, and always selects the node with the smallest current cumulative distance for expansion until it reaches the end point. Finally, the shortest path from the user's current location to the optimal device location is generated as the recommended route for user delivery.

[0049] The candidate device set is first screened by the hard constraint of the user's detour distance to ensure that the geographical distance of all devices meets the requirements. Then, the remaining capacity of the device and the dispersion of the regional load after delivery are calculated based on these parameters, and the multi-objective decision algorithm is executed. The output result is the optimal device and its corresponding shortest path, which is used to update the prediction parameters and adjust the thresholds in the subsequent steps.

[0050] The processing technology logic of the path optimization module takes the user's detour distance as a hard constraint, optimizes the two goals of equipment remaining capacity and regional load dispersion, and uses an improved non-dominated sorting method to determine the optimal delivery path. The calculation process includes the screening of candidate devices, the evaluation of target values, non-dominated sorting, the refinement of comprehensive scores, and the generation of the shortest path to ensure the efficiency of user delivery and the balance of equipment load.

[0051] The feedback adjustment module includes the following: S4.1, Collect and process user behavior feedback data: In the process of collecting and processing user behavior feedback data, it is first necessary to record the relevant information of the user's actual delivery, including the device selected by the user, the time when the delivery occurs, the number of items delivered, and the path actually selected by the user. These actual data are then compared with the recommended device, recommended delivery time, and recommended delivery volume in the path optimization module to calculate three types of deviation data. Specifically, the device selection deviation is determined by judging whether the device actually delivered by the user is consistent with the recommended device. If it is consistent, the deviation is zero, and if it is inconsistent, the deviation is a non-zero value; the time deviation is obtained by calculating the difference between the user's actual delivery time and the recommended delivery time, indicating the time difference between the two; the delivery volume deviation is determined by calculating the difference between the user's actual delivery volume and the recommended delivery volume, reflecting the gap between the actual delivery volume and the predicted value. These deviation data will be used as the basis for subsequent parameter adjustments.

[0052] S4.2, adjust equipment load prediction parameters: When adjusting the prediction parameters of the device load, the actual capacity consumption rate is first calculated by dividing the user's actual delivery volume by the delivery time interval to obtain the capacity consumption of the device per unit time. Next, this actual capacity consumption rate is compared with the capacity consumption rate predicted in the trajectory clustering module, and the predicted value is updated using a nonlinear fusion method. The specific update process is: take the predicted capacity consumption rate as the base value, and then add an adjustment item; the adjustment item is determined by calculating the ratio of the difference between the actual capacity consumption rate and the predicted capacity consumption rate, and introduces an adjustment factor to control the magnitude of the adjustment. The updated capacity consumption rate is then used to recalculate the full load time of the device. The calculation method is: take the current time as the starting point, add the quotient obtained by dividing the remaining capacity of the device by the updated capacity consumption rate, and the result is the time point when the device is expected to be fully loaded.

[0053] S4.3, adjust the maximum detour distance threshold: When adjusting the maximum detour distance threshold in the path screening threshold, first check whether the device actually delivered by the user is the same as the recommended device, and whether the actual detour distance exceeds the current maximum detour distance threshold. If these two conditions are met, the maximum detour distance threshold needs to be adjusted. The specific adjustment method is: take the current maximum detour distance threshold as the base value, plus an adjustment item; the adjustment item is determined by calculating the ratio of the difference between the actual detour distance and the current maximum detour distance threshold, and introduces an adjustment factor to control the magnitude of the adjustment. Through this process, the maximum detour distance threshold is increased to adapt to the greater detour needs that may occur in the actual behavior of the user. The path screening threshold includes the maximum detour distance threshold, the remaining capacity threshold, and the coupling coefficient threshold.

[0054] S4.4, adjust the remaining capacity threshold: For the adjustment of the remaining capacity threshold, first check whether the remaining capacity of the user's device after delivery is lower than the current remaining capacity threshold. If it is lower than the current threshold, the remaining capacity threshold needs to be adjusted. The specific adjustment method is: take the current remaining capacity threshold as the base value and add an adjustment item; the adjustment item is determined by calculating the ratio of the actual remaining capacity of the device after delivery and the current remaining capacity threshold, and introduces an adjustment factor to control the magnitude of the adjustment. Through this process, the remaining capacity threshold is increased to ensure that the recommended devices in the future still have sufficient remaining capacity to meet the needs after delivery.

[0055] S4.5, adjust the coupling coefficient threshold: For the adjustment of the coupling coefficient threshold, first check whether the device actually delivered by the user is different from the recommended device, and whether the coupling coefficient of the actual path is lower than the current coupling coefficient threshold. If these two conditions are met, the coupling coefficient threshold needs to be adjusted. The specific adjustment method is: take the current coupling coefficient threshold as the base value, minus an adjustment item; the adjustment item is determined by calculating the ratio of the difference between the actual coupling coefficient and the current coupling coefficient threshold, and introduces an adjustment factor to control the magnitude of the adjustment. Through this process, the coupling coefficient threshold is lowered, thereby relaxing the path screening conditions and allowing more potential paths to be included in the recommendation range.

[0056] S4.6, closed-loop optimization mechanism: In the closed-loop optimization mechanism, the adjusted equipment load prediction parameters, that is, the updated capacity consumption rate, are fed back to the trajectory clustering module for the prediction calculation of the next equipment full load time. At the same time, the adjusted path screening thresholds, including the maximum detour distance threshold, the remaining capacity threshold, and the coupling coefficient threshold, are fed back to the load prediction module for the subsequent screening process of the candidate equipment set. By continuously collecting user behavior feedback data and repeatedly performing the above parameter and threshold adjustment process, the prediction parameters and screening thresholds gradually approach the actual distribution of user behavior characteristics and the actual status of the equipment, thereby achieving a closed-loop optimization effect of prediction and scheduling.

[0057] The feedback adjustment module analyzes user behavior feedback data, calculates device selection deviation, time deviation and delivery volume deviation, and then adjusts the prediction parameters of device load and path screening thresholds, including the maximum detour distance threshold, remaining capacity threshold and coupling coefficient threshold. The adjustment results are fed back to the previous steps to form a closed-loop optimization mechanism, ultimately improving the accuracy of device full load time prediction and the adaptability of path screening.

[0058] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0059] It should be noted that the system of the present invention can be deployed on the device itself to realize embedded applications, and can also be run on a PC or other terminal with a user interface, thereby meeting a variety of hardware environments and usage requirements.

[0060] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0061] It should be noted that, in this article, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0062] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A multi-mode interactive garbage classification guidance and recycling system, characterized in that: include: Trajectory clustering module, load prediction module, path optimization module and feedback adjustment module; The trajectory clustering module identifies high-frequency delivery paths through spatiotemporal clustering of user movement trajectories, and extracts dynamic prediction parameters of equipment full load time by combining spatiotemporal correlation modeling of equipment load data; The load prediction module quantifies the maximum detour range based on the user's current location and path preference, uses the spatiotemporal prediction results of the equipment capacity consumption rate, and dynamically calculates the equipment load urgency and path selection resistance to generate an indicator reflecting the matching degree of load and path, and dynamically selects a set of candidate devices that meet the remaining capacity requirements; The route optimization module uses the user's detour distance as a hard constraint, the remaining capacity of the equipment and the regional load difference as the optimization target, and solves the optimal delivery route through a multi-objective decision-making algorithm; The feedback adjustment module adjusts the prediction parameters of the equipment load in the spatiotemporal correlation model based on user behavior feedback data, and simultaneously optimizes the path screening threshold to achieve closed-loop optimization of prediction and scheduling.

2. The multi-mode interactive garbage classification guidance and recycling system according to claim 1 is characterized in that: The trajectory clustering module includes the following: High-frequency delivery paths are identified through spatiotemporal clustering of legitimate mobile trajectories actively provided by users, specifically including collecting trajectory point sets uploaded by users, clustering the trajectory point sets in the spatiotemporal dimension using a density-based spatiotemporal clustering algorithm, generating multiple spatiotemporal clusters, and screening out high-frequency delivery paths by counting the frequency of occurrence of trajectory points in each spatiotemporal cluster within a specific time period; at the same time, the spatiotemporal correlation modeling of equipment load data is combined, specifically including constructing a vector autoregression model to analyze the dynamic changes of equipment load in time and space, calculate the equipment load change rate, and then extract the dynamic prediction parameters of the equipment full load time.

3. The multi-mode interactive garbage classification guidance and recycling system according to claim 2 is characterized by: The dynamic prediction parameters include capacity consumption rate and predicted full load time. The predicted full load time is calculated by the ratio of remaining capacity to capacity consumption rate.

4. The multi-mode interactive garbage classification guidance and recycling system according to claim 2 is characterized in that: The load prediction module includes the following: The user's current location is obtained in real time, and the preference feature vector is extracted by analyzing the user's historical delivery behavior data. The maximum detour distance threshold acceptable to the user is calculated to determine the detour range; then the dynamic prediction parameters of the device are obtained from the trajectory clustering module, and the coupling coefficient is calculated using the device capacity consumption rate; finally, a set of candidate devices is dynamically generated through multiple screening of the user's maximum detour range, the device's remaining capacity greater than the remaining capacity threshold, and the coupling coefficient greater than or equal to the preset coupling coefficient threshold.

5. The multi-mode interactive garbage classification guidance and recycling system according to claim 4 is characterized in that: The coupling coefficient is obtained by adjusting the ratio of the spatiotemporal inertia attenuation index to the path topology resistance index and the regional equipment load difference.

6. The multi-mode interactive garbage classification guidance and recycling system according to claim 5, characterized in that: The spatiotemporal inertia attenuation index is calculated by weighting the instantaneous change gradient of the equipment capacity consumption rate and the information entropy of the user's historical delivery time distribution by the inverse of the regional traffic density.

7. The multi-mode interactive garbage classification guidance and recycling system according to claim 5, characterized in that: The path topology resistance index is obtained by multiplying the path network topology depth difference from the user's current location to the device location by the attenuation memory factor of the user's historical detour behavior.

8. The multi-mode interactive garbage classification guidance and recycling system according to claim 4, characterized in that: The path optimization module includes the following: Taking the user detour distance as the hard constraint condition, and the device remaining capacity and regional load difference as the optimization objectives, the geographical distance from the user's current location to the candidate device location is first calculated to verify whether the hard constraint of the user's detour distance is met; then the device remaining capacity is defined as the device's full load capacity minus the current load, and the regional load difference is defined as the sum of the deviation indexes of all device loads and the average load; then the improved non-dominated sorting method is used to construct the initial solution set, and the device remaining capacity and regional load difference after delivery are evaluated for each candidate device one by one, and the Pareto frontier solution set is generated through non-dominated sorting; Then, the target preference factor is introduced to calculate the comprehensive score of the solutions in the Pareto frontier solution set, and the device with the highest comprehensive score is selected as the optimal device; finally, the Dijkstra algorithm is used to generate the shortest path from the user's current location to the optimal device location, thereby solving the optimal delivery path.

9. The multi-mode interactive garbage classification guidance and recycling system according to claim 8, characterized in that: The feedback adjustment module includes the following: Firstly, multiple feedback data of users' actual delivery are collected and compared with the recommendation results generated by the path optimization module, and the equipment selection deviation, delivery time deviation and delivery volume deviation are analyzed. Then, by comparing the difference between the actual capacity consumption rate and the predicted capacity consumption rate, the capacity consumption rate of the equipment is updated by a nonlinear fusion method, and the full load time of the equipment is recalculated according to the updated capacity consumption rate. At the same time, according to the actual detour distance of the user, the remaining capacity of the equipment after delivery and the coupling coefficient of the actual path, the maximum detour distance threshold, the remaining capacity threshold and the coupling coefficient threshold are adjusted respectively. Finally, the updated prediction parameters and path screening threshold are fed back to the trajectory clustering module and the load prediction module.

Images