Path planning method and system

Abstract

The invention discloses a path planning method and system applied to a material distribution system, and the method comprises the steps: processing the obtained distribution information of each distribution task, determining a first distribution mode of each distribution task, and determining a first distribution scheme based on the first distribution mode; determining distribution mode influence information according to the dependency relationship among the distribution tasks; performing primary optimization on the first delivery mode to obtain a second delivery mode corresponding to each optimized delivery task; determining a resource tension degree based on distribution resources in a material distribution system corresponding to the second distribution mode; inputting the second delivery mode, the resource tension degree and the selected delivery condition into a constructed multi-objective optimization model to perform secondary optimization processing on a first delivery path in the first delivery scheme to obtain a second delivery path; and sending the generated control signals to the corresponding distribution terminals. According to the path planning method and system provided by the invention, the distribution efficiency can be remarkably improved.

Description

A path planning method and system Technical Field

[0001] This invention relates to the field of path planning technology, and in particular to a path planning method and system. Background Technology

[0002] During the operation of a delivery system, reasonable route planning can significantly shorten delivery time and travel distance, reduce fuel or electricity consumption, and improve delivery efficiency and resource utilization.

[0003] The existing route planning method is to formulate delivery priorities based on the specific requirements of each delivery task, and then execute the delivery tasks in sequence according to the delivery priorities. As shown in Figure 1, Figure 1 is a schematic diagram of the existing delivery technology route. A is the delivery starting point, B and D are ordinary delivery points, and C is a delivery task point with special requirements. According to the existing route planning method, the resulting delivery scheme is shown in delivery scheme (1) in Figure 1. In delivery scheme (1), C, which has special requirements, will be delivered first, and then B and D will be delivered in sequence according to their distance, and finally the route will return to the starting point A. When C and D are very close, it is obvious that delivering D first will save more time and distance than delivering B according to its distance. It can be seen that the existing technology scheme will lead to a decrease in delivery timeliness, an increase in energy and transportation capacity waste, and thus a significant decrease in the efficiency of the delivery system. Summary of the Invention

[0004] This invention provides a route planning method and system to solve the technical problem of low delivery timeliness in existing technologies, thereby improving delivery efficiency.

[0005] The process involves processing the delivery information of each delivery task to determine a first delivery method for each task and a first delivery plan based on the first delivery method. It also involves determining the impact information of the delivery methods based on the dependencies between the delivery tasks, performing a first-level optimization on the first delivery method based on the impact information and the geographical information of each delivery task, and obtaining a second delivery method corresponding to each optimized delivery task. Furthermore, it involves determining the resource scarcity level based on the delivery resources in the material distribution system corresponding to the second delivery method, inputting the second delivery method, the resource scarcity level, and the selected delivery conditions into a pre-constructed multi-objective optimization model to perform a second-level optimization on the first delivery path in the first delivery plan, and obtaining a second delivery path. Finally, it involves sending the generated control signals to the corresponding delivery terminals, wherein the control signals are obtained at least by the second delivery method and the second delivery path determined by the second delivery plan.

[0006] Preferably, the step of processing the delivery information of each delivery task to determine a first delivery method for each delivery task and determining a first delivery plan based on the first delivery method includes: extracting the delivery information from the delivery task; analyzing the delivery information to determine the first delivery method for each delivery task; clustering the delivery tasks of the same type based on the first delivery method, performing preliminary path planning processing based on the clustering results, and determining the first delivery plan.

[0007] Preferably, determining the delivery method impact information based on the dependency relationships between the various delivery tasks includes: constructing a task dependency graph based on the dependency relationships between the various delivery tasks; and performing a quantitative analysis of the collaborative impact caused by the change in the first delivery method for each of the delivery tasks based on the task dependency graph to determine the delivery method impact information.

[0008] Preferably, the step of performing first-level optimization on the first delivery method based on the delivery method impact information and the geographical information of each delivery task to obtain the optimized second delivery method corresponding to each delivery task includes: identifying abnormal tasks based on the delivery method impact information and the geographical information of each delivery task; correcting the abnormal tasks according to preset conflict resolution rules, and obtaining the second delivery method corresponding to each delivery task from the correction result.

[0009] Preferably, determining the resource scarcity level based on the delivery resources in the material delivery system corresponding to the second delivery method includes: based on the second delivery method, calculating the allocated task quantity and available total quantity of the delivery resources in the material delivery system within the corresponding spatiotemporal region; determining the occupancy ratio of the delivery resources in the spatiotemporal region based on the comparison result of the allocated task quantity and the available total quantity; and using the occupancy ratio to characterize the resource scarcity level of the delivery resources.

[0010] In another aspect, the present invention provides a path planning system applied to a material distribution system, comprising: an acquisition module for processing the acquired delivery information of various delivery tasks, determining a first delivery method for each delivery task, and determining a first delivery scheme based on the first delivery method; a determination module for determining delivery method impact information based on the dependency relationship between the various delivery tasks; a first-level module for performing first-level optimization on the first delivery method based on the delivery method impact information and the geographical information of each delivery task, to obtain a second delivery method corresponding to each delivery task after optimization; a resource module for determining the resource scarcity level based on the delivery resources in the material distribution system corresponding to the second delivery method; a second-level module for inputting the second delivery method, the resource scarcity level, and selected delivery conditions into a pre-constructed multi-objective optimization model to perform second-level optimization on the first delivery path in the first delivery scheme, to obtain a second delivery path; and a control module for sending generated control signals to corresponding delivery terminals, wherein the control signals are obtained at least by the second delivery method and the second delivery scheme determined by the second delivery path.

[0011] Preferably, the acquisition module includes: an extraction unit for extracting the delivery information from the delivery task; an analysis unit for analyzing the delivery information to determine the first delivery method for each delivery task; and a clustering unit for clustering the delivery tasks based on the first delivery method, performing preliminary path planning processing based on the clustering results, and determining the first delivery scheme.

[0012] Preferably, the determining module includes: a dependency graph unit, used to construct a task dependency graph based on the dependency relationships between the various delivery tasks; and a quantification unit, used to perform quantitative analysis on the collaborative impact caused by the change of the first delivery method for each of the delivery tasks based on the task dependency graph, and determine the impact information of the delivery method.

[0013] Preferably, the first-level module includes: an anomaly unit, used to identify anomaly tasks based on the delivery method impact information and the geographical information of each delivery task; and a correction unit, used to correct the anomaly tasks according to preset conflict resolution rules, so as to obtain the second delivery method corresponding to each delivery task from the correction result.

[0014] Preferably, the resource module includes: a statistics unit, used to calculate the allocated task quantity and available total quantity of the delivery resources in the material delivery system within the corresponding spatiotemporal region based on the second delivery method; an occupancy unit, used to determine the occupancy ratio of the delivery resources in the spatiotemporal region based on the comparison result of the allocated task quantity and the available total quantity; and a characterization unit, used to characterize the resource shortage level of the delivery resources using the occupancy ratio.

[0015] Compared to existing technologies, the beneficial effects of this invention are at least one of the following: By introducing a mechanism of dependency relationships between delivery tasks and mutual influence of delivery methods, this invention overcomes the limitations of existing technologies that only set priorities based on the specificity of a single task and mechanically execute them sequentially. Specifically, this invention first generates preliminary delivery methods and routes based on delivery information, then combines task dependencies to explore the synergistic or conflicting effects between delivery methods, and integrates geographic information to perform primary optimization of the initial delivery methods, thereby dynamically adjusting the delivery methods of each task to achieve global synergy. On this basis, it further combines real-time resource scarcity and multi-dimensional delivery conditions to perform secondary multi-objective optimization of the initial routes, avoiding detours, empty runs, or inefficient sequencing caused by handling special tasks in isolation. Compared to the rigid priority path shown in Figure 1 in the background technology (C first, then B, then D), this invention can automatically identify synergy opportunities when C and D are adjacent, rationally rearrange the service order, and effectively shorten the total travel distance and time. It significantly improves delivery timeliness and route rationality, reduces energy consumption and transportation capacity waste, thereby improving the overall intelligence level and operational efficiency of the material distribution system. Attached Figure Description

[0016] Figure 1 is a schematic diagram of prior art delivery in the background of this invention; Figure 2 is a flowchart of a path planning method in one embodiment of this invention; Figure 3 is a structural schematic diagram of a path planning system in one embodiment of this invention; Reference numerals: 11, acquisition module; 12, determination module; 13, first-level module; 14, resource module; 15, second-level module; 16, control module. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] In the description of this invention, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0019] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0020] In the description of this invention, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0021] In delivery systems, rational route planning can effectively shorten delivery time and travel distance, reduce energy consumption, and improve efficiency and resource utilization. Existing methods typically prioritize tasks based on their specific requirements and execute deliveries in that order. However, existing methods often lead to detours due to mechanically following priorities, resulting in decreased timeliness, wasted energy and transportation capacity, and significantly reduced overall system efficiency.

[0022] An embodiment of the present invention provides a path planning method. Specifically, please refer to Figure 2, which shows a flowchart of the path planning method in one embodiment of the present invention, including: S1, processing the obtained delivery information of each delivery task to determine a first delivery method for each delivery task, and determining a first delivery plan based on the first delivery method; S2, determining the delivery method influence information according to the dependency relationship between each delivery task; S3, performing a first-level optimization on the first delivery method based on the delivery method influence information and the geographical information of each delivery task to obtain a second delivery method corresponding to each delivery task after optimization; S4, determining the resource shortage level based on the delivery resources in the material delivery system corresponding to the second delivery method; S5, inputting the second delivery method, resource shortage level, and selected delivery conditions into a pre-constructed multi-objective optimization model to perform a second-level optimization on the first delivery path in the first delivery plan to obtain a second delivery path; S6, sending the generated control signals to the corresponding delivery terminals, wherein the control signals are obtained at least by the second delivery plan determined by the second delivery method and the second delivery path.

[0023] First, the delivery information of each delivery task is processed to determine the primary delivery method for each task, and a primary delivery plan is determined based on the primary delivery method. Delivery information is extracted from the delivery tasks; the delivery information is analyzed to determine the primary delivery method for each task; based on the primary delivery method, the delivery tasks are clustered into similar task groups, and the clustering results are used for preliminary path planning to determine the primary delivery plan.

[0024] Delivery information refers to various core data related to the delivery task, including the type, quantity, weight, volume, and special attributes of the delivered goods (such as fragility, perishability, and temperature control requirements); the specific geographical coordinates of the delivery origin and destination; delivery time requirements (such as earliest and latest delivery times and timeliness levels); recipient information (such as contact person's contact information and detailed delivery address); and the priority of the delivery task. This information forms the foundation for all subsequent delivery planning work. The primary delivery method refers to the specific delivery execution method initially determined based on the delivery information, including express delivery, full truckload delivery, less-than-truckload (LTL) delivery, cold chain delivery, UAV (Unmanned Aerial Vehicle) delivery, and same-city instant delivery, etc. Different delivery methods correspond to different applicable scenarios and resource requirements. The primary delivery plan refers to the initial delivery planning scheme formed based on the initially determined primary delivery method, which corely includes the delivery method arrangement for each delivery task and a preliminary delivery route planning framework.

[0025] First, delivery information is extracted from the delivery task order system or task issuance platform using data acquisition technology. Data cleaning techniques, such as imputation of missing values ​​using mean interpolation or nearest neighbor interpolation, are employed to handle incomplete information. Outlier detection techniques, such as the interquartile range method, are used to remove erroneous data to ensure information accuracy. Subsequently, feature analysis is performed on the cleaned delivery information using multi-attribute decision-making methods such as the Analytic Hierarchy Process (AHP). The process (analytic hierarchy process) combined with fuzzy comprehensive evaluation method constructs an evaluation index system from multiple dimensions such as delivery material attributes, delivery time requirements, delivery distance, and cost budget. It comprehensively scores and ranks each candidate delivery method, selects the method with the best score as the first delivery method, and then uses clustering algorithms such as K-means or density clustering algorithm to cluster similar tasks with the first delivery method as the core clustering feature and the geographical distribution of the delivery start and end points. Delivery tasks that use the same or similar first delivery methods and are geographically adjacent are grouped into one category. Based on the clustering results, a greedy algorithm or genetic algorithm is used for preliminary path planning. The path distance between each task is calculated by traversing the clustered task set, constructing an initial path network, and gradually optimizing it to form a preliminary delivery route and delivery order containing each clustered task group, which is the first delivery plan.

[0026] The purpose of this approach is to quickly develop a feasible initial delivery framework based on fundamental data, providing a foundation for subsequent optimization. This enables preliminary classification of delivery tasks and route planning, improving the efficiency of subsequent optimization. The advantages of these methods are that data cleaning techniques ensure the quality of delivery information and prevent planning deviations caused by erroneous data; the analytic hierarchy process (AHP) combined with fuzzy comprehensive evaluation can comprehensively consider multi-dimensional factors to achieve a scientific and reasonable selection of the first delivery method; clustering algorithms can effectively integrate similar tasks to reduce waste of delivery resources; and greedy algorithms and genetic algorithms can quickly achieve preliminary route planning to ensure the feasibility and basic efficiency of the first delivery plan.

[0027] Secondly, the impact information of delivery methods is determined based on the dependencies between various delivery tasks. A task dependency graph is constructed based on the dependencies between various delivery tasks; based on the task dependency graph, the collaborative impact caused by the change of the first delivery method for each delivery task is quantitatively analyzed to determine the impact information of delivery methods.

[0028] Dependency relationships refer to the sequential constraints or associations between different delivery tasks. These include sequential dependencies (where a subsequent task can only be executed after a preceding task is completed), resource dependencies (requiring the sharing of the same batch of materials or the same delivery equipment), and time-sensitivity dependencies (where a delay in one task leads to delays in other related tasks). For example, completing the delivery of materials for task A before replenishing goods for task B is a typical sequential dependency. Delivery method impact information refers to the quantitative data related to the degree and scope of the impact of a change in the primary delivery method of a delivery task on the delivery efficiency, delivery cost, delivery timeliness, and resource consumption of its related delivery tasks. A task dependency graph is a model that graphically presents the dependency relationships between delivery tasks. Nodes represent individual delivery tasks, and node attributes include task identifier, primary delivery method, and delivery timeliness requirements. Edges represent the dependency relationships between tasks, and edge attributes include dependency type and impact weight.

[0029] First, task analysis techniques are used to identify and extract dependencies by analyzing the constraints and related information of each delivery task. A graph modeling technique is then used to construct a task dependency graph, mapping each delivery task to a node and dependencies to directed edges between nodes. Weights are assigned to these edges, and the strength of the dependency is determined using the Analytic Hierarchy Process (AHP) combined with expert scoring. Next, based on the task dependency graph, an impact propagation analysis method combined with a Bayesian Network (BN) model is employed to quantify the collaborative impact of the first delivery method change. Impact indicators for the delivery method change are defined, including changes in delivery time, delivery costs, and resource conflict probability. Then, the Bayesian Network model is used to infer the propagation path and intensity of the impact of a delivery method change on the dependency graph, calculating the quantitative changes of each related task under different impact indicators. Finally, these quantitative data are integrated to form the delivery method impact information.

[0030] The reason for doing this is to clarify the chain reaction of changes in delivery methods and provide a basis for subsequent optimization. The advantage is that it can accurately grasp the synergistic effect of changes in delivery methods, ensuring the rationality and comprehensiveness of the optimization process. The integrated technologies and steps can ensure the completeness of dependency analysis and the accuracy of impact analysis, avoiding a decline in overall delivery efficiency due to local optimization.

[0031] Furthermore, based on the impact information of delivery methods and the geographical information of each delivery task, the first delivery method is optimized to obtain the second delivery method corresponding to each delivery task. Based on the impact information of delivery methods and the geographical information of each delivery task, abnormal tasks are identified; according to the preset conflict resolution rules, abnormal tasks are corrected, and the corrected results are used to obtain the second delivery method corresponding to each delivery task.

[0032] Geographic information refers to spatial location data related to each delivery task, including the latitude and longitude coordinates of the delivery origin and destination, administrative division information, and surrounding transportation network data such as road type, traffic capacity, and congestion status, as well as the geographical characteristics of the delivery area such as topography, business district distribution, and residential density. This data provides support for analyzing the spatial correlation and route feasibility of delivery tasks. Abnormal task pairs refer to paired delivery tasks identified through analysis of delivery method impact information and geographic information that exhibit delivery method conflicts or spatial mismatches. For example, two tasks may have a strong dependency relationship, but the original first delivery method leads to delivery time conflicts; or two tasks may be geographically adjacent, but significant differences in the first delivery method result in resource waste and route redundancy. Conflict resolution rules refer to pre-defined normative criteria used to correct abnormal task pairs and resolve delivery method conflicts. These include rules for unifying delivery methods based on dependency relationships, rules for integrating delivery methods based on geographical proximity, rules for adjusting delivery methods based on resource adaptability, and rules for prioritizing delivery methods based on timeliness requirements. Secondary delivery methods refer to delivery methods corrected through primary optimization, which are more aligned with task dependencies and geographical spatial characteristics, and are more rational and collaborative than the primary delivery method.

[0033] First, GIS (Geographic Information System) technology is used to perform spatial analysis on the geographic information of each delivery task. Spatial overlay analysis is used to obtain geographic distance and transportation accessibility data between tasks. Combined with the quantitative impact indicators in the delivery method impact information, an association rule mining algorithm is used to identify abnormal task pairs. By calculating the delivery method conflict coefficient and spatial matching degree of the task pair, when the conflict coefficient exceeds the preset threshold or the spatial matching degree is lower than the preset threshold, it is determined to be an abnormal task pair. Then, the preset conflict resolution rules are retrieved and the abnormal task pairs are corrected using the rule reasoning engine. If it is a dependency conflict, the delivery method of the strongly dependent task pair is adjusted to a consistent type that meets the dependency requirements according to the dependency unification rule. If it is a geographic spatial imbalance, the delivery method of adjacent tasks is adjusted to a type that can share delivery resources according to the geographic proximity integration rule. If there are multiple rules applicable, the weighted scoring method is used to determine the optimal correction scheme. After the correction is completed, the second delivery method corresponding to each delivery task can be obtained.

[0034] This is done to correct the inefficiencies of the initial delivery method and improve delivery coordination. By combining dependency effects and geographical features to optimize the delivery method, a good foundation is laid for subsequent route optimization. The integrated technologies and steps can accurately identify anomalies and scientifically resolve conflicts to ensure the overall adaptability of the delivery method and reduce subsequent resource waste and time delays.

[0035] Next, based on the delivery resources in the material distribution system corresponding to the second delivery method, the resource shortage level is determined. Based on the second delivery method, the allocated task volume and the total available volume of delivery resources in the material distribution system within the corresponding spatiotemporal region are statistically analyzed; based on the comparison results of the allocated task volume and the total available volume, the occupancy ratio of delivery resources in the spatiotemporal region is determined; the occupancy ratio is used to characterize the resource shortage level of delivery resources.

[0036] A materials distribution system refers to a comprehensive system that coordinates and manages distribution resources and executes distribution tasks. It includes core components such as a distribution resource management module, a task scheduling module, and a spatiotemporal monitoring module, responsible for integrating various distribution elements to ensure smooth operation of the distribution process. Distribution resources refer to various elements in the materials distribution system that can be used to execute distribution tasks, including delivery vehicles such as vans, refrigerated trucks, and UAVs, delivery personnel, warehousing sites, loading and unloading equipment, and transportation routes. Different distribution methods correspond to different distribution resource requirements. A spatiotemporal region refers to a resource scheduling unit divided by time and space dimensions. The spatial dimension can be divided into different regions according to administrative divisions or latitude and longitude grids. The time dimension can be divided into different time periods according to hours, days, and delivery time periods. The division of spatiotemporal regions can accurately locate the spatiotemporal distribution status of resources. The allocated task volume refers to the number of distribution tasks or corresponding resource occupancy that have been undertaken by the distribution resources corresponding to the second distribution method within a specific spatiotemporal region, such as the allocated transportation mileage of vehicles and the allocated working hours of personnel. Total available resources refer to the maximum number of tasks or the maximum resource supply that delivery resources corresponding to the second delivery method can handle within a specific time and space area, such as the total number of available vehicles and the total available man-hours. Utilization ratio refers to the ratio of allocated tasks to the total available resources, a core indicator for quantifying resource utilization. Resource scarcity refers to the supply and demand matching of delivery resources within a specific time and space area, reflecting whether resources can meet the current delivery task requirements. This is characterized by the utilization ratio; a higher utilization ratio indicates a higher resource scarcity, and vice versa.

[0037] First, based on the second delivery method, the types of delivery resources required for each delivery task are identified. The basic information of various delivery resources, including resource quantity, distribution location, and available time windows, is retrieved through the resource management module of the material delivery system. Simultaneously, combining the time requirements and geographical information of the delivery tasks, spatiotemporal regions are divided using a grid method. The delivery area is divided into spatial grids at fixed latitude and longitude intervals, and time segments are formed in 2-hour units to create a set of spatiotemporal regions. Next, data statistics are used to summarize the allocated task quantity of corresponding delivery resources in each spatiotemporal region. By linking tasks to spatiotemporal regions, the resource occupancy corresponding to the allocated tasks is accumulated. Simultaneously, the total available delivery resources in each spatiotemporal region are calculated. The maximum supply of usable resources is calculated using basic resource information. Then, the ratio of allocated task quantity to available total quantity in each spatiotemporal region is calculated to obtain the occupancy ratio. This is directly calculated using division, i.e., the occupancy ratio equals the allocated task quantity divided by the available total quantity. Finally, the occupancy ratio directly represents the resource tension level, with occupancy ratios of 0 to 0.3 indicating low tension, 0.3 to 0.7 indicating medium tension, and 0.7 to 1 indicating high tension.

[0038] The reason for doing this is to accurately grasp the supply and demand status of delivery resources, so as to provide a basis for resource constraints for subsequent route optimization. Through refined statistics in the spatiotemporal dimensions, we can accurately identify resource-scarce areas and time periods to ensure the feasibility and resource adaptability of subsequent optimization plans. The integrated technologies and steps can achieve quantitative assessment of the degree of resource scarcity, avoid excessive resource occupation or idleness, and improve resource utilization efficiency.

[0039] Then, the second delivery method, resource scarcity, and selected delivery conditions are input into the constructed multi-objective optimization model to perform secondary optimization on the first delivery path in the first delivery plan, resulting in the second delivery path. The selected delivery conditions refer to the core constraints and optimization guidelines preset based on actual delivery needs, including delivery cost constraints, delivery timeliness requirements, service quality standards, and resource usage restrictions. Delivery cost constraints specify the maximum cost threshold for a single delivery or overall delivery; delivery timeliness requirements stipulate the latest delivery time for each task; service quality standards include indicators such as material integrity rate and on-time delivery rate; and resource usage restrictions specify the maximum usage intensity of specific delivery resources. The multi-objective optimization model refers to a mathematical model built based on multiple mutually constraining optimization objectives to find the optimal solution under the premise of satisfying the constraints. The core optimization objectives here include minimizing the total delivery cost, minimizing the total delivery time, maximizing resource utilization, and improving the on-time delivery rate. Secondary optimization refers to the in-depth optimization of the delivery path based on the primary delivery method optimization, which is a further adjustment and improvement of the initial path in the first delivery plan. The second delivery route refers to the final delivery route determined after secondary optimization. Compared with the first delivery route, it is better suited to the optimized delivery method and resource constraints, while also meeting the selected delivery conditions.

[0040] First, the input parameters of the multi-objective optimization model are defined, including the resource demand parameters corresponding to the second delivery method, the resource constraint parameters corresponding to the resource scarcity level, and the constraint thresholds and objective weights corresponding to the selected delivery conditions. Next, the objective functions of the multi-objective optimization model are constructed, including a delivery cost minimization function (total cost equals transportation cost plus labor cost plus resource scheduling cost), a total delivery time minimization function (total time is the cumulative time of each task's delivery path), and a resource utilization maximization function (utilization rate equals the ratio of actual used resources to available resources). Simultaneously, the upper limit of the occupancy ratio corresponding to the resource scarcity level and the time-cost constraint in the selected delivery conditions are transformed into model constraints. The model construction can use a weighted sum method to transform the multi-objective model into a single-objective optimization. Then, intelligent optimization algorithms such as genetic algorithms or particle swarm optimization algorithms are used to solve the model. The first delivery path is used as the initial solution population, and the path information is encoded as gene fragments. The population is iteratively updated through crossover and mutation operations. The fitness value of each individual is calculated and determined by the weighted objective function values. Iteration stops when the preset number of iterations is reached or the fitness value stabilizes. The path corresponding to the optimal solution is the second delivery path. The reason for doing this is to achieve global optimization of delivery routes based on adapting delivery methods and resource status. By optimizing multiple objectives, the overall performance of the delivery solution is improved while taking into account core requirements such as cost, timeliness and resource utilization. The integrated technologies and steps ensure the scientific nature and feasibility of the optimization process, making the final route more in line with actual delivery scenarios.

[0041] Finally, the generated control signals are sent to the corresponding delivery terminals. These control signals are derived from the second delivery scheme determined by the second delivery method and the second delivery path. Control signals refer to electrical or data signals carrying delivery execution instructions, containing core execution information of the second delivery scheme. Specifically, these include delivery method instructions such as temperature control parameters for cold chain delivery and flight modes for UAV delivery; delivery path instructions such as detailed route coordinate sequences and turning node information; and task execution requirements such as loading / unloading sequence, delivery time nodes, and material verification standards. Delivery terminals refer to the equipment or carriers that directly execute delivery tasks, including vehicle-mounted terminals for delivery vehicles, flight control terminals for UAVs, and handheld terminals for delivery personnel. These terminals have signal reception, instruction parsing, and task execution feedback functions. The second delivery scheme refers to the final delivery execution scheme composed of the second delivery method and the second delivery path. It integrates the optimized delivery method and path information and serves as the core basis for guiding the implementation of delivery tasks.

[0042] First, based on the second delivery plan, the core execution information corresponding to each delivery task is extracted. Then, the operation requirements corresponding to the second delivery method and the route data corresponding to the second delivery path are converted into standardized control signal data through signal encoding technology such as JSON (JavaScript Object Notation). Next, wireless communication technologies such as 4G and 5G and IoT communication technologies are used to build a signal transmission channel. A point-to-point communication connection is established between the dispatch center of the material distribution system and each delivery terminal. Subsequently, the dispatch center sends control signals in a combination of batch sending and single-point confirmation. First, the corresponding control signals are sent in batches according to the delivery terminal type and task affiliation. Then, the communication receipt mechanism confirms whether each delivery terminal has successfully received the signal. If the signal is not successfully received, the retransmission mechanism is initiated until the signal is successfully received. After receiving the control signal, the delivery terminal decodes the signal data through the built-in parsing module and converts it into directly executable operation instructions to guide delivery personnel or automated delivery equipment to carry out delivery work.

[0043] This is done to translate the optimized delivery plan into actual executable instructions to ensure the precise implementation of delivery tasks. Standardized signal transmission and terminal execution ensure the consistency and accuracy of the delivery plan, improving delivery efficiency and task completion quality. The integrated technologies and procedures enable efficient transmission and reliable reception of control signals, ensuring that each stage of delivery proceeds in an orderly manner according to the optimized plan.

[0044] One embodiment of the material distribution system of the present invention is an instant retail same-city delivery system. This system is used to undertake the delivery service of goods orders from local supermarkets and convenience stores, with a coverage radius of 3 to 5 kilometers and a delivery time requirement of 1 hour. The following are definitions of terms for this system: Delivery task refers to the delivery request generated after a consumer places an order on the platform, such as the delivery request for snacks, beverages, and daily necessities; Delivery information includes product type, quantity, weight, delivery origin (supermarket / convenience store address), delivery destination (consumer's delivery address latitude and longitude), earliest delivery time, latest delivery time, consumer contact information, and order priority; First delivery method refers to the delivery method initially determined based on order information, such as electric vehicle delivery, personal delivery, and concentrated delivery by small vans; First delivery plan refers to the route planning after clustering according to the initial delivery method, such as the planned roving delivery route after clustering electric vehicle delivery orders in the same business district; Dependency relationship refers to the association between different orders, such as the association of multiple orders placed by the same consumer at the same time requiring combined delivery, or the sequential delivery association of orders from the same batch from the same supermarket; Delivery method impact information refers to the impact of a change in the delivery method of an order on related orders, such as changing an order from electric vehicle delivery to personal delivery potentially leading to earlier delivery times for other orders in the same batch; Geographic information includes road distribution, traffic light locations, business district distribution, residential density, and the number of traffic congestion periods within the delivery area. According to the above; the second delivery method refers to the delivery method after primary optimization, such as consolidating multiple previously scattered electric vehicle delivery orders into a small van for centralized delivery; delivery resources include delivery electric vehicles, small vans, delivery personnel, and delivery stations within the system; resource scarcity refers to the supply and demand of delivery resources in a specific time period and area, such as the ratio of the number of orders allocated to electric vehicle delivery personnel in a business district during the evening peak hours to the total number of available delivery personnel; selected delivery conditions include time constraints of delivery within 1 hour, cost constraints of a single delivery cost not exceeding 20 yuan, and service quality requirements of an on-time delivery rate of over 98%; multi-objective optimization model refers to a model constructed with the objectives of minimizing delivery costs, shortening delivery time, and maximizing delivery personnel utilization; the second delivery route refers to the final delivery route after secondary optimization, such as a patrol delivery route optimized to avoid congested road sections; the second delivery plan refers to the final execution plan after integrating and optimizing the delivery method and delivery route; control signals refer to instruction data including delivery method instructions, such as route navigation information for electric vehicle delivery, and delivery task requirements, such as instruction data for checking the goods checklist; delivery terminals refer to the delivery APP carried by the delivery personnel and the positioning and navigation terminal on the electric vehicle.

[0045] Another embodiment of the present invention constructs a hybrid delivery network coupled with a "physical-information" dual-domain architecture, aiming to support intelligent delivery scenarios involving collaborative operations between trucks and drones. This network comprises three types of nodes: task nodes (representing customer delivery points), connection nodes (supporting truck parking, drone take-off and landing, and cargo handover, serving as the coupling hub between the vehicle network and the drone network), and warehouse nodes (the starting and ending points of delivery tasks). Edges in the network are divided into two categories: vehicle edges connect warehouses and connection points, representing travel time and distance on ground roads; drone edges connect connection points and task points, between task points, or between task points and connection points, characterizing the time, distance, and energy consumption characteristics of drone flights.

[0046] Based on this network, each delivery task can flexibly choose a service mode: it can be delivered directly by truck (if parking is allowed at the task location), or the truck can transport the goods to a pick-up point and then hand them over to a drone for last-mile delivery. This "vehicle-drone collaboration" mechanism significantly improves the flexibility and coverage of the delivery system, and is especially suitable for traffic-congested areas or scenarios with high timeliness requirements.

[0047] To achieve globally optimal scheduling, the problem is modeled as a mixed-integer linear programming (MILP) model, whose objective function comprehensively considers economic costs, environmental impact, service quality, and airspace load. Total costs include truck operating costs, drone flight costs, and time penalty costs. Weight setting employs a dual-track mechanism of "expert judgment + data-driven": first, initial weights for criteria such as transportation cost, timeliness, and operational complexity are obtained through the Analytic Hierarchy Process (AHP); then, combined with a large number of historical delivery plans and their actual comprehensive scores, multiple linear regression is used to fit the relationship between each cost item and the overall performance. The normalized regression coefficients are used as dynamic weights, supporting adaptive adjustments based on scenarios—for example, automatically increasing the time penalty weight during peak promotional periods to prioritize delivery timeliness.

[0048] The model also introduces strict spatiotemporal synchronization constraints: if the drone k executes a task from the docking point p, it must complete the cargo handover with its corresponding truck at point p, meaning that their arrival times must match and their dwell times must overlap. This constraint ensures the physical feasibility of vehicle-machine collaboration.

[0049] To address the high solution complexity and difficulty in real-time application of MILP models, a hierarchical-collaborative solution strategy is proposed. The top layer employs a simplified model or heuristic algorithm to quickly determine the task allocation scheme (i.e., who serves each task and from which connection point), significantly reducing the decision space. The bottom layer, with a fixed allocation scheme, meticulously plans truck routes and UAV flight sequences to accurately satisfy synchronization constraints and calculate actual costs. By feeding back the bottom layer results (such as total cost and synchronization difficulty) to the top layer, an iterative optimization loop is formed, gradually approaching a high-quality feasible solution.

[0050] This method significantly improves computational efficiency while ensuring scheduling rationality, providing a feasible intelligent optimization framework for large-scale, highly dynamic "vehicle-machine collaborative" delivery systems.

[0051] Another embodiment of the present invention provides a path planning system. Specifically, please refer to Figure 3, which shows a schematic diagram of the path planning system in one embodiment of the present invention, including: an acquisition module 11, used to process the acquired delivery information of each delivery task, determine a first delivery method for each delivery task, and determine a first delivery plan based on the first delivery method; a determination module 12, used to determine the delivery method impact information according to the dependency relationship between each delivery task; a first-level module 13, used to perform first-level optimization on the first delivery method based on the delivery method impact information and the geographical information of each delivery task, to obtain a second delivery method corresponding to each delivery task after optimization; a resource module 14, used to determine the resource shortage level based on the delivery resources in the material delivery system corresponding to the second delivery method; a second-level module 15, used to input the second delivery method, resource shortage level and selected delivery conditions into a pre-constructed multi-objective optimization model to perform second-level optimization on the first delivery path in the first delivery plan, to obtain a second delivery path; and a control module 16, used to send the generated control signals to the corresponding delivery terminals, wherein the control signals are obtained at least by the second delivery plan determined by the second delivery method and the second delivery path.

[0052] Preferably, the acquisition module 11 includes: an extraction unit for extracting delivery information from delivery tasks; an analysis unit for analyzing the delivery information and determining the first delivery method for each delivery task; and a clustering unit for clustering delivery tasks of the same type based on the first delivery method, performing preliminary path planning processing based on the clustering results, and determining the first delivery plan.

[0053] Preferably, the determining module 12 includes: a dependency graph unit, used to construct a task dependency graph based on the dependency relationships between various delivery tasks; and a quantification unit, used to perform quantitative analysis on the collaborative impact caused by the change of the first delivery method for each delivery task based on the task dependency graph, and to determine the impact information of the delivery method.

[0054] Preferably, the first-level module 13 includes: an anomaly unit, used to identify anomaly tasks based on delivery method impact information and geographic information of each delivery task; and a correction unit, used to correct the anomaly tasks according to preset conflict resolution rules, so as to obtain the second delivery method corresponding to each delivery task from the correction result.

[0055] Preferably, the resource module 14 includes: a statistics unit, used to calculate the allocated task quantity and available total quantity of delivery resources in the corresponding time and space area of ​​the material delivery system based on the second delivery method; an occupancy unit, used to determine the occupancy ratio of delivery resources in the time and space area based on the comparison result of the allocated task quantity and the available total quantity; and a characterization unit, used to characterize the resource shortage level of delivery resources by the occupancy ratio.

[0056] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0057] Accordingly, embodiments of the present invention provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform steps in the path planning method of the above embodiments, such as steps S1 to S6 as shown in FIG1.

[0058] This invention emphasizes the interrelationships between delivery tasks and the impact of resource scarcity on delivery route planning, addressing the inefficiencies of traditional route planning methods through a multi-stage optimization strategy. First, in the primary optimization stage, this invention considers not only geographical distance but also in-depth analysis of the dependencies between various delivery tasks and their impact on the overall delivery method, thereby enabling fine-tuning of the initial delivery sequence and route. This dynamic adjustment mechanism based on task correlation can more flexibly handle complex situations in actual operation; for example, when two or more delivery points are geographically close but have different demand characteristics, the system can intelligently determine the optimal access order. Second, in the secondary optimization stage, this scheme introduces resource scarcity as an important variable to further optimize delivery routes, ensuring that existing resources are maximized even with limited resources, reducing empty mileage and waiting time. This invention does not only focus on the efficiency of completing individual delivery tasks but also takes a global perspective, comprehensively considering multiple factors such as task interaction, real-time traffic conditions, and resource status to achieve intelligent reorganization and optimization of delivery routes, effectively improving the response speed and service quality of the delivery system, reducing operating costs, and increasing customer satisfaction.

[0059] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A path planning method applied to a material distribution system, characterized in that, include: The delivery information of each delivery task is processed to determine a first delivery method for each delivery task, and a first delivery plan is determined based on the first delivery method; the delivery method impact information is determined according to the dependency relationship between each delivery task. Based on the impact information of the delivery method and the geographical information of each delivery task, the first delivery method is optimized to obtain the second delivery method corresponding to each delivery task after optimization; based on the delivery resources in the material distribution system corresponding to the second delivery method, the resource shortage level is determined. The second delivery method, the resource scarcity level, and the selected delivery conditions are input into the constructed multi-objective optimization model to perform secondary optimization on the first delivery path in the first delivery scheme to obtain the second delivery path; the generated control signals are sent to the corresponding delivery terminals, wherein the control signals are obtained at least by the second delivery scheme determined by the second delivery method and the second delivery path.

2. The path planning method as described in claim 1, characterized in that, The step of processing the delivery information of each delivery task to determine a first delivery method for each delivery task and determining a first delivery plan based on the first delivery method includes: extracting the delivery information from the delivery task; analyzing the delivery information to determine the first delivery method for each delivery task; clustering the delivery tasks of the same type based on the first delivery method; performing preliminary path planning processing based on the clustering results; and determining the first delivery plan.

3. The path planning method as described in claim 1, characterized in that, The step of determining the impact information of delivery methods based on the dependencies between the various delivery tasks includes: constructing a task dependency graph based on the dependencies between the various delivery tasks; and performing quantitative analysis on the collaborative impact caused by the change of the first delivery method for each of the delivery tasks based on the task dependency graph to determine the impact information of delivery methods.

4. The path planning method as described in claim 1, characterized in that, The step of performing a first-level optimization of the first delivery method based on the delivery method impact information and the geographical information of each delivery task to obtain the optimized second delivery method corresponding to each delivery task includes: identifying abnormal tasks based on the delivery method impact information and the geographical information of each delivery task; correcting the abnormal tasks according to preset conflict resolution rules, and obtaining the second delivery method corresponding to each delivery task from the correction result.

5. The path planning method as described in claim 1, characterized in that, The determination of resource scarcity based on the delivery resources in the material delivery system corresponding to the second delivery method includes: based on the second delivery method, calculating the allocated task quantity and the total available quantity of the delivery resources in the material delivery system within the corresponding spatiotemporal region; determining the occupancy ratio of the delivery resources in the spatiotemporal region based on the comparison result of the allocated task quantity and the total available quantity; and using the occupancy ratio to characterize the resource scarcity of the delivery resources.

6. A route planning system, applied in a material distribution system, characterized in that, include: The acquisition module is used to process the acquisition information of each delivery task, determine the first delivery method for each delivery task, and determine the first delivery plan based on the first delivery method. The determination module is used to determine the impact information of the delivery method based on the dependency relationship between each of the delivery tasks; the first-level module is used to perform first-level optimization on the first delivery method based on the impact information of the delivery method and the geographical information of each of the delivery tasks to obtain the optimized second delivery method corresponding to each of the delivery tasks. The resource module is used to determine the resource scarcity level based on the delivery resources in the material delivery system corresponding to the second delivery method; The secondary module is used to input the second delivery method, the resource scarcity level, and the selected delivery conditions into the constructed multi-objective optimization model to perform secondary optimization processing on the first delivery path in the first delivery scheme to obtain the second delivery path; the control module is used to send the generated control signals to the corresponding delivery terminals, wherein the control signals are obtained at least by the second delivery scheme determined by the second delivery method and the second delivery path.

7. The path planning system as described in claim 6, characterized in that, The acquisition module includes: an extraction unit for extracting the delivery information from the delivery task; an analysis unit for analyzing the delivery information and determining the first delivery method for each delivery task; and a clustering unit for clustering the delivery tasks based on the first delivery method, performing preliminary path planning processing based on the clustering results, and determining the first delivery scheme.

8. The path planning system as described in claim 6, characterized in that, The determining module includes: a dependency graph unit, used to construct a task dependency graph based on the dependency relationships between the various delivery tasks; and a quantification unit, used to perform quantitative analysis on the collaborative impact caused by the change of the first delivery method for each delivery task based on the task dependency graph, and determine the impact information of the delivery method.

9. The path planning system as described in claim 6, characterized in that, The first-level module includes: an anomaly unit, used to identify anomaly tasks based on the delivery method impact information and the geographical information of each delivery task; and a correction unit, used to correct the anomaly tasks according to preset conflict resolution rules, so as to obtain the second delivery method corresponding to each delivery task from the correction result.

10. The path planning system as described in claim 6, characterized in that, The resource module includes: a statistics unit, used to calculate the allocated task quantity and available total quantity of the delivery resources in the material delivery system within the corresponding spatiotemporal region based on the second delivery method; an occupancy unit, used to determine the occupancy ratio of the delivery resources in the spatiotemporal region based on the comparison result of the allocated task quantity and the available total quantity; and a characterization unit, used to characterize the resource shortage level of the delivery resources using the occupancy ratio.

Images