Intelligent path planning method and system for logistics drones

Abstract

The present invention provides an intelligent path planning method and system for logistics drones, which relates to the field of data processing technology, including: reading multiple RFID tag information of multiple packages to be delivered; obtaining the standard logistics performance of M idle drones of the same model; performing joint optimization of logistics task allocation to obtain M delivery allocation strategies and M logistics delivery paths; fitting to the delivery verification map to locate M verification tag sequences; during the loading process, performing loading deviation verification through an onboard scanner; during the execution of package delivery, performing path execution deviation verification through M verification tag sequences. The present invention solves the technical problem that the path planning method of the prior art is usually based on fixed maps or preset rules, and it is difficult to dynamically adapt to multiple packages, complex paths and real-time delivery needs, resulting in insufficient efficiency and reliability of logistics delivery.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an intelligent path planning method and system for logistics drones. Background Art

[0002] With the rapid development of e-commerce and express delivery industries, traditional manual delivery and ground transportation systems are gradually facing problems such as low delivery efficiency, waste of resources and high labor costs. As an innovative delivery method, drones can provide more efficient, flexible and intelligent solutions, especially when dealing with remote areas, urban traffic congestion areas and peak delivery.

[0003] In drone delivery, path planning is the key to achieving efficient delivery. However, the flight path of a drone is limited by many factors, such as flight distance, flight time, and energy consumption. Current path planning mostly relies on preset routes or is based on simple rules, and lacks dynamic optimization of actual delivery situations. Especially for situations with multiple packages and multiple drones, it is difficult to dynamically adjust the flight path in real time according to changes in delivery demand, resulting in low delivery efficiency and insufficient reliability. Summary of the invention

[0004] This application provides an intelligent path planning method and system for logistics drones, aiming to solve the technical problem that the path planning methods in the prior art are usually based on fixed maps or preset rules, and are difficult to dynamically adapt to multiple packages, complex routes and real-time delivery needs, resulting in insufficient efficiency and reliability of logistics delivery.

[0005] The first aspect disclosed in the present application provides an intelligent path planning method for logistics drones, the method comprising: reading multiple RFID tag information of multiple packages to be delivered through a local RFID scanner; interacting with a logistics management cloud to obtain standard logistics performance of M idle drones of the same model; performing joint optimization of logistics task allocation based on the standard logistics performance and multiple RFID tag information to obtain M delivery allocation strategies and M logistics delivery paths; fitting the M logistics delivery paths to a delivery verification map to locate M verification tag sequences; in the process of mapping the multiple packages to be delivered to the M idle drones according to the M delivery allocation strategies, performing loading deviation verification through M airborne scanners pre-configured on the M idle drones; after mapping the multiple packages to be delivered to the M idle drones according to the M delivery allocation strategies, the M idle drones perform path execution deviation verification through the M verification tag sequences when executing package delivery according to the M logistics delivery paths.

[0006] The second aspect disclosed in the present application provides an intelligent path planning system for logistics drones, the system is used for the above-mentioned intelligent path planning method for logistics drones, and the system includes: a tag information reading module, which is used to read multiple RFID tag information of multiple packages to be delivered through a local RFID scanner; a standard logistics performance acquisition module, which is used to interact with the logistics management cloud to obtain the standard logistics performance of M idle drones of the same model; a joint optimization module, which is used to perform joint optimization of logistics task allocation based on the standard logistics performance and multiple RFID tag information, and obtain M delivery allocation strategies and M logistics delivery paths; a tag sequence verification positioning module , used to fit the M logistics delivery paths to the delivery verification map to locate M verification label sequences; a loading deviation verification module, used to perform loading deviation verification through M airborne scanners pre-configured by the M idle drones during the process of mapping the multiple packages to be delivered to the M idle drones according to the M delivery allocation strategies; a path execution deviation verification module, used to perform path execution deviation verification through the M verification label sequences during the process of the M idle drones executing package delivery according to the M logistics delivery paths after mapping the multiple packages to be delivered to the M idle drones according to the M delivery allocation strategies.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] Through RFID scanning technology, the basic information of the packages to be delivered can be collected quickly and in batches. The accurately collected RFID tag information provides key parameters for drone path planning. By interacting with the logistics management cloud, the standard logistics performance parameters of idle drones can be obtained in real time. By matching the actual performance of the drone with the delivery task requirements, the task allocation of each drone is ensured to be reasonable. Through the joint optimization algorithm, the package information and drone performance are comprehensively analyzed to generate efficient delivery distribution strategies and logistics delivery paths. Reasonable path planning reduces unnecessary flight distances, reduces energy consumption, and improves the delivery efficiency of drones. The logistics delivery path is fitted to the delivery verification map positioning verification tag sequence to ensure that the drone is in the correct position during flight. The delivery process always follows the predetermined path to reduce the risk of path deviation. The setting of verification tags along the way provides a real-time verification mechanism for path execution, further ensuring the accurate execution of delivery tasks. During the loading process, the onboard RFID scanner is used to verify whether the packages loaded by each drone comply with the allocation strategy to avoid delivery failures due to loading errors. The automated loading verification process reduces the need for manual inspection and improves the loading efficiency and accuracy of drones. During the package delivery process, the path execution deviation verification is performed by verifying the tag sequence. At each delivery point, the delivery point information and package information are verified by RFID to ensure that the package can be accurately delivered to the target location. Combined with the path execution deviation verification, high-precision logistics distribution services can be achieved.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic flow chart of an intelligent path planning method for a logistics drone provided in an embodiment of the present application.

[0011] Figure 2 A schematic diagram of the structure of an intelligent path planning system for a logistics drone provided in an embodiment of the present application.

[0012] Explanation of the reference numerals: label information reading module 10 , standard logistics performance acquisition module 20 , joint optimization module 30 , label sequence verification and positioning module 40 , loading deviation verification module 50 , path execution deviation verification module 60 . DETAILED DESCRIPTION

[0013] The embodiments of the present application provide an intelligent path planning method and system for logistics drones, thereby solving the technical problem that the path planning methods in the prior art are usually based on fixed maps or preset rules, and are difficult to dynamically adapt to multiple packages, complex routes and real-time delivery needs, resulting in insufficient efficiency and reliability of logistics delivery.

[0014] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced below in conjunction with the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0015] Embodiment 1, as Figure 1 As shown, the embodiment of the present application provides an intelligent path planning method for a logistics drone, the method comprising:

[0016] The local RFID scanner is used to read the multiple RFID tag information of the multiple packages to be delivered.

[0017] Get multiple parcels to be delivered, each of which is pre-installed with RFID tags, which contain key information about the parcels, such as the parcel's unique identification code (ID), parcel weight, size, target delivery address, delivery time limit, etc. Use a local RFID scanner to read the RFID tags on all the parcels to be delivered. When the parcel passes through the scanner, the scanner automatically identifies the content of the RFID tag on the parcel and obtains multiple RFID tag information. In this way, it is possible to ensure that detailed information of each parcel is available at any time during the delivery process, and this information is used for subsequent route planning and task allocation.

[0018] Interactive logistics management cloud to obtain the standard logistics performance of M idle drones of the same model.

[0019] Interact with the logistics management cloud, which includes a real-time database for managing the information of all drones, especially the status of idle drones. By interactively querying all drones of the same model, M drones that are currently idle and can perform tasks are obtained. M is the number of idle drones, which is determined according to the actual drone situation. Idle drones refer to those that are not occupied by other tasks and can be put into tasks immediately. The standard logistics performance of the M idle drones is obtained, including key information such as the load capacity, flight speed, maximum flight distance and energy budget of each idle drone, providing data support for subsequent path planning, task allocation and flight execution.

[0020] The logistics task allocation is jointly optimized according to the standard logistics performance and the information of multiple RFID tags to obtain M delivery allocation strategies and M logistics delivery paths.

[0021] According to the standard logistics performance and multiple RFID tag information, with the load capacity of the drone as a constraint, the packages are arranged and combined according to the weight information to generate multiple package combination plans. Based on the package combination plan and the target address, the initial flight path is planned for each combination. The generated initial strategy (including the assigned package combination and flight path) is evaluated for logistics fitness. The fitness evaluation indicators include delivery time limit deviation, flight time and distance, flight energy consumption, and load balance. Optimization algorithms, such as genetic algorithms or particle swarm optimization algorithms, are used to maximize logistics fitness. The optimal task allocation strategy and path planning are selected, and finally M delivery allocation strategies and M logistics delivery paths corresponding to M drones are generated.

[0022] The M logistics delivery paths are fitted to the delivery verification map to locate M verification tag sequences.

[0023] Retrieve map information of the delivery area, including the geographic location of each delivery point and the distribution of surrounding verification tags. Verification tags refer to RFID tags distributed along the delivery path, which are used for path deviation verification. Fit each logistics delivery path to the delivery verification map to ensure that the path passes through the designated verification point. The constraints of the path fitting include flight distance, time and performance limitations of the drone. According to the location of the delivery path, match the verification tag sequence along the way. Each path needs to correspond to a set of verification tags for one-by-one verification during the flight. Generate a set of verification tag sequences for each logistics delivery path, and obtain M verification tag sequences. These verification tag sequences are used for path deviation verification and delivery point location confirmation during the actual flight.

[0024] In the process of mapping and loading the multiple packages to be delivered to the M idle drones according to the M delivery allocation strategies, loading deviation verification is performed through M airborne scanners pre-configured on the M idle drones.

[0025] According to the generated delivery allocation strategy, a specific set of packages to be delivered is assigned to each idle drone. Each drone is equipped with an onboard scanner that can scan the packages to be loaded in real time and identify their RFID tag information. During the loading process, each package is moved to the drone loading area by an operator or a conveyor belt. The drone's onboard scanner automatically scans the package's RFID tag and identifies the package information. The scan content includes the package's unique identification code (ID), weight, size, etc. The package information identified by the scanner is compared one by one with the package list in the allocation strategy to ensure that the package information complies with the policy content. If the scanned package is not in the allocation list, an alarm will be issued and the loading of the package will be prevented. By verifying each package during the loading process, it is ensured that the package matches the allocation strategy.

[0026] After the multiple packages to be delivered are mapped and loaded onto the M idle drones according to the M delivery allocation strategies, the M idle drones perform path execution deviation verification through the M verification tag sequences when executing package delivery according to the M logistics delivery paths.

[0027] Before taking off, each idle drone loads the planned logistics delivery path into the flight control system, including the verification tag sequence and its geographic coordinates along the way. The drone uses the built-in navigation system, such as GPS, inertial navigation and map data, to fly along the planned path. During the package delivery process, the drone uses the onboard scanner to identify the RFID verification tags of the verification points along the way, and compares the scanning time and geographic location of each verification tag with the verification tag sequence in the planned path. Specifically, the sequence and order of the verification tags in the actual path and the planned path are compared. If the drone does not scan the verification tag in sequence or deviates from the verification tag position, it is considered a path deviation. If a deviation is detected, the drone control system will automatically correct the flight path, or send an alarm to the cloud and request to re-plan the path. Through the deviation verification mechanism, the drone can adapt to dynamic changes and complete efficient delivery.

[0028] Furthermore, the logistics task allocation is jointly optimized according to the standard logistics performance and the information of multiple RFID tags to obtain M delivery allocation strategies and M logistics delivery paths. The method includes:

[0029] A plurality of logistics-related information is obtained by calling from the plurality of RFID tag information, wherein the logistics-related information includes package quality, target address and delivery time limit; a load capacity, a flight speed scale and an energy budget scale are obtained from the standard logistics performance; the load capacity is used as a package combination constraint, and the permutation and combination results of the plurality of packages to be delivered are judged according to the plurality of package qualities extracted from the plurality of logistics-related information, so as to obtain N groups of initial allocation strategies; the logistics task allocation is jointly optimized for the N groups of initial allocation strategies according to the standard logistics performance and the plurality of logistics-related information, so as to obtain N logistics planning fitnesses and N groups of logistics delivery paths; the M delivery allocation strategies and the M logistics delivery paths are located from the N groups of initial allocation strategies and the N groups of logistics delivery paths by serializing the N logistics planning fitnesses.

[0030] By scanning the RFID tag of each package, the logistics-related information related to the package is extracted, including the package quality, target address and delivery time limit. The package quality is weight and volume, the target address is geographic coordinates or specific distribution point, and the delivery time limit is the time window or deadline for delivery.

[0031] The standard logistics performance of currently idle drones is called from the logistics management cloud, including load capacity, flight speed scale and energy budget scale, where the load capacity is the maximum load weight of each drone, the flight speed scale is the normal flight speed range of the drone, such as 10-15m / s, and the energy budget scale is the budget value of battery capacity or flight energy consumption, indicating the maximum energy consumption that the drone battery can support. Furthermore, the performance of each drone is parameterized to generate standardized performance data to ensure comparability between the performance data and the package logistics information, such as consistent weight units.

[0032] According to the load capacity of the drone, load constraints are imposed on the permutations and combinations of packages, that is, the total weight of packages carried by a single drone is less than or equal to the load capacity of the drone. According to the logistics-related information (package quality, target address and delivery time limit), the packages to be delivered are permuted and combined to generate all possible allocation plans. Each set of plans ensures that the packages will not be allocated repeatedly and the total weight meets the load constraints. The package combinations are allocated to M drones and N groups of initial allocation strategies are generated. Each group of strategies defines the package list of each drone and the estimated load of each drone. In addition, the package allocation of each group of strategies does not overlap, ensuring that all packages to be delivered are reasonably allocated.

[0033] Through the optimization algorithm, efficient allocation strategies and logistics delivery paths are selected from N groups of initial allocation strategies, and N logistics planning fitness and N groups of optimized logistics delivery paths are generated. The optimization goal is to maximize the overall efficiency of logistics task allocation. Specific indicators include minimizing delivery time deviation, minimizing flight time, minimizing energy consumption, optimizing load balance, etc. The specific optimization process is elaborated in detail in subsequent steps and will not be repeated here.

[0034] Sort the fitness of N logistics plans from high to low according to the fitness value to ensure that the optimal strategy is selected first. According to the serialization results, the strategy with the highest fitness is selected as the final delivery allocation plan to obtain M delivery allocation strategies. For each group of selected allocation strategies, the corresponding logistics delivery path is extracted to obtain M logistics delivery paths. The whole process ensures the efficiency of the delivery path, the rationality of package distribution and the optimal utilization of drone performance.

[0035] Furthermore, the N groups of initial allocation strategies are jointly optimized for logistics task allocation according to the standard logistics performance and multiple logistics association information to obtain N logistics planning fitness and N groups of logistics delivery paths. The method also includes:

[0036] The first group of initial allocation strategies is disassembled to obtain M initial allocation strategies of the M idle drones; the multiple logistics association information are reorganized according to the M initial allocation strategies to obtain M groups of logistics association information; flight path optimization is performed according to the M groups of logistics association information and standard logistics performance to obtain M first flight paths, wherein the first path identifier of the first flight path comprises a first delivery time limit deviation, a first flight time, a first flight distance and a first flight energy consumption; M groups of package qualities are extracted from the M groups of logistics association information, and load balancing evaluation is performed based on the M groups of package qualities to obtain M load balancing coefficients; the M first path identifiers and the M load balancing coefficients are loaded into a task allocation evaluation function to perform task allocation fitness evaluation to obtain a first logistics planning fitness; and by analogy, the N groups of initial allocation strategies are evaluated for task allocation fitness to obtain the N logistics planning fitness.

[0037] An analysis object is randomly selected from the N groups of initial allocation strategies as the first group of initial allocation strategies. This random selection ensures the traversal analysis of all N groups of initial allocation strategies. The first group of initial allocation strategies contains the allocation and path planning results of all packages to be delivered. According to the drone ID, the first group of initial allocation strategies is disassembled into M initial allocation strategies, and each sub-strategy corresponds to a specific task of an idle drone.

[0038] For each initial allocation strategy, the logistics information related to the allocated packages is extracted. The information of each package includes the quality, target address, and delivery time limit. According to the mission of the drone, the package information allocated to each drone is recombined into a complete logistics-related information to obtain M groups of logistics-related information. The logistics-related information includes the quality, target address, and delivery time limit of all packages allocated to the corresponding drone.

[0039] The flight path is optimized by using logistics information and drone performance data. The constraints include that the flight distance cannot exceed the maximum flight distance of the drone, the delivery order must meet the time limit requirements of the package, and the path is within the energy consumption range of the drone. For multiple delivery points, the flight path is optimized to obtain the path with the shortest total flight distance, and M first flight paths are obtained. The specific flight path optimization process is detailed in subsequent steps and will not be repeated here. The first path identification of the first flight path includes the first delivery time limit deviation, the first flight time, the first flight distance and the first flight energy consumption. The first delivery time limit deviation is the deviation between the actual delivery time of each package and its time limit. The first flight time is the estimated flight time required for the drone to execute the entire path. The first flight distance is the total flight distance for the drone to complete the path. The first flight energy consumption is the energy consumption of the drone predicted based on the path length and the package weight.

[0040] From the logistics-related information of each drone, the package masses of all packages assigned to the drone are extracted, and the package masses are summed up to obtain the actual load of each drone. The load balance evaluation is performed using the actual load and the load capacity of the drone, that is, the ratio of the actual load of the drone to the load capacity is calculated. The load balance coefficient of the corresponding drone is obtained after calculation, and the ideal value of the load balance coefficient is close to 100%.

[0041] Substitute the M first path identifiers and the M load balancing coefficients into the task allocation evaluation function, calculate the fitness of each initial allocation strategy, obtain the M sample planning fitness, and perform mean calculation to obtain the first logistics planning fitness. The M first flight paths are stored as the first group of logistics delivery paths in association with the first logistics planning fitness.

[0042] By analogy, all N groups of initial allocation strategies are traversed, and the above process is repeated to obtain the N logistics planning fitness.

[0043] Further, performing flight path optimization according to the M groups of logistics association information and standard logistics performance to obtain M first flight paths, the method further includes:

[0044] The first initial allocation strategy of the first idle drone is called from the M initial allocation strategies, wherein the first initial allocation strategy includes K packages to be delivered; K package masses, K target addresses and K delivery time limits of the K packages to be delivered are extracted from the first group of logistics-related information; with the flight distance as a constraint, flight trajectory fitting is performed according to the K target addresses to obtain H sample flight paths; with the flight speed scale as a constraint, delivery simulation is performed on the H sample flight paths to obtain H sample time parameters, H sample distance parameters and H groups of delivery time information, wherein each group of delivery time information includes K simulated delivery times; the Pearson correlation coefficient is used to calculate the H sample time limit deviations of the H groups of delivery time information and the K delivery time limits; the K package masses and the flight speed scale are used as energy consumption analysis Constraints are set, load energy consumption is predicted during the delivery simulation of the H sample flight paths to obtain H sample energy consumption parameters; the H sample time parameters, H sample distance parameters, H sample time limit deviations and H sample energy consumption parameters are weightedly solved using a preset weight configuration to obtain H sample path reliabilities; the H sample path reliabilities are serialized, and a first flight path is called from the H sample flight paths based on a sorting extreme value; a first delivery time limit deviation, a first flight time, a first flight distance and a first flight energy consumption of the first flight path are obtained from the H sample time parameters, H sample distance parameters, H sample time limit deviations and H sample energy consumption parameters as a first path identifier; and so on, flight path optimization is performed according to the M groups of logistics association information and standard logistics performance to obtain M first flight paths.

[0045] An analysis object is randomly selected from the M initial allocation strategies as the first initial allocation strategy. This random selection ensures the subsequent traversal analysis of all strategies. The first initial allocation strategy corresponds to the first idle drone. The first initial allocation strategy includes a package list of K packages to be delivered. The package list content includes a unique identification code (ID) for each package.

[0046] In the first set of logistics-related information, the logistics information corresponding to the unique identification codes (IDs) of the K packages to be delivered is searched, and the K package qualities, K target addresses, and K delivery time limits are obtained.

[0047] Taking the flight distance as a constraint, ensure that the total flight distance of the UAV cannot exceed its maximum flight range. The maximum flight range is affected by the battery capacity and load. The flight trajectory is fitted according to the K target addresses to generate H sample flight paths that meet the constraints.

[0048] The flight speed scale of the drone is used as a constraint, such as 10–15 m / s. During the simulation, it is assumed that the drone performs the task at an average flight speed (such as 12 m / s). For each sample flight path, the total flight distance (sample distance parameter) is obtained. The total flight distance is obtained by adding up each segment. The total flight time (sample time parameter) is calculated based on the total flight distance and the average flight speed. The total flight time is equal to the total flight distance divided by the average flight speed. For each package in the path, its specific delivery time is simulated. Assuming that each delivery operation requires a fixed time, such as 30 seconds, for the K packages in the path, the respective simulated delivery time is recorded to obtain K simulated delivery times. For each sample flight path, the total flight time is added to the simulated delivery time of each point to obtain the total time of the path as the delivery time information.

[0049] The time limit deviation is defined as the difference between the simulated delivery time and the target delivery time limit. For each sample flight path, the Pearson correlation coefficient of the delivery time information of the path is calculated with the delivery time limit. The obtained correlation coefficient range is [-1,1]. The closer it is to 1, the higher the match between the path and the target delivery time limit. The calculated correlation coefficient is used as the time limit deviation of the path. The corresponding time limit deviation is calculated for each path to obtain H sample time limit deviations.

[0050] The energy consumption of a drone is mainly determined by the total flight distance, flight speed, total load and the fixed energy consumption each time a package is delivered. For each sample path, the total load of the drone is calculated. The total load is the sum of the masses of the K packages carried. The flight energy consumption is predicted, which is proportional to the flight distance, total load and flight speed. The delivery energy consumption is predicted, which is proportional to the number of delivery points. Each time a package is delivered, a fixed energy consumption is consumed. The flight energy consumption is added to the delivery energy consumption to obtain the total energy consumption, which is the sample energy consumption parameter of the corresponding path. The corresponding total energy consumption is calculated for each path to obtain H sample energy consumption parameters.

[0051] The H sample time parameters, H sample distance parameters, H sample time limit deviations and H sample energy consumption parameters are normalized to eliminate the influence of different parameter dimensions, and a weight is set for each indicator to reflect its importance in the path reliability. The weight is set according to specific needs. For example, the time parameter is set to 30%, the distance parameter is set to 20%, the time limit deviation is set to 40%, and the energy consumption parameter is set to 10%. Through weighted calculation, the time, distance, time limit deviation and energy consumption of each path are comprehensively evaluated to obtain the reliability of the H sample paths. The higher the reliability, the better the corresponding path.

[0052] The reliability of the H sample paths is sorted in descending order to form a serialization result, and the path with the highest reliability is selected from the serialization result as the first flight path.

[0053] Extract the sample time parameters of the first flight path from the H sample time parameters to obtain the first flight time; extract the sample distance parameters of the first flight path from the H sample distance parameters to obtain the first flight distance; extract the sample time limit deviation of the first flight path from the H sample time limit deviations to obtain the first delivery time limit deviation; extract the sample energy consumption parameters of the first flight path from the H sample energy consumption parameters to obtain the first flight energy consumption. Summarize the first delivery time limit deviation, the first flight time, the first flight distance and the first flight energy consumption to form the first path identification, which provides a basis for subsequent optimization and execution.

[0054] Similarly, based on the flight path optimization of a completed drone delivery mission, the same logic and methods are applied to optimize the flight paths of the remaining drone delivery missions, and finally the optimal flight paths of M drones are generated, and M first flight paths are obtained. Each path is evaluated for reliability, providing a clear plan for drones to perform logistics tasks.

[0055] Furthermore, taking the K package masses and the flight speed scale as energy consumption analysis constraints, performing load energy consumption prediction during the delivery simulation of the H sample flight paths to obtain H sample energy consumption parameters, the method further includes:

[0056] Compare the H sample energy consumption parameters with the energy budget scale to obtain H energy consumption deviation vectors; and eliminate the paths that do not meet the energy budget from the H sample flight paths according to the H energy consumption deviation vectors.

[0057] The energy budget scale is the budget value of battery capacity or flight energy consumption, which indicates the maximum energy consumption that the drone battery can support. For each sample path, its energy consumption deviation is calculated, that is, the difference between the sample energy consumption parameter and the energy budget scale, and the energy consumption deviation vector is obtained. If the difference is less than or equal to 0, it means that the energy consumption is within the budget; if the difference is greater than 0, it means that the energy consumption of the path exceeds the budget.

[0058] According to the energy consumption deviation vector, all paths with deviations greater than 0, that is, paths with energy consumption exceeding the budget, are eliminated to ensure that all remaining paths meet the energy consumption constraints of the drone. This process ensures the feasibility of the flight path and avoids energy shortage when the drone is performing tasks.

[0059] Furthermore, the task allocation evaluation function is constructed, and the task allocation evaluation function is as follows:

[0060] ;

[0061] Among them, A is the fitness of logistics planning, r is the first delivery time limit deviation, T is the first flight time, D is the first flight distance, E is the first flight energy consumption, L fis the load balancing factor, L c For load-bearing capacity.

[0062] Specifically, the task allocation evaluation function is as follows:

[0063] ;

[0064] The task allocation evaluation function is used to comprehensively consider multiple indicators such as delivery time, flight efficiency, energy consumption, load balance, etc., integrate these factors together through mathematical expression, and provide a scientific fitness evaluation for each task allocation plan, so as to evaluate the task planning fitness A of the UAV. By combining and calculating multiple key logistics indicators, the efficiency and rationality of the task allocation plan are measured. The higher the final calculated fitness value A, the better the allocation plan.

[0065] Furthermore, the M first path identifiers and the M load balancing coefficients are loaded into the task allocation evaluation function to perform task allocation fitness evaluation to obtain the first logistics planning fitness. The method further includes:

[0066] The M first path identifiers and M load balancing coefficients are loaded into the task allocation evaluation function to perform task allocation fitness evaluation to obtain M sample planning fitnesses; the mean of the M sample planning fitnesses is calculated to obtain the first logistics planning fitnesses; the M first flight paths are stored as a first group of logistics delivery paths in association with the first logistics planning fitnesses.

[0067] The M first path identifiers and M load balancing coefficients are loaded into the task allocation evaluation function. The task allocation evaluation function is used to calculate the fitness of the task allocation by integrating multiple indicators. After calculation, M sample planning fitnesses are obtained. The mean of the M sample planning fitnesses is calculated to evaluate the overall fitness of the entire logistics delivery plan, and the first logistics planning fitness is calculated. The M first flight paths are stored as the first group of logistics delivery paths in association with the first logistics planning fitness, providing data support for the execution and subsequent optimization of logistics tasks.

[0068] In summary, the intelligent path planning method for logistics drones provided in the embodiments of the present application has the following technical effects:

[0069] Through RFID scanning technology, the basic information of the packages to be delivered can be collected quickly and in batches. The accurately collected RFID tag information provides key parameters for drone path planning. By interacting with the logistics management cloud, the standard logistics performance parameters of idle drones can be obtained in real time. By matching the actual performance of the drone with the delivery task requirements, the task allocation of each drone is ensured to be reasonable. Through the joint optimization algorithm, the package information and drone performance are comprehensively analyzed to generate efficient delivery distribution strategies and logistics delivery paths. Reasonable path planning reduces unnecessary flight distances, reduces energy consumption, and improves the delivery efficiency of drones. The logistics delivery path is fitted to the delivery verification map positioning verification tag sequence to ensure that the drone is in the correct position during flight. The delivery process always follows the predetermined path to reduce the risk of path deviation. The setting of verification tags along the way provides a real-time verification mechanism for path execution, further ensuring the accurate execution of delivery tasks. During the loading process, the onboard RFID scanner is used to verify whether the packages loaded by each drone comply with the allocation strategy to avoid delivery failures due to loading errors. The automated loading verification process reduces the need for manual inspection and improves the loading efficiency and accuracy of drones. During the package delivery process, the path execution deviation verification is performed by verifying the tag sequence. At each delivery point, the delivery point information and package information are verified by RFID to ensure that the package can be accurately delivered to the target location. Combined with the path execution deviation verification, high-precision logistics distribution services can be achieved.

[0070] Embodiment 2 is based on the same inventive concept as the intelligent path planning method for logistics drones in the previous embodiment. Figure 2 As shown, the embodiment of the present application provides an intelligent path planning system for a logistics drone, the system comprising:

[0071] The tag information reading module 10 is used to read multiple RFID tag information of multiple packages to be delivered through a local RFID scanner; the standard logistics performance acquisition module 20 is used to interact with the logistics management cloud to obtain the standard logistics performance of M idle drones of the same model; the joint optimization module 30 is used to perform joint optimization of logistics task allocation based on the standard logistics performance and multiple RFID tag information to obtain M delivery allocation strategies and M logistics delivery paths; the verification tag sequence positioning module 40 is used to fit the M logistics delivery paths to the delivery verification map to locate M verification tag sequences; the loading deviation verification module 50 is used to perform loading deviation verification through M airborne scanners pre-configured on the M idle drones during the process of mapping the multiple packages to be delivered to the M idle drones according to the M delivery allocation strategies; the path execution deviation verification module 60 is used to perform path execution deviation verification through the M verification tag sequences during the process of the M idle drones executing package delivery according to the M logistics delivery paths after mapping the multiple packages to be delivered to the M idle drones according to the M delivery allocation strategies.

[0072] Furthermore, the system also includes a logistics delivery path positioning module to perform the following operation steps:

[0073] A plurality of logistics-related information is obtained by calling from the plurality of RFID tag information, wherein the logistics-related information includes package quality, target address and delivery time limit; a load capacity, a flight speed scale and an energy budget scale are obtained from the standard logistics performance; the load capacity is used as a package combination constraint, and the permutation and combination results of the plurality of packages to be delivered are judged according to the plurality of package qualities extracted from the plurality of logistics-related information, so as to obtain N groups of initial allocation strategies; the logistics task allocation is jointly optimized for the N groups of initial allocation strategies according to the standard logistics performance and the plurality of logistics-related information, so as to obtain N logistics planning fitnesses and N groups of logistics delivery paths; the M delivery allocation strategies and the M logistics delivery paths are located from the N groups of initial allocation strategies and the N groups of logistics delivery paths by serializing the N logistics planning fitnesses.

[0074] Furthermore, the system also includes a logistics planning fitness acquisition module to perform the following operation steps:

[0075] The first group of initial allocation strategies is disassembled to obtain M initial allocation strategies of the M idle drones; the multiple logistics association information are reorganized according to the M initial allocation strategies to obtain M groups of logistics association information; flight path optimization is performed according to the M groups of logistics association information and standard logistics performance to obtain M first flight paths, wherein the first path identifier of the first flight path comprises a first delivery time limit deviation, a first flight time, a first flight distance and a first flight energy consumption; M groups of package qualities are extracted from the M groups of logistics association information, and load balancing evaluation is performed based on the M groups of package qualities to obtain M load balancing coefficients; the M first path identifiers and the M load balancing coefficients are loaded into a task allocation evaluation function to perform task allocation fitness evaluation to obtain a first logistics planning fitness; and by analogy, the N groups of initial allocation strategies are evaluated for task allocation fitness to obtain the N logistics planning fitness.

[0076] Furthermore, the system further includes a first flight path acquisition module to perform the following operation steps:

[0077] The first initial allocation strategy of the first idle drone is called from the M initial allocation strategies, wherein the first initial allocation strategy includes K packages to be delivered; K package masses, K target addresses and K delivery time limits of the K packages to be delivered are extracted from the first group of logistics-related information; with the flight distance as a constraint, flight trajectory fitting is performed according to the K target addresses to obtain H sample flight paths; with the flight speed scale as a constraint, delivery simulation is performed on the H sample flight paths to obtain H sample time parameters, H sample distance parameters and H groups of delivery time information, wherein each group of delivery time information includes K simulated delivery times; the Pearson correlation coefficient is used to calculate the H sample time limit deviations of the H groups of delivery time information and the K delivery time limits; the K package masses and the flight speed scale are used as energy consumption analysis Constraints are set, load energy consumption is predicted during the delivery simulation of the H sample flight paths to obtain H sample energy consumption parameters; the H sample time parameters, H sample distance parameters, H sample time limit deviations and H sample energy consumption parameters are weightedly solved using a preset weight configuration to obtain H sample path reliabilities; the H sample path reliabilities are serialized, and a first flight path is called from the H sample flight paths based on a sorting extreme value; a first delivery time limit deviation, a first flight time, a first flight distance and a first flight energy consumption of the first flight path are obtained from the H sample time parameters, H sample distance parameters, H sample time limit deviations and H sample energy consumption parameters as a first path identifier; and so on, flight path optimization is performed according to the M groups of logistics association information and standard logistics performance to obtain M first flight paths.

[0078] Furthermore, the system further includes a path elimination module to perform the following operation steps:

[0079] Compare the H sample energy consumption parameters with the energy budget scale to obtain H energy consumption deviation vectors; and eliminate the paths that do not meet the energy budget from the H sample flight paths according to the H energy consumption deviation vectors.

[0080] Furthermore, the system also includes an evaluation function building module to perform the following operation steps:

[0081] Construct the task allocation evaluation function, which is as follows:

[0082] ;

[0083] Among them, A is the fitness of logistics planning, r is the first delivery time limit deviation, T is the first flight time, D is the first flight distance, E is the first flight energy consumption, L f is the load balancing factor, L c For load-bearing capacity.

[0084] Furthermore, the system further includes an associated storage module to perform the following operation steps:

[0085] The M first path identifiers and M load balancing coefficients are loaded into the task allocation evaluation function to perform task allocation fitness evaluation to obtain M sample planning fitnesses; the mean of the M sample planning fitnesses is calculated to obtain the first logistics planning fitnesses; the M first flight paths are stored as a first group of logistics delivery paths in association with the first logistics planning fitnesses.

[0086] Through the above detailed description of the intelligent path planning method for logistics drones in this specification, those skilled in the art can clearly understand the intelligent path planning system for logistics drones in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0087] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent path planning method for logistics drones, characterized in that: The method comprises: Reading multiple RFID tag information of multiple packages to be delivered through a local RFID scanner; Interactive logistics management cloud to obtain the standard logistics performance of M idle drones of the same model; Jointly optimize the logistics task allocation according to the standard logistics performance and multiple RFID tag information to obtain M delivery allocation strategies and M logistics delivery paths; Fitting the M logistics delivery paths to the delivery validation map to locate M verification tag sequences; In the process of mapping and loading the multiple packages to be delivered to the M idle drones according to the M delivery allocation strategies, performing loading deviation verification through M airborne scanners pre-configured on the M idle drones; After mapping and loading the multiple packages to be delivered to the M idle drones according to the M delivery allocation strategies, the M idle drones perform path execution deviation verification through the M verification tag sequences when executing package delivery according to the M logistics delivery paths; Joint optimization of logistics task allocation is performed according to the plurality of RFID tag information and M logistics performance information to obtain M delivery allocation strategies and M logistics delivery paths, the method comprising: Retrieving and obtaining a plurality of logistics-related information from the plurality of RFID tag information, wherein the logistics-related information includes package quality, target address, and delivery time limit; Extracting the load capacity, flight speed scale and energy budget scale from the standard logistics performance; Taking the load capacity as the package combination constraint, judging the permutation and combination results of the multiple packages to be delivered according to the multiple package masses extracted from the multiple logistics related information, and obtaining N groups of initial allocation strategies; Jointly optimizing the logistics task allocation of the N groups of initial allocation strategies according to the standard logistics performance and multiple logistics related information to obtain N logistics planning fitness and N groups of logistics delivery paths; By serializing the N logistics planning fitnesses, the M delivery allocation strategies and the M logistics delivery paths are located from the N groups of initial allocation strategies and the N groups of logistics delivery paths; The N groups of initial allocation strategies are jointly optimized for logistics task allocation according to the standard logistics performance and multiple logistics association information to obtain N logistics planning fitness and N groups of logistics delivery paths, and the method further includes: Disassembling the first group of initial allocation strategies to obtain M initial allocation strategies for the M idle drones; Reorganize the plurality of logistics association information according to the M initial allocation strategies to obtain M groups of logistics association information; Perform flight path optimization according to the M groups of logistics association information and standard logistics performance to obtain M first flight paths, wherein the first path identifier of the first flight path includes a first delivery time limit deviation, a first flight time, a first flight distance, and a first flight energy consumption; Extracting M groups of package qualities from the M groups of logistics-related information, and performing load balancing evaluation based on the M groups of package qualities to obtain M load balancing coefficients; Loading the M first path identifiers and the M load balancing coefficients into the task allocation evaluation function to perform task allocation fitness evaluation to obtain the first logistics planning fitness; Similarly, the task allocation fitness of the N groups of initial allocation strategies is evaluated to obtain the fitness of the N logistics plans; Construct the task allocation evaluation function, which is as follows: Among them, A is the logistics planning fitness, r is the first delivery time limit deviation, T is the first flight time, D is the first flight distance, E is the first flight energy consumption, Lf is the load balance coefficient, and Lc is the load capacity.

2. The intelligent path planning method for a logistics drone according to claim 1, characterized in that: Optimizing the flight path according to the M groups of logistics association information and standard logistics performance to obtain M first flight paths, the method further includes: Calling a first initial allocation strategy of a first idle drone from the M initial allocation strategies, wherein the first initial allocation strategy includes K packages to be delivered; Extracting K package masses, K target addresses and K delivery time limits of the K packages to be delivered from the first group of logistics-related information; Taking the flight distance as a constraint, performing flight trajectory fitting according to the K target addresses to obtain H sample flight paths; Taking the flight speed scale as a constraint, performing a delivery simulation on the H sample flight paths, obtaining H sample time parameters, H sample distance parameters and H groups of delivery time information, wherein each group of delivery time information includes K simulated delivery times; The Pearson correlation coefficient is used to calculate the H sample time limit deviations of the H groups of delivery time information and the K delivery time limits; Taking the K package masses and the flight speed scale as energy consumption analysis constraints, load energy consumption prediction is performed during the delivery simulation of the H sample flight paths to obtain H sample energy consumption parameters; The H sample time parameters, the H sample distance parameters, the H sample time limit deviations and the H sample energy consumption parameters are weighted and solved using a preset weight configuration to obtain H sample path reliability; Serializing the H sample path reliabilities, and calling a first flight path from the H sample flight paths based on a ranking extreme value; The first delivery time limit deviation, the first flight time, the first flight distance and the first flight energy consumption of the first flight path are obtained from the H sample time parameters, the H sample distance parameters, the H sample time limit deviations and the H sample energy consumption parameters as the first path identifier; By analogy, flight paths are optimized based on the M groups of logistics association information and standard logistics performance to obtain M first flight paths.

3. The intelligent path planning method for a logistics drone as claimed in claim 2, characterized in that: Taking the K package masses and the flight speed scale as energy consumption analysis constraints, performing load energy consumption prediction during the delivery simulation of the H sample flight paths to obtain H sample energy consumption parameters, the method further includes: Comparing the H sample energy consumption parameters with the energy budget scale to obtain H energy consumption deviation vectors; The paths that do not meet the energy budget in the H sample flight paths are eliminated according to the H energy consumption deviation vectors.

4. The intelligent path planning method for a logistics drone according to claim 1, characterized in that: The M first path identifiers and the M load balancing coefficients are loaded into a task allocation evaluation function to perform task allocation fitness evaluation to obtain a first logistics planning fitness, the method further comprising: Loading the M first path identifiers and the M load balancing coefficients into the task allocation evaluation function to perform task allocation fitness evaluation to obtain M sample planning fitnesses; Calculating the mean of the M sample planning fitnesses to obtain the first logistics planning fitness; The M first flight paths are stored as a first group of logistics delivery paths in association with the first logistics planning fitness.

5. Intelligent path planning system for logistics drones, characterized by: A system for implementing the intelligent path planning method for a logistics drone according to any one of claims 1 to 4, the system comprising: A tag information reading module, used for reading multiple RFID tag information of multiple packages to be delivered through a local RFID scanner; The standard logistics performance acquisition module is used to interact with the logistics management cloud to obtain the standard logistics performance of M idle drones of the same model; A joint optimization module is used to perform joint optimization of logistics task allocation according to the standard logistics performance and multiple RFID tag information to obtain M delivery allocation strategies and M logistics delivery paths; A verification tag sequence positioning module is used to fit the M logistics delivery paths to the delivery validation map to locate M verification tag sequences; A loading deviation verification module, used for performing loading deviation verification through M airborne scanners pre-configured on the M idle drones during the process of mapping and loading the multiple packages to be delivered to the M idle drones according to the M delivery allocation strategies; A path execution deviation verification module is used to map and load the multiple packages to be delivered to the M idle drones according to the M delivery allocation strategies, and then perform path execution deviation verification through the M verification tag sequences when the M idle drones execute package delivery according to the M logistics delivery paths; The system also includes a logistics delivery path positioning module to perform the following operation steps: A plurality of logistics-related information is obtained by calling from the plurality of RFID tag information, wherein the logistics-related information includes package quality, target address and delivery time limit; a load capacity, a flight speed scale and an energy budget scale are obtained from the standard logistics performance; the load capacity is used as a package combination constraint, and the permutation and combination results of the plurality of packages to be delivered are judged according to the plurality of package qualities extracted from the plurality of logistics-related information, to obtain N groups of initial allocation strategies; the N groups of initial allocation strategies are jointly optimized for logistics task allocation according to the standard logistics performance and the plurality of logistics-related information, to obtain N logistics planning fitnesses and N groups of logistics delivery paths; the M delivery allocation strategies and M logistics delivery paths are located from the N groups of initial allocation strategies and the N groups of logistics delivery paths by serializing the N logistics planning fitnesses; The system also includes a logistics planning fitness acquisition module to perform the following operation steps: Disassemble the first group of initial allocation strategies to obtain M initial allocation strategies of the M idle drones; reorganize the multiple logistics association information according to the M initial allocation strategies to obtain M groups of logistics association information; optimize the flight path according to the M groups of logistics association information and standard logistics performance to obtain M first flight paths, wherein the first path identifier of the first flight path comprises a first delivery time limit deviation, a first flight time, a first flight distance and a first flight energy consumption; extract M groups of package qualities from the M groups of logistics association information, and perform load balancing evaluation based on the M groups of package qualities to obtain M load balancing coefficients; load the M first path identifiers and the M load balancing coefficients into the task allocation evaluation function to perform task allocation fitness evaluation to obtain the first logistics planning fitness; and so on, perform task allocation fitness evaluation on the N groups of initial allocation strategies to obtain the N logistics planning fitness; The system also includes an evaluation function building module to perform the following operation steps: Construct the task allocation evaluation function, which is as follows: Among them, A is the logistics planning fitness, r is the first delivery time limit deviation, T is the first flight time, D is the first flight distance, E is the first flight energy consumption, Lf is the load balance coefficient, and Lc is the load capacity.

Images