Unmanned Aerial Vehicle Logistics Distribution Receiving Management System Based on the Internet of Things

Abstract

The present invention belongs to the field of logistics planning, and particularly relates to an unmanned aerial vehicle logistics distribution receiving management system based on the Internet of Things. The system includes at least one unmanned aerial vehicle, a distribution module, and a distribution control module; the distribution module configures the optimal order distribution path for the unmanned aerial vehicle by means of a preset regional dynamic distribution network and the current distribution order information. The distribution control module controls the unmanned aerial vehicle to perform order distribution according to the optimal order distribution path; wherein, the regional dynamic distribution network is composed of a regional distribution point subnet and a comprehensive regional target distribution point subnet, and the two are connected by a connection relationship constructed by the historical optimal distribution path of each target distribution point. Moreover, each connection relationship is configured with a security control key constructed by a hash fingerprint, which is used to adjust the real-time abnormal distribution trajectory, so as to ensure the efficiency and security of the distribution process and realize the intelligent logistics distribution management of unmanned aerial vehicles based on the Internet of Things.

Description

Technical Field

[0001] The present invention belongs to the field of logistics planning, and particularly relates to an unmanned aerial vehicle logistics distribution receiving management system based on the Internet of Things. Background Art

[0002] The current unmanned aerial vehicle logistics distribution receiving management system is still in the development stage, and there are a series of problems to be solved urgently. In terms of information security, the existing system has limited encryption technology and is difficult to resist complex network attacks. The risk of criminals stealing goods information and tampering with order data is relatively high, seriously damaging the interests of logistics enterprises and consumers. For example, hackers modify the goods delivery address, resulting in wrong delivery of goods and disputes. In addition, since the delivery orders are different at each location, how to adjust the control scope of the delivery area according to the orders is also a key problem existing at present, which will affect the delivery efficiency.

[0003] For example, the Chinese patent application with the publication number CN104156843A discloses a logistics distribution management system and its method, which optimizes logistics distribution management by means of dividing distribution grids, etc. However, it does not fully consider the characteristics of unmanned aerial vehicle distribution, and there are deficiencies in information security protection and compatibility with unmanned aerial vehicles. It cannot effectively solve the key problems of unmanned aerial vehicle logistics distribution receiving management systems in aspects such as information security and equipment coordination, and it is difficult to meet the needs of modern logistics development. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention proposes an unmanned aerial vehicle logistics distribution receiving management system based on the Internet of Things. The system includes at least one unmanned aerial vehicle, a distribution module, and a delivery control module; the distribution module configures the optimal order delivery path for the unmanned aerial vehicle by means of a preset regional dynamic distribution network and current delivery order information. The delivery control module controls the unmanned aerial vehicle to perform order delivery according to this path; wherein, the regional dynamic distribution network is composed of a regional distribution point subnet and a comprehensive regional target distribution point subnet, and the two are connected by the connection relationship constructed by the historical optimal delivery path of each target distribution point. Moreover, each connection relationship is configured with a security control key constructed by a hash fingerprint, which is used to adjust the real-time abnormal delivery trajectory, so as to ensure the efficiency and security of the delivery process and realize the intelligent logistics distribution management of unmanned aerial vehicles based on the Internet of Things.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] An unmanned aerial vehicle logistics distribution receiving management system based on the Internet of Things, including at least one unmanned aerial vehicle, a distribution module, and a delivery control module;

[0007] The distribution module is used to configure the optimal order delivery path for at least one unmanned aerial vehicle through a preset regional dynamic distribution network and current delivery order information;

[0008] The delivery control module controls at least one unmanned aircraft to perform order delivery based on the configured optimal order delivery path.

[0009] The regional dynamic delivery network includes a regional delivery point subnet and a comprehensive regional target delivery point subnet; the regional delivery point subnet and the comprehensive regional target delivery point subnet are connected through the connection relationship constructed by the historical optimal delivery path of each target delivery point, and each connection relationship is configured with a security control key constructed by a hash fingerprint to adjust the real-time abnormal delivery trajectory.

[0010] Specifically, the regional dynamic delivery network includes a regional delivery point subnet and a comprehensive regional target delivery point subnet;

[0011] The construction steps of the comprehensive regional target delivery point subnet include:

[0012] Obtain the order delivery site, the corresponding delivery area information, and the historical order successful delivery information, and analyze to obtain N target position points successfully delivered by the i-th delivery site in the current delivery area before the current moment t, and construct the single-point target delivery point set corresponding to the i-th delivery site;

[0013] Based on the shortest time delivery path from the i-th delivery site to any target delivery point in the single-point target delivery point set, construct the optimal delivery connection relationship corresponding to any target delivery point;

[0014] Based on the single-point target delivery point set of the i-th delivery site and the corresponding optimal delivery connection relationship, obtain the target delivery point subnet corresponding to the i-th delivery site through the automatic annotation algorithm combined with the topology algorithm, and perform real-time annotation on the position and historical delivery frequency of each target delivery point.

[0015] Based on the historical delivery frequency corresponding to the target delivery point, analyze the delivery change curve of the corresponding target delivery point with 1 month as the time unit, and configure the delivery change curve to the corresponding target delivery point.

[0016] Specifically, the construction steps of the comprehensive regional target delivery point subnet further include:

[0017] Repeat the process of obtaining the target delivery point subnet corresponding to the i-th delivery site above, obtain the target delivery point subnet corresponding to each delivery site in the current delivery area, and perform annotation on the position of the target delivery point and the corresponding delivery change curve;

[0018] When the same target delivery point receives the deliveries from at least two different delivery sites, dynamically annotate the delivery information through the delivery site serial number.

[0019] Based on the target delivery point location information corresponding to all target delivery point subnets within the same delivery area, the delivery frequency of a single target delivery point per unit time, and the number of delivery stations performing deliveries per unit time, the current delivery area is adaptively grid-divided through an adaptive clustering algorithm combined with a grid division algorithm to obtain a comprehensive area target delivery point subnet.

[0020] Specifically, the acquisition steps of adaptively grid-dividing the current delivery area through an adaptive clustering algorithm combined with a grid division algorithm include:

[0021] Set the total number of delivery stations within the current delivery area as M, and the set of target delivery points of the i-th delivery station as , the delivery frequency of target delivery point j receiving deliveries from the i-th delivery station per unit time is , the shortest time path from the i-th delivery station to target delivery point j is , and the position coordinates of target delivery point j in the regional dynamic delivery network are ;

[0022] Based on the coordinates of any two target delivery points on the comprehensive area target delivery point subnet, the distance between the corresponding two target delivery points is calculated through the distance formula built into the adaptive clustering algorithm ;

[0023] Set the clustering radius threshold and the load radius threshold , and based on the delivery frequency of target delivery point j receiving deliveries from the i-th delivery station per unit time and the number of delivery stations corresponding to the goods received by target delivery point j per unit time, the delivery control load corresponding to each target delivery point is obtained;

[0024] Based on the clustering radius threshold , the load radius threshold and the delivery control load corresponding to each target delivery point, a clustering rule is constructed;

[0025] According to the clustering rule and the delivery control load corresponding to each target delivery point in the comprehensive area target delivery point subnet , through the adaptive clustering algorithm, the number of clusters and the information of the target delivery points corresponding to each cluster are obtained.

[0026] Specifically, the construction process of the clustering rule includes:

[0027] Select the target delivery point corresponding to the largest delivery control load as the starting clustering target delivery point, calculate the distance between the starting clustering target delivery point and any non-starting clustering target delivery point , and obtain the delivery control load corresponding to the corresponding target delivery point;

[0028] When the k-th target delivery point satisfies , it is determined whether is less than the load radius threshold. If it is satisfied, it is determined that the k-th target delivery point belongs to the initial clustering cluster corresponding to the starting clustering target delivery point; where is the delivery control load corresponding to the target delivery point j, is the delivery control load corresponding to the target delivery point k;

[0029] Repeat the above clustering judgment process to search the set of target delivery points corresponding to the subnet of the target delivery points in the comprehensive area. When the distance between any remaining target delivery point and the starting clustering target delivery point is greater than or the delivery control load corresponding to the current clustering target delivery point and the target delivery points that have been searched and meet the conditions is greater than or equal to the load radius threshold, stop searching for the target delivery points corresponding to the initial clustering cluster, and obtain the initial clustering cluster with the search completed;

[0030] Repeat the above process of the target delivery points corresponding to the initial clustering cluster to obtain the clustering cluster information and the number of clustering clusters corresponding to the remaining unclustered target delivery points.

[0031] Specifically, the construction steps of the subnet of the target delivery points in the comprehensive area further include:

[0032] Based on the number of target delivery points, the corresponding delivery control load, and the number of clustering target delivery points in each clustering cluster obtained, construct a sub-region load weight factor;

[0033] Input the sub-region load weight factor and the target delivery point information in the subnet of the target delivery points in the comprehensive area into the grid algorithm, and combine the preset delivery time window constraint matrix and the spatio-temporal reachability criterion to divide the subnet of the target delivery points in the comprehensive area to obtain a subnet of the target delivery points in the comprehensive area with dynamic sub-regions.

[0034] Specifically, the steps of controlling at least one unmanned aerial vehicle for order delivery include:

[0035] Obtain the delivery order and delivery station information corresponding to the current moment. According to the delivery location corresponding to the order, through the dynamic sub-regions and the corresponding target delivery point information included in the subnet of the target delivery points in the comprehensive area, make a preliminary allocation of the order;

[0036] Specifically, the preliminary allocation includes:

[0037] Match according to the delivery order information corresponding to the current moment and the target delivery point information in each sub-region of the subnet of the target delivery points in the comprehensive area. If the target delivery point corresponding to the same sub-region is matched, divide the corresponding order into the initial delivery set of the corresponding sub-region;

[0038] If no match is found, the current target delivery point is determined as a new target delivery point. According to the location information of the new target delivery point in the subnet of the target delivery points in the comprehensive area, the remaining delivery control load of each sub-region, and the reachability discrimination probability, the new target delivery point is divided into the sub-region with the shortest successful delivery time and the highest reachability probability. At the same time, the order corresponding to the new target delivery point is divided into the initial delivery set of the corresponding sub-region.

[0039] Specifically, the steps of controlling at least one unmanned aerial vehicle for order delivery further include:

[0040] Based on the initial delivery set corresponding to each sub-region, the rated load capacity and the rated cruising range of the unmanned aerial vehicle, the same initial delivery set corresponding to the same delivery site is allocated to obtain the set of unmanned aerial vehicles with allocated orders within the same sub-region.

[0041] If the orders corresponding to the set of unmanned aerial vehicles with allocated orders within the same sub-region are for the same target delivery point and the target delivery point is included in the sub-region corresponding to the subnet of the target delivery points in the comprehensive area, then through the connection relationship configured between the corresponding nodes of the subnet of the target delivery points in the comprehensive area and the subnet of the regional delivery points, the optimal order delivery path is configured for the set of unmanned aerial vehicles. At the same time, through the built-in safety control key, the flight trajectory of the optimal order delivery path is determined for abnormality.

[0042] Specifically, the steps of controlling at least one unmanned aerial vehicle for order delivery further include:

[0043] If each unmanned aerial vehicle in the set of unmanned aerial vehicles with allocated orders within the same sub-region corresponds to an order that contains at least two target delivery points, and at least one target delivery point has the same position as the existing target delivery point in the sub-region corresponding to the subnet of the target delivery points in the comprehensive area, then the connection relationship between the corresponding target delivery point and the delivery site included in the subnet of the target delivery points in the comprehensive area is used as the starting flight delivery path.

[0044] And by using the position coordinates and distances between the remaining target delivery points, through the shortest time path planning algorithm combined with the preset delivery time window constraint matrix and the spatio-temporal reachability criterion, the shortest delivery time path between the target delivery points within the corresponding sub-region is planned.

[0045] The planned shortest delivery time path is configured into the set of unmanned aerial vehicles, and through the configured safety control key, the flight trajectory of the shortest delivery time path is determined for abnormality.

[0046] Specifically, the steps of controlling at least one unmanned aerial vehicle for order delivery further include:

[0047] If each order corresponding to an unmanned aerial vehicle concentrated in the same sub-region with allocated orders has at least two target delivery points, but for at least one unmanned aerial vehicle, all the target delivery points of the order it carries are not the same as the positions of the existing target delivery points in the sub-region corresponding to the subnet of the target delivery points in the comprehensive region, then the unmanned aerial vehicle whose target delivery points of the carried order are not the same as the positions of the existing target delivery points in the corresponding sub-region is marked as a key planning aircraft;

[0048] According to the position information of the target delivery points corresponding to the orders carried by the key planning aircraft and the sub-region information corresponding to the initial delivery set, with the shortest delivery time and the maximum spatio-temporal reachability as the constraint conditions, combined with the path planning algorithm, the initial target delivery points that meet the shortest delivery time and the maximum spatio-temporal reachability of the key planning aircraft and the corresponding shortest delivery time path are obtained;

[0049] Mark the target delivery points corresponding to the orders of the key planning aircraft in the sub-region corresponding to the initial delivery set, and monitor the delivery control load corresponding to the marked sub-region. If it is greater than the load radius threshold, re-cluster and divide the target delivery points in the corresponding sub-region so that the load radius threshold is met.

[0050] Specifically, the construction process of the security control key includes:

[0051] According to the historical optimal delivery path, obtain the corresponding optimal path node sequence, time window constraint, path static attributes, physical parameters, and context identifier, and through binarization processing, obtain the constraint sequence feature space corresponding to the historical optimal delivery path;

[0052] According to the optimal path node sequence in the constraint sequence feature space, through the hash algorithm, obtain the SHA-256 hash value corresponding to each path node, and generate the root hash by recursively merging the hash values of adjacent path nodes;

[0053] According to the time window constraint, path static attributes, and physical parameters, through the keyed-hash message authentication code algorithm, obtain the spatio-temporal attribute hash corresponding to the historical optimal delivery path;

[0054] Concatenate the root hash, spatio-temporal attribute hash, and context identifier, and add a dynamic salt value to obtain the final hash fingerprint;

[0055] Embed the final hash fingerprint into the optimal path node sequence through the blockchain algorithm to obtain the security control key corresponding to the historical optimal delivery path.

[0056] Compared with the prior art, the beneficial effects of the present invention are:

[0057] In view of the deficiencies of the prior art, the present invention constructs a security control key and the corresponding historical optimal delivery path to ensure the security of the delivery path. The construction process includes binary processing, hash operation, etc. of the key information of the historical optimal delivery path. The generated security control key can be used to determine abnormal flight trajectories, effectively resist network attacks, reduce the risk of cargo information being stolen and order data being tampered with, avoid disputes such as wrong delivery of goods, and safeguard the interests of logistics enterprises and consumers. In addition, in terms of delivery efficiency, the system constructs a subnet of comprehensive regional target delivery points, uses an adaptive clustering algorithm combined with a grid division algorithm, considers various factors such as delivery stations, delivery frequencies, and shortest-time paths for adaptive grid division, and constructs a load weight factor for sub-regions. Combining the delivery time window constraint matrix and the spatio-temporal reachability criterion to divide dynamic sub-regions, so as to efficiently allocate orders according to the order delivery locations. At the same time, reasonable path planning is carried out according to the rated load capacity, rated cruising range of the unmanned aerial vehicle and the order situation. In case of special situations, re-clustering and division can be carried out to optimize the delivery, greatly improving the delivery efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 FIG. is a module diagram of the unmanned aerial vehicle logistics delivery receiving management system based on the Internet of Things according to an embodiment of the present invention;

[0059] Figure 2 FIG. is a flowchart for constructing a subnet of comprehensive regional target delivery points according to an embodiment of the present invention;

[0060] Figure 3 FIG. is a flowchart for constructing clustering rules according to an embodiment of the present invention;

[0061] Figure 4 FIG. is a flowchart for constructing a security control key according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0062] The unmanned aerial vehicle logistics delivery in urban or remote areas faces problems such as low cargo delivery efficiency, unreasonable delivery path planning, and lack of information security guarantee. For example, urban delivery is prone to congestion resulting in low efficiency, and delivery path planning in remote areas is difficult. At the same time, information is easily leaked and tampered with, affecting the normal operation of logistics delivery and the interests of all parties. Therefore, please refer to Figure 1 , an embodiment provided by the present invention: an unmanned aerial vehicle logistics delivery receiving management system based on the Internet of Things, including at least one unmanned aerial vehicle, a distribution module, and a delivery control module;

[0063] The distribution module is used to configure the optimal order delivery path for at least one unmanned aerial vehicle through a preset regional dynamic delivery network and current delivery order information;

[0064] The delivery control module controls at least one unmanned aerial vehicle to perform order delivery based on the configured optimal order delivery path;

[0065] The regional dynamic distribution network includes a regional distribution point subnet and a comprehensive regional target distribution point subnet; the regional distribution point subnet and the comprehensive regional target distribution point subnet are connected through a connection relationship constructed by a historical optimal distribution path of each target distribution point, and each connection relationship is configured with a security control key constructed by a hash fingerprint to adjust real-time abnormal distribution trajectories.

[0066] The system builds a regional dynamic distribution network, integrates regional distribution point subnetworks and comprehensive regional target distribution point subnetworks, and uses the historical optimal distribution path of each target distribution point to build a connection relationship and configure a security control key. It can effectively adjust the real-time abnormal distribution trajectory, greatly improving the safety and reliability of distribution. This is because the security control key constructed by hash fingerprint is generated based on the key information of the historical optimal distribution path. During the distribution process, the current trajectory can be compared with the historical optimal trajectory characteristics in real time. Once an abnormality occurs, it can be discovered and adjusted in time to ensure the safe delivery of goods.

[0067] Further, in this embodiment, the regional dynamic distribution network includes a regional distribution point subnet and a comprehensive regional target distribution point subnet;

[0068] For further information, see Figure 2 The steps of constructing the comprehensive regional target distribution point subnet in this embodiment include:

[0069] Obtain the order delivery site and the corresponding delivery area information and the successful delivery information of historical orders, and analyze and obtain the N target location points that the i-th delivery site in the current delivery area successfully delivered before the current time t, and construct a single-point target delivery point set corresponding to the i-th delivery site;

[0070] Furthermore, in this embodiment, the N target location points that were successfully delivered before the current time t are the user location points where the i-th delivery site of the historical order successful delivery information successfully delivered goods.

[0071] Based on the shortest time delivery path from the i-th delivery station in the single-point target delivery point set to any target delivery point in the single-point target delivery point set, the optimal delivery connection relationship corresponding to any target delivery point is constructed;

[0072] Based on the single-point target distribution point set of the i-th distribution station and the corresponding optimal distribution connection relationship, the target distribution point subnet corresponding to the i-th distribution station is obtained through the automatic labeling algorithm combined with the topological algorithm, and the location and historical distribution frequency of each target distribution point are labeled in real time;

[0073] Based on the historical delivery frequency corresponding to the target delivery point, analyze the delivery change curve of the corresponding target delivery point with 1 month as the time unit, and configure the delivery change curve to the corresponding target delivery point;

[0074] Repeat the process of obtaining the target delivery point subnet corresponding to the i-th delivery site above, obtain the target delivery point subnet corresponding to each delivery site in the current delivery area, and mark the position of the target delivery point and the corresponding delivery change curve;

[0075] When the same target delivery point receives deliveries from at least two different delivery sites, dynamically mark the delivery information through the delivery site serial number;

[0076] Based on the target delivery point position information corresponding to all target delivery point subnets in the same delivery area, the delivery frequency of a single target delivery point per unit time, and the number of delivery sites for which deliveries are made per unit time, perform adaptive grid division on the current delivery area through an adaptive clustering algorithm combined with a grid division algorithm to obtain a comprehensive regional target delivery point subnet;

[0077] Further, in this embodiment, the steps of performing adaptive grid division on the current delivery area through an adaptive clustering algorithm combined with a grid division algorithm include:

[0078] Set the total number of delivery sites in the current delivery area to M, the set of target delivery points of the i-th delivery site to , the delivery frequency of target delivery point j receiving deliveries from the i-th delivery site per unit time to , the shortest time path from the i-th delivery site to target delivery point j to , and the position coordinates of target delivery point j in the regional dynamic delivery network to ;

[0079] Based on the coordinates of any two target delivery points on the comprehensive regional target delivery point subnet, calculate the distance between the corresponding two target delivery points through the distance formula built into the adaptive clustering algorithm ;

[0080] Set the clustering radius threshold and the load radius threshold , and obtain the delivery control load corresponding to each target delivery point based on the delivery frequency of target delivery point j receiving deliveries from the i-th delivery site per unit time and the number of delivery sites for which the goods received by target delivery point j are delivered per unit time;

[0081] Based on the clustering radius threshold , the load radius threshold and the delivery control load corresponding to each target delivery point, construct a clustering rule;

[0082] According to the clustering rules and combining with the delivery control load corresponding to each target delivery point in the comprehensive regional target delivery point subnet , through the adaptive clustering algorithm, obtain the number of clusters and the information of the corresponding target delivery points in each cluster.

[0083] Further, please refer to Figure 3 , the construction process of the clustering rules in this embodiment includes:

[0084] Select the target delivery point corresponding to the largest delivery control load as the starting clustering target delivery point, and calculate the distance between the starting clustering target delivery point and any non-starting clustering target delivery point , and obtain the delivery control load corresponding to the corresponding target delivery point;

[0085] When the k-th target delivery point satisfies , judge Whether it is less than the load radius threshold. If it is satisfied, it is judged that the k-th target delivery point belongs to the initial cluster corresponding to the starting clustering target delivery point; where is the delivery control load corresponding to target delivery point j, is the delivery control load corresponding to target delivery point k;

[0086] Repeat the above clustering judgment process to search the set of corresponding target delivery points in the comprehensive regional target delivery point subnet. When the distance between any remaining target delivery point and the starting clustering target delivery point is greater than or the delivery control load corresponding to the current clustering target delivery point and the target delivery points that have been searched and meet the conditions is greater than or equal to the load radius threshold, stop searching for the target delivery points corresponding to the initial cluster, and obtain the initial cluster with the search completed;

[0087] Further, in this embodiment, the target delivery points that have been searched and meet the conditions are the target delivery points that have been determined to belong to the initial cluster.

[0088] Repeat the above process of the target delivery points corresponding to the initial cluster clustering to obtain the cluster information and the number of clusters corresponding to the remaining unclustered target delivery points.

[0089] Based on the number of target delivery points, the corresponding delivery control load, and the number of clustering target delivery points in each cluster obtained, construct the sub-region load weight factor;

[0090] Input the sub-region load weight factor and the target delivery point information in the integrated regional target delivery point subnet into the grid algorithm. Combining the preset delivery time window constraint matrix and the spatio-temporal reachability criterion, divide the integrated regional target delivery point subnet to obtain an integrated regional target delivery point subnet with dynamic sub-regions.

[0091] Further, the steps for obtaining the delivery time window constraint matrix in this embodiment include:

[0092] Collect the delivery time data of a large number of historical orders, including information such as the order placement time, the shipping time, the arrival time at each delivery node, and the final delivery time.

[0093] Further, in this embodiment, these data should cover the delivery situations of different time periods, different delivery regions, different delivery stations, and different types of orders.

[0094] According to the collected historical delivery time data, through statistical analysis methods, obtain the average duration, the shortest duration, the longest duration, and the duration distribution of deliveries in different time periods; for example, analyze the delivery time differences in different time periods such as weekdays and weekends, day and night, as well as the delivery time differences between different delivery regions.

[0095] Based on the analyzed delivery time distribution and combined with the actual business requirements, determine the delivery time window range for each target delivery point or delivery region.

[0096] Further, the delivery time window refers to the time period during which delivery is allowed to arrive; for example, if the delivery efficiency in a certain region is high in the morning on weekdays and the average delivery duration is 1 - 2 hours, then the delivery time window for this region can be set to 9:00 - 11:00 on weekdays.

[0097] Construct a delivery time window constraint matrix with the target delivery points or delivery regions as rows and different delivery time window ranges as columns.

[0098] Further, the elements in the delivery time window constraint matrix indicate whether delivery to the corresponding target delivery point is allowed within the corresponding time interval; for example, if delivery to target delivery point A is allowed within the time interval of 9:00 - 10:00, the corresponding element in the matrix is set to 1; if not, it is set to 0; at the same time, according to the actual situation, different weights can be assigned to different delivery time windows, and the weights can reflect the importance or priority of the time window; for example, for some urgent orders or orders with high requirements for delivery time, the weights of the corresponding delivery time windows can be set higher.

[0099] Further, the steps for obtaining the spatio-temporal reachability criterion in this embodiment of the distribution include:

[0100] Obtain the factors affecting the spatio-temporal accessibility criterion, and through the principal component analysis method, obtain the main influencing factors for judging spatio-temporal accessibility;

[0101] Furthermore, in this embodiment, the main influencing factors for judging spatio-temporal accessibility mainly include geographical location, traffic conditions, weather conditions, the performance of unmanned aerial vehicles, etc. The geographical location determines the distance between the delivery starting point and the target delivery point; traffic conditions (such as road congestion, no-fly zones, etc.) will affect the flight speed and path selection of unmanned aerial vehicles; weather conditions (such as wind speed, rainfall, fog, etc.) will have an impact on the flight safety and efficiency of unmanned aerial vehicles; the performance of unmanned aerial vehicles (such as maximum flight speed, endurance mileage, etc.) also limits the area that can be reached within a certain period of time.

[0102] Quantify the determined main influencing factors. For example, for the geographical location, it can be quantified by calculating the straight-line distance or the actual flight distance between the delivery starting point and the target delivery point; for traffic conditions, according to historical traffic data or real-time traffic information, the degree of road congestion can be divided into different levels, and a corresponding speed decay coefficient can be set for each level to reflect the impact on the flight speed of unmanned aerial vehicles; for example, the congestion levels are divided into mild congestion, moderate congestion, and severe congestion, corresponding to speed decay coefficients of 0.8, 0.6, and 0.4 respectively; for weather conditions, set corresponding limiting conditions or speed adjustment coefficients according to the impact degree of different weather conditions on the flight of unmanned aerial vehicles; when the wind speed exceeds a certain threshold, the flight speed of the unmanned aerial vehicle needs to be reduced by a certain proportion; for the performance of unmanned aerial vehicles, clarify its maximum flight speed, endurance mileage and other parameters, and adjust according to the actual situation.

[0103] Based on the quantified main influencing factors, establish a spatio-temporal accessibility model. Furthermore, the method adopted in this embodiment is a calculation model based on distance and time, such as considering the flight speed and flight time of unmanned aerial vehicles under different conditions. Assume that the flight speed of the unmanned aerial vehicle under no wind and no congestion conditions is the standard flight speed, and the speed adjusted according to traffic conditions and weather conditions is the simulated flight speed. Then the flight time from the delivery starting point to the target delivery point can be calculated by dividing the distance by the simulated flight speed;

[0104] According to the established spatio-temporal accessibility model, set the spatio-temporal accessibility criterion; for example, set a time threshold and a distance threshold. When the flight time from the delivery starting point to the target delivery point is less than the time threshold, and the flight distance is less than the endurance mileage of the unmanned aerial vehicle and less than the distance threshold, it is considered that the target delivery point is spatio-temporally reachable, and the criterion is true; otherwise, the criterion is false.

[0105] In addition, in this embodiment, the criteria are further refined and adjusted according to actual business requirements. For example, when considering the situation of collaborative distribution by multiple unmanned aerial vehicles, or combining the delivery time window constraint, it is ensured that the target delivery point can be reached within the allowed time range.

[0106] In this process, a single-point target delivery point set is constructed by obtaining the order delivery site and historical delivery information, and then the optimal delivery connection relationship is established. The target delivery point subnet is generated by using the automatic annotation algorithm combined with the topology algorithm, which can clearly present the delivery relationship of each delivery point. The position and historical delivery frequency of the target delivery point are annotated in real time, and the delivery change curve is analyzed on a monthly basis, which helps to accurately grasp the change law of delivery demand. At the same time, through the adaptive clustering algorithm combined with the grid division algorithm for adaptive grid division, factors such as the total number of delivery sites, delivery frequency, shortest time path, and target delivery point coordinates are fully considered, the clustering rules are constructed and the clustering information is obtained, and then the dynamic sub-region is divided by combining the sub-region load weight factor, the delivery time window constraint matrix, and the spatio-temporal reachability criterion, which can make the delivery area division more reasonable and efficient. For example, the delivery time window constraint matrix is obtained through statistical analysis of a large amount of historical delivery time data, and the delivery time window range is determined according to different time periods, delivery regions, delivery sites, and order types. The constructed time window constraint matrix can effectively limit the delivery time and improve the delivery efficiency. The spatio-temporal reachability criterion comprehensively considers factors such as geographical location, traffic conditions, weather conditions, and the performance of unmanned aerial vehicles, establishes a model and sets the criterion through quantitative processing, which can ensure that the unmanned aerial vehicle can reach the target delivery point in space and time, and improve the feasibility and success rate of delivery.

[0107] Furthermore, in this embodiment, the steps of controlling at least one unmanned aerial vehicle to perform order delivery include:

[0108] Obtain the delivery order and delivery site information corresponding to the current moment, and according to the delivery position corresponding to the order, perform an initial allocation of the order through the dynamic sub-region and the corresponding target delivery point information included in the comprehensive regional target delivery point subnet.

[0109] Specifically, the initial allocation includes:

[0110] Match the delivery order information corresponding to the current moment with the target delivery point information in each sub-region of the comprehensive regional target delivery point subnet. If the corresponding target delivery point in the same sub-region is matched, the corresponding order is divided into the initial delivery set of the corresponding sub-region.

[0111] If no match is found, the current target delivery point is determined as a new target delivery point. According to the position information of the new target delivery point in the subnet of the comprehensive regional target delivery points, the remaining delivery control load of each sub-region, and the reachability discrimination probability, the new target delivery point is divided into the sub-region with the shortest successful delivery time and the highest reachability probability. At the same time, the order corresponding to the new target delivery point is divided into the initial delivery set of the corresponding sub-region.

[0112] Based on the initial delivery set corresponding to each sub-region, the rated load capacity and the rated cruising range of the unmanned aerial vehicle, the same initial delivery set corresponding to the same delivery site is allocated to obtain the set of unmanned aerial vehicles with the allocated orders within the same sub-region.

[0113] If the orders corresponding to the set of unmanned aerial vehicles with the allocated orders within the same sub-region are for the same target delivery point and this target delivery point is included in the sub-region corresponding to the subnet of the comprehensive regional target delivery points, then through the connection relationship configured between the corresponding nodes of the subnet of the comprehensive regional target delivery points and the subnet of the regional delivery points, the optimal order delivery path is configured for the set of unmanned aerial vehicles. At the same time, through the built-in safety control key, the flight trajectory of the optimal order delivery path is determined for anomalies.

[0114] If each unmanned aerial vehicle in the set of unmanned aerial vehicles with the allocated orders within the same sub-region corresponds to an order that contains at least two target delivery points, and at least one target delivery point has the same position as the existing target delivery point within the sub-region corresponding to the subnet of the comprehensive regional target delivery points, then the connection relationship between the corresponding target delivery point included in the subnet of the comprehensive regional target delivery points and the delivery site is used as the starting flight delivery path.

[0115] And using the position coordinates and distances between the remaining target delivery points, through the shortest time path planning algorithm combined with the preset delivery time window constraint matrix and the spatio-temporal reachability criterion, the shortest delivery time path planning between the target delivery points within the corresponding sub-region is carried out.

[0116] The planned shortest delivery time path is configured into the set of unmanned aerial vehicles, and through the configured safety control key, the flight trajectory of the shortest delivery time path is determined for anomalies.

[0117] If each unmanned aerial vehicle in the set of unmanned aerial vehicles with the allocated orders within the same sub-region corresponds to an order that contains at least two target delivery points, but all the target delivery points of the order carried by at least one unmanned aerial vehicle are not in the same position as the existing target delivery points within the sub-region corresponding to the subnet of the comprehensive regional target delivery points, then the unmanned aerial vehicle whose target delivery points of the carried order are not in the same position as the existing target delivery points within the corresponding sub-region is marked as a key planning aircraft.

[0118] According to the location information of the delivery order target distribution points corresponding to the key planning aircraft and the sub-region information corresponding to the initial delivery set, combined with the path planning algorithm with the shortest delivery time and the maximum spatio-temporal reachability as the constraint conditions, the initial target distribution points that meet the shortest delivery time and the maximum spatio-temporal reachability of the key planning aircraft and the corresponding shortest delivery time paths are obtained;

[0119] Mark the order target distribution points corresponding to the key planning aircraft within the sub-regions corresponding to the initial delivery set, and monitor the delivery control load corresponding to the marked sub-regions. If it is greater than the load radius threshold, re-cluster and divide the target distribution points within the corresponding sub-regions to meet the load radius threshold.

[0120] During the delivery process of unmanned aircraft orders, first, based on the dynamic sub-regions and target distribution point information of the comprehensive regional target distribution point subnet, perform the initial order allocation. If the same sub-region target distribution points are matched, directly divide the orders. If not, divide the new target distribution points and orders according to the location, remaining delivery control load, and reachability discrimination probability. This method can quickly and reasonably allocate the orders to the appropriate sub-regions, improving the accuracy and efficiency of order allocation and laying a good foundation for subsequent deliveries.

[0121] Secondly, based on the initial delivery set of the sub-region, the rated load capacity and the rated cruising range of the unmanned aircraft, perform the secondary allocation to obtain the set of unmanned aircraft with allocated orders, which ensures the reasonable load of the unmanned aircraft, avoids delivery failures caused by overloading or insufficient voyage, and guarantees the feasibility of the delivery.

[0122] When the set of unmanned aircraft corresponds to a single target distribution point, configure the optimal path using the subnet node connection relationship and monitor for anomalies through the safety control key, which can ensure the optimal and safe delivery path; if each unmanned aircraft corresponds to multiple target distribution points and some target distribution points are in the same position as the existing target points in the comprehensive regional target distribution point subnet, use the existing connection relationship as the starting path, and combine the shortest time path planning algorithm, the delivery time window constraint matrix, and the spatio-temporal reachability criterion to plan the subsequent paths. This not only utilizes the existing delivery relationships but also comprehensively considers various factors to optimize the paths, improving the delivery efficiency and ensuring reachability within the specified time; if there are key planning aircraft, plan the paths with the shortest delivery time and the maximum spatio-temporal reachability as the constraints, mark the target distribution points and monitor the sub-region delivery control load, and re-cluster and divide when the load radius threshold is exceeded, which guarantees the efficient delivery of special orders and at the same time maintains the balance of the sub-region delivery control load, avoiding local overload from affecting the overall delivery efficiency.

[0123] Furthermore, please refer to Figure 4 , the construction process of the safety control key in this embodiment includes:

[0124] Based on the historical optimal delivery route, obtain the corresponding optimal route node sequence, time window constraints, route static attributes, physical parameters, and context identifiers, and through binarization processing, obtain the constraint sequence feature space corresponding to the historical optimal delivery route;

[0125] Furthermore, in this embodiment, the optimal route node sequence is a set of longitude and latitude coordinates corresponding to the flight path, the time window constraint is the earliest and latest arrival times of each node, the route static attributes are the maximum altitude limit, airspace type, and priority weight, and the physical parameters include the average speed, turning radius threshold, and flight energy consumption coefficient under different loads; the context identifiers include the delivery site ID, sub-region ID, hash salt value, etc.; the hash salt value is the date attribute of the current day.

[0126] According to the optimal route node sequence in the constraint sequence feature space, through the hash algorithm, obtain the SHA-256 hash value corresponding to each route node, and generate the root hash by recursively merging the hash values of adjacent route nodes;

[0127] According to the time window constraints, route static attributes, and physical parameters, through the keyed-hash message authentication code algorithm, obtain the spatio-temporal attribute hash corresponding to the historical optimal delivery route;

[0128] Concatenate the root hash, spatio-temporal attribute hash, and context identifiers, and add a dynamic salt value to obtain the final hash fingerprint;

[0129] Embed the final hash fingerprint into the optimal route node sequence through the blockchain algorithm to obtain the security control key corresponding to the historical optimal delivery route.

[0130] Furthermore, the process of using the security control key to determine the abnormality of the flight trajectory of the optimal order delivery route in this embodiment includes:

[0131] When the drone is flying, real-time collect trajectory data (coordinates, speed, altitude) and timestamps, according to the same node sequence as the historical route, and extract the current time window status (remaining delivery time, airspace control update) of each route node block;

[0132] Calculate the instant hash according to the information of each route node and the current time window status, and gradually construct a MerkleTree to generate the real-time root hash. If the number of route nodes changes (such as detouring), trigger dynamic block reorganization;

[0133] Match the generated real-time root hash with the root hash corresponding to the historical optimal delivery route stored in the blockchain of the optimal historical route. If they match, the route structure is legal; if they do not match, mark it as potentially structurally abnormal (such as deviating from the preset route);

[0134] Recalculate the spatio-temporal attribute hash based on the current time window and physical state (such as the remaining battery level), and compare it with the spatio-temporal attribute hash of the historical optimal delivery route. If the difference exceeds a preset threshold (such as a time deviation > 10%), mark it as a spatio-temporal anomaly;

[0135] Regenerate the final hash fingerprint using the daily salt value and compare it with the stored final hash fingerprint. If it matches exactly: the route is trustworthy. If there is a partial match: trigger a risk level assessment (such as a yellow warning). If it does not match: raise a red alert and freeze the control authority;

[0136] When an anomaly is detected, locate the deviated node based on the block difference of the current Merkle Tree, and generate an alternative sub-path through the minimum repair path algorithm:

[0137] If a key leakage risk is detected (such as the same salt value triggering anomalies multiple times), automatically enable the alternative key and recalculate the hash chain.

[0138] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make changes, modifications, substitutions, and variations to the above embodiments without departing from the spirit and scope protected by the present invention and the claims. All of these fall within the protection scope of the present invention.

Claims

1. An unmanned aerial vehicle logistics distribution receiving management system based on the Internet of Things, characterized in that, It includes at least one unmanned aerial vehicle, a distribution module, and a delivery control module; The distribution module is used to configure the optimal order delivery path for at least one unmanned aerial vehicle through a preset regional dynamic distribution network and current delivery order information; The delivery control module controls at least one unmanned aerial vehicle to perform order delivery based on the configured optimal order delivery path; The regional dynamic distribution network includes a regional delivery point subnet and a comprehensive regional target delivery point subnet; the regional delivery point subnet and the comprehensive regional target delivery point subnet are connected through the connection relationship constructed by the historical optimal delivery path of each target delivery point, and each connection relationship is configured with a security control key constructed by a hash fingerprint to adjust the real-time abnormal delivery trajectory; The construction steps of the comprehensive regional target delivery point subnet include: Obtain the order delivery site, the corresponding delivery area information, and the historical order successful delivery information, and analyze to obtain N target location points successfully delivered by the i-th delivery site in the current delivery area before the current time t, and construct the single-point target delivery point set corresponding to the i-th delivery site; Based on the shortest time delivery path from the i-th delivery site to any target delivery point in the single-point target delivery point set, construct the optimal delivery connection relationship corresponding to any target delivery point; Based on the single-point target delivery point set of the i-th delivery site and the corresponding optimal delivery connection relationship, through an automatic annotation algorithm combined with a topology algorithm, obtain the target delivery point subnet corresponding to the i-th delivery site, and perform real-time annotation on the position and historical delivery frequency of each target delivery point; Based on the historical delivery frequency corresponding to the target delivery point, analyze the delivery change curve of the corresponding target delivery point with 1 month as the time unit, and configure the delivery change curve to the corresponding target delivery point; Repeat the process of obtaining the target delivery point subnet corresponding to the i-th delivery site, obtain the target delivery point subnet corresponding to each delivery site in the current delivery area, and perform annotation on the target delivery point position and the corresponding delivery change curve; When the same target delivery point receives deliveries from at least two different delivery sites, dynamically annotate the delivery information through the delivery site serial number; Based on the target delivery point position information corresponding to all target delivery point subnets in the same delivery area, the delivery frequency per unit time of a single target delivery point, and the number of delivery sites performing deliveries per unit time, perform adaptive grid division on the current delivery area through an adaptive clustering algorithm combined with a grid division algorithm to obtain the comprehensive regional target delivery point subnet; The steps of performing adaptive grid division on the current delivery area through an adaptive clustering algorithm combined with a grid division algorithm include: Set the total number of distribution stations in the current distribution area as M, and the set of target delivery points of the i-th distribution station is , and the delivery frequency of the target delivery point j received from the i-th distribution station per unit time is , the shortest time path from the i-th distribution station to the target delivery point j is , and the position coordinates of the target delivery point j in the regional dynamic distribution network are ; Based on the coordinates of any two target delivery points on the comprehensive regional target delivery subnet, the distance between the corresponding two target delivery points is calculated through the distance formula built into the adaptive clustering algorithm ; Set the clustering radius threshold and the load radius threshold , and obtain the delivery control load corresponding to each target delivery point based on the delivery frequency of the i-th delivery site received by the target delivery point j within a unit time and the number of delivery sites corresponding to the goods received by the target delivery point j within a unit time. Based on the clustering radius threshold , the load radius threshold and the delivery control load corresponding to each target delivery point, a clustering rule is constructed; According to the clustering rules and combining the distribution control loads corresponding to each target distribution point in the comprehensive regional target distribution point subnet , through an adaptive clustering algorithm, obtain the number of clusters and the information of the corresponding target distribution points in each cluster; The construction steps of the comprehensive regional target delivery point subnet further include: Based on the number of target delivery points and the corresponding delivery control load in each clustering cluster obtained, and the number of target delivery points in the clustering, construct a sub-region load weight factor; Input the sub-region load weight factor and the target delivery point information in the comprehensive regional target delivery point subnet into the grid algorithm. Combine the preset delivery time window constraint matrix and the spatio-temporal reachability criterion to divide the comprehensive regional target delivery point subnet, and obtain a comprehensive regional target delivery point subnet with dynamic sub-regions.

2. The unmanned aerial vehicle logistics distribution receiving management system based on the Internet of Things according to claim 1, characterized in that, The construction process of the clustering rule includes: Select the target delivery point corresponding to the maximum delivery control load as the starting clustering target delivery point, and calculate the distance between the starting clustering target delivery point and any non-starting clustering target delivery point , and obtain the delivery control load corresponding to the corresponding target delivery point; When the k-th target delivery point satisfies , judge is less than the load radius threshold . If it is satisfied, it is judged that the k-th target delivery point belongs to the initial cluster corresponding to the starting cluster target delivery point; where is the delivery control load corresponding to the target delivery point j, is the delivery control load corresponding to the target delivery point k; Repeat the above clustering judgment process to search for the corresponding set of target delivery points in the subnet of the comprehensive regional target delivery points. When the distance between any remaining target delivery point and the starting clustering target delivery point is greater than or when the delivery control load corresponding to the current clustering target delivery point and the target delivery points that have been searched and meet the conditions is greater than or equal to the load radius threshold, stop the search for the target delivery points corresponding to the initial clustering cluster, and obtain the initial clustering cluster with the search completed; Repeat the above process of the target delivery points corresponding to the initial clustering clusters to obtain the clustering cluster information and the number of clustering clusters corresponding to the remaining unclustered target delivery points.

3. The unmanned aerial vehicle logistics distribution receiving management system based on the Internet of Things according to claim 2, wherein The step of controlling at least one unmanned aerial vehicle to perform order delivery includes: Obtain the delivery order and delivery station information corresponding to the current moment. According to the delivery location corresponding to the order, perform an initial allocation of the order through the dynamic sub-regions and the corresponding target delivery point information included in the comprehensive regional target delivery point subnet. The initial allocation includes: Match according to the delivery order information corresponding to the current moment and the target delivery point information in each sub-region of the comprehensive regional target delivery point subnet. If the target delivery point corresponding to the same sub-region is matched, divide the corresponding order into the initial delivery set of the corresponding sub-region; If not matched, determine that the current target delivery point is a new target delivery point. According to the position information of the new target delivery point in the comprehensive regional target delivery point subnet, the remaining delivery control load of each sub-region, and the reachability discrimination probability, divide the new target delivery point into the sub-region with the shortest successful delivery time and the highest reachability probability. At the same time, divide the order corresponding to the new target delivery point into the initial delivery set of the corresponding sub-region.

4. The unmanned aerial vehicle logistics distribution receiving management system based on the Internet of Things according to claim 3, wherein, The step of controlling at least one unmanned aerial vehicle to perform order delivery further includes: Based on the initial delivery set corresponding to each sub-region, the rated load capacity and the rated cruising range of the unmanned aerial vehicle, allocate the same initial delivery set corresponding to the same delivery station to obtain the set of unmanned aerial vehicles with allocated orders in the same sub-region; If the orders corresponding to the set of unmanned aerial vehicles with allocated orders in the same sub-region correspond to the same target delivery point and the target delivery point is included in the sub-region corresponding to the comprehensive regional target delivery point subnet, then through the connection relationship configured between the corresponding nodes of the comprehensive regional target delivery point subnet and the regional delivery point subnet, perform the optimal order delivery path configuration for the set of unmanned aerial vehicles, and at the same time, through the built-in safety control key, perform the abnormal determination of the flight trajectory of the optimal order delivery path.

5. The unmanned aerial vehicle logistics distribution receiving management system based on the Internet of Things according to claim 4, characterized in that The step of controlling at least one unmanned aerial vehicle to perform order delivery further includes: If each unmanned aerial vehicle corresponding to the set of unmanned aerial vehicles with allocated orders in the same sub-region corresponds to an order that includes at least two target delivery points, and at least one target delivery point has the same position as the existing target delivery point in the sub-region corresponding to the comprehensive regional target delivery point subnet, then use the connection relationship between the corresponding target delivery point and the delivery station included in the comprehensive regional target delivery point subnet as the starting flight delivery path; Using the corresponding position coordinates and distances between the remaining target delivery points, through the shortest time path planning algorithm in combination with the preset delivery time window constraint matrix and the spatio-temporal reachability criterion, perform the shortest delivery time path planning between the target delivery points within the corresponding sub-region; Configure the planned shortest delivery time path into the set of unmanned aerial vehicles, and through the configured safety control key, determine the abnormality of the flight trajectory of the shortest delivery time path.

6. The Internet of Things-based unmanned aerial vehicle logistics distribution receiving management system according to claim 5, characterized in that The step of controlling at least one unmanned aerial vehicle to perform order delivery further includes: If each unmanned aerial vehicle corresponding to the order in the set of unmanned aerial vehicles with allocated orders in the same sub-region has at least two target delivery points in its order, but all the target delivery points of the order carried by at least one unmanned aerial vehicle are different from the positions of the existing target delivery points in the sub-region corresponding to the subnet of the comprehensive region target delivery points, then mark the unmanned aerial vehicle whose target delivery points of the carried order are different from the positions of the existing target delivery points in the corresponding sub-region as the key planning aircraft; According to the position information of the target delivery points of the order carried by the key planning aircraft and the sub-region information corresponding to the initial delivery set, with the shortest delivery time and the maximum spatio-temporal reachability as the constraint conditions and combined with the path planning algorithm, obtain the initial target delivery points that meet the shortest delivery time and the maximum spatio-temporal reachability of the key planning aircraft and the corresponding shortest delivery time path; Mark the order target delivery points corresponding to the key planning aircraft in the sub-region corresponding to the initial delivery set, and monitor the delivery control load corresponding to the marked sub-region. If it is greater than the load radius threshold, re-cluster and divide the target delivery points in the corresponding sub-region so as to meet the load radius threshold.

7. The unmanned aerial vehicle logistics distribution receiving management system based on the Internet of Things according to claim 6, characterized in that, The construction process of the safety control key includes: According to the historical optimal delivery path, obtain the corresponding optimal path node sequence, time window constraint, path static attributes, physical parameters and context identifier, and through binarization processing, obtain the constraint sequence feature space corresponding to the historical optimal delivery path; According to the optimal path node sequence in the constraint sequence feature space, through the hash algorithm, obtain the SHA-256 hash value corresponding to each path node, and generate the root hash by recursively merging the hash values of adjacent path nodes; According to the time window constraint, path static attributes and physical parameters, through the keyed-hash message authentication code algorithm, obtain the spatio-temporal attribute hash corresponding to the historical optimal delivery path; Concatenate the root hash, spatio-temporal attribute hash and context identifier, and add a dynamic salt value to obtain the final hash fingerprint; Embed the final hash fingerprint into the optimal path node sequence through the blockchain algorithm to obtain the safety control key corresponding to the historical optimal delivery path.

Images