Method and device for determining candidate priorities of take-off and landing points of unmanned aerial vehicle based on complex network model, medium and equipment

Abstract

The invention relates to the technical field of unmanned aerial vehicles, in particular to an unmanned aerial vehicle take-off and landing point candidate priority determination method and device based on a complex network model, a medium and equipment, and the method comprises the steps: obtaining a candidate priority corresponding to an initial unmanned aerial vehicle take-off and landing point according to a take-off and landing point weight, an air route weight, a node betweenness, a clustering coefficient and a preset initial weight; m candidate unmanned aerial vehicle take-off and landing points are screened out, the difference between the candidate take-off and landing point prediction scores of the M candidate unmanned aerial vehicle take-off and landing points is measured according to the air route abnormal degree, the air route length and the second traffic flow, and the screening scores of the M candidate unmanned aerial vehicle take-off and landing points are obtained to represent the screening accuracy. And as the basis of updating the initial weight set and the candidate priorities, M is used for screening target unmanned aerial vehicle take-off and landing points and providing setting positions for unmanned aerial vehicle take-off and landing facilities, so that the effect of balancing the distribution efficiency and distribution intensity of the unmanned aerial vehicle corresponding to each target unmanned aerial vehicle take-off and landing point is achieved, and the overall working efficiency of the unmanned aerial vehicle is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV), and in particular to a method, device, medium and equipment for determining the priority of candidate take-off and landing points of unmanned aerial vehicle (UAV) based on a complex network model. Background Art

[0002] Traffic congestion in cities has seriously affected the delivery speed of traditional logistics vehicles, while drones can fly straight in the air, avoiding traffic jams and traffic lights on the ground, greatly shortening the delivery time, optimizing the work structure of logistics delivery personnel, meeting the transportation needs of various distances within the city, and improving transportation safety and efficiency. Therefore, with the rapid development of drone technology, drones have more and more extensive applications in logistics and distribution and other fields.

[0003] In the prior art, the execution of various tasks by drones relies on the constructed complex network model. By analyzing the various take-off and landing points in the complex network model of the drone and the areas covered by their flight routes, the length of each route and the traffic flow, etc., an influence coefficient is set for each factor, and the priority of each preset drone take-off and landing point is measured in combination with various factors and their weights. Locations located at the intersection of multiple routes or covering a wide area, close to longer routes or routes with higher flight frequency are selected as take-off and landing points, so as to facilitate drones to quickly access different routes, improve mission response speed, reduce flight distance and time when performing tasks, and improve efficiency.

[0004] However, the accuracy of the above-mentioned take-off and landing point screening depends on the accuracy of the weights corresponding to various factors. The existing technology usually uses artificially set fixed weights, which are difficult to adapt well to the dynamic changes of the route network in the complex network model, and difficult to flexibly adjust according to factors such as different mission types and seasonal changes. For example, with changes in mission requirements, the opening of new routes, the adjustment of existing routes, the impact of bad weather, etc., the above method is difficult to quickly and accurately re-evaluate the priority of each preset drone take-off and landing point, making it difficult to screen out suitable take-off and landing points, reducing the flight efficiency, safety and reliability of drones.

[0005] Therefore, when measuring the candidate priorities of UAV take-off and landing points based on complex network models, how to improve the accuracy of candidate priorities and thus improve the rationality of site selection for UAV take-off and landing points becomes an urgent problem to be solved. Summary of the invention

[0006] In view of the above technical problems, the technical solution adopted by the present invention is a method for determining the priority of candidate take-off and landing points of unmanned aerial vehicles based on a complex network model. The method for determining the priority of candidate take-off and landing points of unmanned aerial vehicles comprises the following steps:

[0007] S1000, obtain the initial drone take-off and landing point set A = {A1, A2, ..., A n , ..., A N}、Predicted flight route set B={B1,B2,……,B n , ..., B N}、The take-off and landing point weight set Q={Q1,Q2,……,Q n , ..., Q N}、Route weight set P={P1,P2,……,P n , ..., P N}、Node betweenness set JS={JS1,JS2,……,JS n , ..., JS N}、Clustering coefficient set JL={JL1,JL2,……,JL n , ..., JL N}、Route abnormality degree set D={D1,D2,……,D n , ..., D N}、Route length set E={E1,E2,……,E n ,……,E N} and the second traffic flow set F = {F1, F2, ..., F n , ..., F N}, where A n refers to the nth initial drone take-off and landing point, B n = {B n1 , B n2 , ..., B ni , ..., B nI(n)}, B ni A n The corresponding i-th predicted flight route, Q n A n The corresponding take-off and landing point weight, P n = {P n1 , P n2 , ..., P ni , ..., P nI(n)}, P ni Refers to B ni The corresponding route weight, JS n A n The corresponding node betweenness, JL n A n The corresponding clustering coefficient, D n ={D n1 , D n2 , ..., D ni , ..., D nI(n)}, D ni Refers to Bni The corresponding route abnormality degree, E n ={E n1 , E n2 ,……,E ni ,……,E nI(n)}, E ni Refers to B ni The corresponding route length, F n ={F n1 , F n2 , ..., F ni , ..., F nI(n)}, F ni Refers to B ni The corresponding second traffic flow, n=1, 2, ..., N, N refers to the total number of initial drone take-off and landing points, i=1, 2, ..., I(n), I(n) refers to the total number of predicted flight routes corresponding to the nth initial drone take-off and landing point.

[0008] S2000: According to A, B, Q, P, JS, JL and the preset initial weight set QZ = {QZ1, QZ2, QZ3, QZ4, QZ5}, a candidate priority set HX = {HX1, HX2, ..., HX n , ..., HX N}, where QZ1 refers to the preset weight of the take-off and landing point weight, QZ2 refers to the preset weight of the route weight, QZ3 refers to the preset weight of the number of routes, QZ4 refers to the preset weight of the node betweenness, QZ5 refers to the preset weight of the clustering coefficient, and A n Corresponding candidate priority HX n Meet the following conditions:

[0009] HX n =QZ1×Q n +QZ2×∑ i＝1 I(n) (P ni / I(n))+QZ3×I(n)+QZ4×JS n +QZ5×JL n .

[0010] S3000, based on HX, selects M candidate UAV take-off and landing points from N initial UAV take-off and landing points.

[0011] S4000: Based on D, E and F, obtain the candidate take-off and landing point prediction score set FZ = {FZ1, FZ2, ..., FZ n , ..., FZ N}, where A n Corresponding candidate take-off and landing point prediction score FZ n Meet the following conditions:

[0012] FZ n =β1×(Σ i＝1 I(n) e^(-D ni ))+β2×(∑ n＝1 N (Σ i＝1 I(n) E ni ))+β3×(∑ n＝1 N (Σ i＝1 I(n) F ni )), where β1

[0013] refers to the preset abnormal weight, β2 refers to the preset route length weight, β3 refers to the preset second flow weight, and e refers to the natural constant.

[0014] S5000: According to FZ, the screening scores FS of M candidate drone take-off and landing points are obtained, where FS meets the following conditions:

[0015] FS=(Σ n＝1 N (FZ n -FZ0) 2 / N, where the candidate take-off and landing point prediction mean FZ0 ​​= Σ n＝1 N (FZ n ) / N.

[0016] S6000, if FS>FS 0 , then update QZ, and use the updated QZ to replace the preset initial weight set, and repeat step S2000 until FS ≤ FS 0 , get FS≤FS 0 The candidate priority set HX when FS≤FS 0 The candidate priority set HX at this time is used to screen the target UAV take-off and landing point from the N initial UAV take-off and landing points, FS 0 Refers to the preset screening score threshold.

[0017] The present invention also provides a device for determining the priority of candidate take-off and landing points of unmanned aerial vehicles based on a complex network model, and the device for determining the priority of candidate take-off and landing points of unmanned aerial vehicles comprises:

[0018] The data acquisition module is used to obtain the initial drone take-off and landing point set A = {A1, A2, ..., A n , ..., A N}、Predicted flight route set B={B1,B2,……,B n , ..., B N}、The take-off and landing point weight set Q={Q1,Q2,……,Q n , ..., Q N}、Route weight set P={P1,P2,……,P n , ..., P N}、Node betweenness set JS={JS1,JS2,……,JS n , ..., JS N}、Clustering coefficient set JL={JL1,JL2,……,JL n , ..., JL N}、Route abnormality degree set D={D1,D2,……,D n , ..., D N}、Route length set E={E1,E2,……,E n ,……,E N} and the second traffic flow set F = {F1, F2, ..., F n , ..., F N}, where A n refers to the nth initial drone take-off and landing point, B n = {B n1 , B n2 , ..., B ni , ..., B nI(n)}, B ni A n The corresponding i-th predicted flight route, Q n A n The corresponding take-off and landing point weight, P n = {P n1 , P n2 , ..., P ni , ..., P nI(n)}, P ni Refers to B ni The corresponding route weight, JS n A n The corresponding node betweenness, JL n A n The corresponding clustering coefficient, D n ={D n1 , D n2 , ..., D ni , ..., D nI(n)}, D ni Refers to B ni The corresponding route abnormality degree, E n ={E n1 , E n2 ,……,E ni ,……,E nI(n)}, E ni Refers to Bni The corresponding route length, F n ={F n1 , F n2 , ..., F ni , ..., F nI(n)}, F ni Refers to B ni The corresponding second traffic flow, n=1, 2, ..., N, N refers to the total number of initial drone take-off and landing points, i=1, 2, ..., I(n), I(n) refers to the total number of predicted flight routes corresponding to the nth initial drone take-off and landing point.

[0019] The priority acquisition module is used to obtain the candidate priority set HX = {HX1, HX2, ..., HX5} according to A, B, Q, P, JS, JL and the preset initial weight set QZ = {QZ1, QZ2, QZ3, QZ4, QZ5} n , ..., HX N}, where QZ1 refers to the preset weight of the take-off and landing point weight, QZ2 refers to the preset weight of the route weight, QZ3 refers to the preset weight of the number of routes, QZ4 refers to the preset weight of the node betweenness, QZ5 refers to the preset weight of the clustering coefficient, and A n Corresponding candidate priority HX n Meet the following conditions:

[0020] HX n =QZ1×Q n +QZ2×∑ i＝1 I(n) (P ni / I(n))+QZ3×I(n)+QZ4×JS n +QZ5×JL n .

[0021] The take-off and landing point screening module is used to screen M candidate drone take-off and landing points from N initial drone take-off and landing points according to HX.

[0022] The first landing and take-off point scoring module is used to obtain the candidate landing and take-off point prediction score set FZ = {FZ1, FZ2, ..., FZ n , ..., FZ N}, where A n Corresponding candidate take-off and landing point prediction score FZ n Meet the following conditions:

[0023] FZ n =β1×(Σ i＝1 I(n) e^(-D ni ))+β2×(∑ n＝1 N (Σi＝1 I(n) E ni ))+β3×(∑ n＝1 N (Σ i＝1 I(n) F ni )), where β1

[0024] It refers to the preset abnormal weight, β2 refers to the preset route length weight, and β3 refers to the preset second traffic weight.

[0025] The second take-off and landing point scoring module is used to obtain the screening scores FS of M candidate drone take-off and landing points according to FZ, where FS meets the following conditions:

[0026] FS=(Σ n＝1 N (FZ n -FZ0) 2 / N, where the candidate take-off and landing point prediction mean FZ0 ​​= Σ n＝1 N (FZ n ) / N.

[0027] Candidate priority determination module, used if FS>FS 0 , then update QZ, and use the updated QZ to replace the preset initial weight set, and repeat step S2000 until FS ≤ FS 0 , get FS≤FS 0 The candidate priority set HX when FS≤FS 0 The candidate priority set HX at this time is used to screen the target UAV take-off and landing point from the N initial UAV take-off and landing points, FS 0 Refers to the preset screening score threshold.

[0028] The present invention also provides a non-transitory computer-readable storage medium, in which at least one instruction or at least one program is stored, and the at least one instruction or at least one program is loaded and executed by a processor to implement the above-mentioned method for determining the candidate priority of drone take-off and landing points based on a complex network model.

[0029] The present invention also provides an electronic device, comprising a processor and the above-mentioned non-transitory computer-readable storage medium.

[0030] The present invention has at least the following beneficial effects: according to A, B, Q, P, JS, JL and the preset initial weight set QZ = {QZ1, QZ2, QZ3, QZ4, QZ5}, a candidate priority set HX = {HX1, HX2, ..., HX n , ..., HX N}, according to HX, M candidate drone take-off and landing points are selected from the N initial drone take-off and landing points, and according to D, E and F, the candidate take-off and landing point prediction score set FZ = {FZ1, FZ2, ..., FZ n , ..., FZ N}, according to FZ, obtain the screening score FS of M candidate drone take-off and landing points, if FS>FS 0 , then update QZ, and use the updated QZ to replace the preset initial weight set, and repeat the above steps until FS ≤ FS 0 , get FS≤FS 0 The candidate priority set HX at this time is used to screen the target UAV take-off and landing point from the N initial UAV take-off and landing points. It can be seen that by measuring the differences between the candidate take-off and landing point prediction scores between the M candidate UAV take-off and landing points, the screening scores FS of the M candidate UAV take-off and landing points are obtained to characterize the screening accuracy of the M candidate UAV take-off and landing points, which is used as the basis for updating the initial weight set and the candidate priority set to screen the target UAV take-off and landing point and provide a setting location for the UAV take-off and landing facilities, thereby achieving the effect of balancing the distribution efficiency and distribution intensity of the UAVs corresponding to each target UAV take-off and landing point, thereby improving the overall work efficiency of the UAVs. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0032] Figure 1 A flowchart of a method for determining the priority of candidate take-off and landing points for unmanned aerial vehicles based on a complex network model provided in Example 1 of the present invention.

[0033] Figure 2 A flowchart of an execution computer program of a complex network model construction system for selecting the site of a UAV take-off and landing point provided in the second embodiment of the present invention.

[0034] Figure 3 A flowchart of a method for generating an initial site selection for a UAV take-off and landing point based on multiple factors provided in Example 3 of the present invention.

[0035] Figure 4 A flowchart of an execution computer program of a UAV take-off and landing point site selection system based on the take-off and landing point candidate priorities provided in Embodiment 4 of the present invention.

[0036] Figure 5A schematic diagram of the structure of a device for determining the priority of candidate take-off and landing points for unmanned aerial vehicles based on a complex network model provided in Example 5 of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0038] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It is understood that, where appropriate, the above-mentioned terms used to distinguish similar objects can be interchanged so that the present invention can also implement other embodiments other than the above-mentioned illustrated embodiments or described embodiments. In addition, the terms "including" and "having" and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0039] Embodiment 1

[0040] The first embodiment provides a method for determining the priority of candidate take-off and landing points for unmanned aerial vehicles based on a complex network model. The method for determining the priority of candidate take-off and landing points for unmanned aerial vehicles includes the following steps: Figure 1 As shown:

[0041] S1000, obtain the initial drone take-off and landing point set A = {A1, A2, ..., A n , ..., A N}、Predicted flight route set B={B1,B2,……,B n , ..., B N}、The take-off and landing point weight set Q={Q1,Q2,……,Q n , ..., Q N}、Route weight set P={P1,P2,……,P n , ..., P N}、Node betweenness set JS={JS1,JS2,……,JS n , ..., JS N}、Clustering coefficient set JL={JL1,JL2,……,JL n , ..., JL N}、Route abnormality degree set D={D1,D2,……,D n , ..., D N}、Route length set E={E1,E2,……,E n ,……,E N} and the second traffic flow set F = {F1, F2, ..., F n , ..., F N}, where A n refers to the nth initial drone take-off and landing point, B n = {B n1 , B n2 , ..., B ni , ..., B nI(n)}, B ni A n The corresponding i-th predicted flight route, Q n A n The corresponding take-off and landing point weight, P n = {P n1 , P n2 , ..., P ni , ..., P nI(n)}, P ni Refers to B ni The corresponding route weight, JS n A n The corresponding node betweenness, JL n A n The corresponding clustering coefficient, D n ={D n1 , D n2 , ..., D ni , ..., D nI(n)}, D ni Refers to B ni The corresponding route abnormality degree, E n ={E n1 , E n2 ,……,E ni ,……,E nI(n)}, E ni Refers to B ni The corresponding route length, F n ={F n1 , F n2 , ..., F ni , ..., F nI(n)}, F ni Refers to B ni The corresponding second traffic flow, n=1, 2, ..., N, N refers to the total number of initial drone take-off and landing points, i=1, 2, ..., I(n), I(n) refers to the total number of predicted flight routes corresponding to the nth initial drone take-off and landing point.

[0042] Among them, the initial UAV take-off and landing point may refer to a take-off and landing point that can be used as the layout location of the UAV take-off and landing facilities in a geographical area where the UAV take-off and landing facilities need to be sited. There are predicted flight routes suitable for UAV flights to transport goods between the initial UAV take-off and landing points. The location of the initial UAV take-off and landing point and the trajectory of the predicted flight route can be determined by the implementer based on actual conditions.

[0043] The take-off and landing point weight can represent the possibility of setting up drone take-off and landing facilities at the corresponding initial drone take-off and landing point, the route weight can represent the possibility of the predicted flight route as the flight path of the drone, and the node betweenness can represent the influence of the corresponding initial drone take-off and landing point relative to other initial drone take-off and landing points. The clustering coefficient can represent the degree of clustering between the corresponding initial drone take-off and landing point and other initial drone take-off and landing points.

[0044] The degree of route abnormality can be used to characterize the degree of damage to buildings, human life and property when a drone encounters a flight abnormality in the corresponding predicted flight route. The first traffic flow can refer to the traffic flow within a certain range corresponding to the initial drone take-off and landing point, which can be used to characterize the demand for drone transportation of goods at the initial drone take-off and landing point. The average of the first traffic flow of the two initial drone take-off and landing points corresponding to each predicted flight route can be used as the second traffic flow of the corresponding predicted flight route to characterize the degree of guarantee of the corresponding predicted flight route for drone transportation of goods.

[0045] In a specific implementation, S1000 further includes the following steps:

[0046] S1100, obtain the take-off and landing point location importance set H = {H1, H2, ..., H n , ..., H N}, the first traffic flow set C = {C1, C2, ..., C n , ..., C N} and the first building interference level set G = {G1, G2, ..., G n , ..., G N}, where H n A n The corresponding take-off and landing point location importance, C n A n The corresponding first traffic flow, G n A n The corresponding first building interference level.

[0047] In a specific implementation, S1100 further includes the following steps:

[0048] S1110, obtain the position coordinate set Z={Z1, Z2, . . . , Z n , ..., Z N}, where Z n A n The corresponding position coordinates.

[0049] S1120, based on B and Z, obtain the take-off and landing point location importance set H corresponding to A = {H1, H2, ..., H n , ..., H N}, where A n The corresponding take-off and landing point position importance H n Meet the following conditions:

[0050] H n =(2 / ((1+e^(-I(n)))-1))×(e^(-∑ j＝1 N d(Z n , Z j )) / N), where e is the natural index, d(Z n , Z j ) refers to Z n and Z j The distance between.

[0051] Among them, the location coordinates may refer to the geographical location coordinates of the corresponding initial UAV take-off and landing point, including the longitude and latitude of the corresponding position of the initial UAV take-off and landing point.

[0052] The more the number of other initial UAV take-off and landing points connected to the nth initial UAV take-off and landing point and the closer the distance between the nth initial UAV take-off and landing point and the other initial UAV take-off and landing points, the higher the importance of the take-off and landing point position corresponding to the nth initial UAV take-off and landing point.

[0053] In the above, based on the number of other initial UAV take-off and landing points connected to the initial UAV take-off and landing point, and the distance between the initial UAV take-off and landing point and other initial UAV take-off and landing points, the importance of the take-off and landing point position corresponding to each initial UAV take-off and landing point is measured, which serves as the basis for obtaining the take-off and landing point weight corresponding to each initial UAV take-off and landing point, thereby improving the rationality of the take-off and landing point weight.

[0054] S1200: Based on H, C and G, obtain the take-off and landing point weight set Q = {Q1, Q2, ..., Q n , ..., Q N}, where A n The corresponding take-off and landing point weight Q n Meet the following conditions:

[0055] Q n =α1×H n +α2×Cn +α3×e^(-G n ), wherein α1 refers to the preset first position weight, α2 refers to the preset first flow weight, and α3 refers to the preset building interference weight.

[0056] Among them, the importance of the take-off and landing point location can be used to characterize the importance of the location of the corresponding initial drone take-off and landing point in the drone transportation task. The first building interference degree can be used to characterize the interference degree of the buildings near the initial drone take-off and landing point on the drone take-off and landing when the drone takes off and lands at the corresponding initial drone take-off and landing point.

[0057] The greater the importance of the take-off and landing point corresponding to the initial UAV take-off and landing point, the greater the first traffic flow, and the smaller the first building interference level, the higher the possibility of setting up UAV take-off and landing facilities at the initial UAV take-off and landing point.

[0058] Therefore, the importance of the take-off and landing point location and the first traffic flow are positively correlated with the take-off and landing point weight, and the first building interference degree is negatively correlated with the take-off and landing point weight. Based on H, C and G, the take-off and landing point weight set Q can be obtained as the basis for constructing a complex network model and selecting the location of the drone take-off and landing point.

[0059] S1300, based on C, obtain the second traffic flow set F corresponding to B = {F1, F2, ..., F n , ..., F N}, where B n The corresponding second traffic flow list F n ={F n1 , F n2 , ..., F ni , ..., F nI(n)}, F ni Equal to B ni The average value of the first traffic flow at the corresponding two initial drone take-off and landing points.

[0060] Among them, the average value of the first traffic flow of the two initial drone take-off and landing points corresponding to each predicted flight route can be used as the second traffic flow of the corresponding predicted flight route to characterize the degree of security of the corresponding predicted flight route for drone transportation of goods, as the basis for obtaining the route weight corresponding to the predicted flight route.

[0061] The average value of the take-off and landing point position importance of the two initial UAV take-off and landing points corresponding to each predicted flight route can represent the route position importance of the corresponding predicted flight route, which serves as the basis for obtaining the route weight corresponding to the predicted flight route.

[0062] S1400, based on B and H, obtain the route location importance set K corresponding to B = {K1, K2, ..., K n , ..., K N}, where B n The corresponding route location importance list K n ={K n1 , K n2 , ..., K ni , ..., K nI(n)}, K ni Equal to B ni The average value of the importance of the take-off and landing point locations of the two corresponding initial UAV take-off and landing points.

[0063] S1500, based on B, D, E, F and K, obtain the route weight set P corresponding to B = {P1, P2, ..., P n , ..., P N}, where P n = {P n1 , P n2 , ..., P ni , ..., P nI(n)}, B ni The corresponding route weight P ni Meet the following conditions:

[0064] P ni =β1×e^(-D ni )+β2×E ni / ((∑ n＝1 N (Σ i＝1 I(n) E ni )) / (Σ n＝1 N I(n)))+β3×F ni +β4×K ni , where β

[0065] 1 refers to the preset abnormal weight, β2 refers to the preset route length weight, β3 refers to the preset second traffic weight, and β4 refers to the preset second position weight.

[0066] Among them, the smaller the route abnormality corresponding to the predicted flight route, the longer the route length, the more the second traffic flow, and the higher the importance of the route position, the higher the possibility that the predicted flight route is the flight path of the drone.

[0067] Therefore, the route weight corresponding to the predicted flight route is negatively correlated with the degree of route abnormality, and positively correlated with the route length, the second traffic flow, and the importance of the route location. Based on B, D, E, F, and K, the route weight set P corresponding to B can be obtained as the basis for constructing a complex network model and selecting the location of the drone take-off and landing point.

[0068] Among them, the specific values ​​of α1, α2, α3, β1, β2, β3 and β4 can be set by the implementer according to actual conditions.

[0069] S1600, based on A, B, Q and P, takes the initial UAV take-off and landing point as the node, the predicted flight route as the edge, the take-off and landing point weight corresponding to the initial UAV take-off and landing point as the node value of the corresponding node, and the route weight corresponding to the predicted flight route as the edge value of the corresponding edge, to obtain the complex network model T, wherein the complex network model T is used to screen the target UAV take-off and landing point from A.

[0070] Among them, the complex network model T is used to screen the target UAV take-off and landing points from A. The target UAV take-off and landing points can be used as the location for setting up UAV take-off and landing facilities to provide support for UAV delivery logistics and other tasks.

[0071] S1700: According to T and E, obtain the shortest path between any two nodes in T.

[0072] S1800, determining the number of shortest paths passing through the nodes corresponding to the nth initial UAV take-off and landing point as the node betweenness JS corresponding to the nth initial UAV take-off and landing point n .

[0073] S1900, traverse n = 1, 2, ..., N, and obtain the node betweenness set JS = {JS1, JS2, ..., JS n , ..., JS N}.

[0074] Among them, the more shortest paths passing through the nodes corresponding to the nth initial UAV take-off and landing point, the greater the influence of the nth initial UAV take-off and landing point, and the higher the possibility of setting up UAV take-off and landing facilities at the initial UAV take-off and landing point.

[0075] S1910, based on A and B, obtain the clustering coefficient set JL = {JL1, JL2, ..., JL n , ..., JL N}, where JL n Meet the following conditions:

[0076] JL n =ZS n / (I(n)×(I(n)-1)), where, ZSn It refers to the total number of routes between the I(n) initial UAV take-off and landing points connected to the nth initial UAV take-off and landing point.

[0077] In a specific implementation, S1100 further includes the following steps:

[0078] S1130, obtain the first building density set J 1 ={J 1 1. J 1 2, ..., J 1 n , ..., J 1 N} and the first building height set J 2 ={J 2 1. J 2 2, ..., J 2 n , ..., J 2 N}, where J 1 n A n The first building density within the corresponding first preset range, J 2 n ={J 2 n1 , J 2 n2 , ..., J 2 nu , ..., J 2 nU(n)}, J 2 nu A n The height of the u-th first building in the corresponding first preset range, J 2 nu >J 0 , J 0 It refers to the preset building height threshold, u=1, 2, ..., U(n), U(n) refers to A n The total number of first buildings within the corresponding first preset range.

[0079] S1140, according to J 1 and J 2 , obtain the first building interference level set G = {G1, G2, ..., G n , ..., G N}, where A n The corresponding first building interference level G n Meet the following conditions:

[0080] G n =(2 / ((1+e^(-J1 n ))-1))×(γ1×(Σ u＝1 U(n) J 2 nu ) / U(n)+γ2×max(J 2 n )), wherein max() refers to the maximum value function, γ1 refers to the preset first height weight, and γ2 refers to the preset second height weight.

[0081] Among them, the first preset range may refer to a ground area range with the initial UAV take-off and landing point as the center and the first preset length as the radius or side length. There are several buildings within the first preset range. When the density and height of the buildings within the first preset range are different, different degrees of interference will be caused to the UAV when taking off and landing at the corresponding initial UAV take-off and landing point. Therefore, the first building density and the first building height within the first preset range corresponding to the initial UAV take-off and landing point are used to measure the interference degree of the first building corresponding to the initial UAV take-off and landing point, thereby improving the accuracy of obtaining the take-off and landing point weight.

[0082] Among them, the specific values ​​of the first preset length and the preset building height threshold can be set by the implementer according to actual conditions.

[0083] Since the buildings within the first preset range have different heights, lower buildings have less interference with the take-off and landing of the drone. Therefore, this embodiment sets a preset building height threshold J 0 The buildings within the first preset range are screened, and the higher buildings are screened out to characterize the interference degree of the first building corresponding to the initial UAV take-off and landing point, so as to reduce the influence of more low buildings on the interference degree of the first building, thereby improving the measurement accuracy of the interference degree of the first building.

[0084] The specific values ​​of γ1 and γ2 can be set by the implementer according to actual conditions.

[0085] In a specific implementation, S1000 further includes the following steps:

[0086] S1920, obtain the second building density set R 1 = {R 1 1. R 1 2, ..., R 1 n , ..., R 1 N}, the second building height set R 2 = {R 2 1. R 2 2, ..., R 2 n , ..., R2 N}、Population density set Y 1 = {Y 1 1. Y 1 2, ..., Y 1 n , ..., Y 1 N} and the population category importance set Y 2 = {Y 2 1. Y 2 2, ..., Y 2 n , ..., Y 2 N}, where R 1 n = {R 1 n1 , R 1 n2 , ..., R 1 ni , ..., R 1 nI(n)}, R 1 ni Refers to B ni The second building density within the corresponding second preset range, R 2 n = {R 2 n1 , R 2 n2 , ..., R 2 ni , ..., R 2 nI(n)}, R 2 ni = {R 2 ni1 , R 2 ni2 , ..., R 2 niv , ..., R 2 niV(ni)}, R 2 niv Refers to B ni The height of the vth second building in the corresponding second preset range, Y 1 n = {Y 1 n1 , Y 1 n2 , ..., Y 1 ni , ..., Y 1 nI(n)},Y 1ni Refers to B ni The corresponding population density within the second preset range, Y 2 n = {Y 2 n1 , Y 2 n2 , ..., Y 2 ni , ..., Y 2 nI(n)},Y 2 ni = {Y 2 ni1 , Y 2 ni2 , ..., Y 2 niw , ..., Y 2 niW(ni)},Y 2 niw Refers to B ni The importance of the category corresponding to the wth population in the corresponding second preset range, v = 1, 2, ..., V (n), V (n) refers to A n The total number of second buildings within the corresponding second preset range, w = 1, 2, ..., W (ni), W (ni) refers to B ni The total population within the corresponding second preset range.

[0087] S1930, according to R 1 , R 2 , Y 1 and Y 2 , we get the route anomaly degree set D = {D1, D2, ..., D n , ..., D N}, where D n ={D n1 , D n2 , ..., D ni , ..., D nI(n)}, B ni The corresponding route abnormality degree D ni Meet the following conditions:

[0088] D ni =δ1×D 1 ni +δ2×D 2 ni , where δ1 refers to the preset building weight, δ2 refers to the preset first population weight,

[0089] D 1 ni =(2 / ((1+e^(-R1 ni ))-1))×(γ1×(Σ v＝1 V(n) R 2 niv ) / V(n)+γ2×max(R 2 ni )),

[0090] D 2 ni =(2 / ((1+e^(-Y 1 ni ))-1))×(ε1×(Σ w＝1 W(n) Y 2 niv ) / W(n)+ε2×max(Y 2 ni )), ε1 refers to the preset second population weight, and ε2 refers to the preset third population weight.

[0091] The importance of a population category may refer to the degree of security required for the population of the corresponding category, or may refer to the severity of the security impact on the population of the corresponding category.

[0092] The second preset range may refer to a three-dimensional airspace area range that is based on the trajectory of the predicted flight route and is a second preset length away from each trajectory point of the predicted flight route. There are a number of buildings and a number of people within the second preset range. When the density, height, population density and importance of population categories of the buildings within the second preset range are different, it will affect the degree of damage to buildings, human life and property when an abnormal flight situation occurs on the corresponding predicted flight route.

[0093] Therefore, the route abnormality degree corresponding to each predicted flight route is measured according to the second building density, the second building height, the population density and the importance of the population category corresponding to the predicted flight route, thereby improving the accuracy of the route weight.

[0094] Among them, the specific values ​​of δ1, δ2, ε1 and ε2 can be set by the implementer according to actual conditions.

[0095] S2000: According to A, B, Q, P, JS, JL and the preset initial weight set QZ = {QZ1, QZ2, QZ3, QZ4, QZ5}, a candidate priority set HX = {HX1, HX2, ..., HX n , ..., HX N}, where QZ1 refers to the preset weight of the take-off and landing point weight, QZ2 refers to the preset weight of the route weight, QZ3 refers to the preset weight of the number of routes, QZ4 refers to the preset weight of the node betweenness, QZ5 refers to the preset weight of the clustering coefficient, and A n Corresponding candidate priority HX n Meet the following conditions:

[0096] HX n =QZ1×Q n +QZ2×∑ i＝1 I(n) (P ni / I(n))+QZ3×I(n)+QZ4×JS n +QZ5×JL n .

[0097] Among them, the greater the take-off and landing point weight corresponding to the initial UAV take-off and landing point, the greater the average route weight corresponding to the predicted flight route connected to the initial UAV take-off and landing point, the more routes of the predicted flight route connected to the initial UAV take-off and landing point, and the greater the node betweenness and clustering coefficient corresponding to the initial UAV take-off and landing point, the higher the possibility of setting up UAV take-off and landing facilities at the initial UAV take-off and landing point.

[0098] Therefore, in combination with Q, P, JS, JL and the corresponding preset initial weights, the candidate priority corresponding to each initial UAV take-off and landing point is obtained, which indicates that the corresponding initial UAV take-off and landing point is determined as the target UAV take-off and landing point to set the priority of the UAV take-off and landing facilities.

[0099] Among them, the specific values ​​of QZ1, QZ2, QZ3, QZ4 and QZ5 can be set by the implementer according to actual conditions.

[0100] As mentioned above, the take-off and landing point weights, average route weights, predicted number of flight routes, node betweenness and clustering coefficient corresponding to the initial UAV take-off and landing points, and the preset initial weights are comprehensively considered to measure the candidate priority of each initial UAV take-off and landing point to be determined as the target UAV take-off and landing point, which improves the accuracy of the candidate priority and thus improves the accuracy of the screening of target UAVs.

[0101] S3000, based on HX, selects M candidate UAV take-off and landing points from N initial UAV take-off and landing points.

[0102] In a specific implementation, S3000 further includes the following steps:

[0103] S3100, determine the initial UAV take-off and landing points corresponding to the largest M candidate priorities in HX as candidate UAV take-off and landing points.

[0104] S4000: Based on D, E and F, obtain the candidate take-off and landing point prediction score set FZ = {FZ1, FZ2, ..., FZ n , ..., FZ N}, where A n Corresponding candidate take-off and landing point prediction score FZ n Meet the following conditions:

[0105] FZ n =β1×(Σ i＝1 I(n) e^(-D ni ))+β2×(∑ n＝1 N (Σ i＝1 I(n) E ni ))+β3×(∑ n＝1 N (Σ i＝1 I(n) F ni )), where β1

[0106] refers to the preset abnormal weight, β2 refers to the preset route length weight, β3 refers to the preset second flow weight, and e refers to the natural constant.

[0107] Among them, the smaller the route abnormality degree of all predicted flight routes corresponding to the candidate UAV take-off and landing points, the longer the route length, and the larger the second traffic flow, it means that the candidate UAV take-off and landing points have a lower demand for UAV take-off and landing facilities, and the higher the possibility of setting up UAV take-off and landing facilities at the candidate UAV take-off and landing points. Correspondingly, the larger the candidate take-off and landing point prediction score corresponding to the candidate UAV take-off and landing points.

[0108] As described above, based on D, E and F, the possibility of setting up UAV take-off and landing facilities at the candidate UAV take-off and landing points is measured, and then the candidate take-off and landing point prediction score set FZ is obtained, which improves the accuracy of the candidate take-off and landing point prediction scores.

[0109] S5000: According to FZ, the screening scores FS of M candidate drone take-off and landing points are obtained, where FS meets the following conditions:

[0110] FS=(Σ n＝1 N (FZ n -FZ0) 2 / N, where the candidate take-off and landing point prediction mean FZ0 ​​= Σ n＝1 N (FZ n ) / N.

[0111] In order to ensure that the delivery efficiency and delivery intensity of the drones at each take-off and landing point are as balanced as possible, this embodiment hopes that the prediction scores of the candidate take-off and landing points between the selected target drone take-off and landing points are as consistent as possible.

[0112] In the above, by measuring the differences between the candidate take-off and landing point prediction scores between the M candidate drone take-off and landing points, the screening scores FS of the M candidate drone take-off and landing points are obtained to characterize the screening accuracy of the M candidate drone take-off and landing points, which is used as the basis for updating the initial weight set and the candidate priority set, and the screened take-off and landing points are used as the target drone take-off and landing points to provide setting locations for drone take-off and landing facilities, thereby achieving the effect of balancing the distribution efficiency and distribution intensity of the drones corresponding to each target drone take-off and landing point, thereby improving the overall work efficiency of the drones.

[0113] S6000, if FS>FS 0 , then update QZ, and use the updated QZ to replace the preset initial weight set, and repeat step S2000 until FS ≤ FS 0 , get FS≤FS 0 The candidate priority set HX when FS≤FS 0 The candidate priority set HX at this time is used to screen the target UAV take-off and landing point from the N initial UAV take-off and landing points, FS 0 Refers to the preset screening score threshold.

[0114] Among them, the preset screening score threshold FS 0 The specific value can be set by the implementer according to the actual situation.

[0115] The candidate priority corresponding to each initial UAV take-off and landing point is obtained by comprehensively considering the take-off and landing point weights, average route weights, predicted number of flight routes, node betweenness and clustering coefficient corresponding to the initial UAV take-off and landing point, as well as the preset initial weights. Therefore, the accuracy of the preset initial weights affects the accuracy of the candidate priority, and thus affects the rationality of the screening of candidate UAV take-off and landing points.

[0116] The larger the screening score FS of the M candidate drone take-off and landing points, the greater the difference in delivery efficiency and delivery intensity between the M candidate drone take-off and landing points. Therefore, when FS>FS 0 Update QZ when the default initial weight set is replaced by the updated QZ, and repeat step S2000 to obtain new M candidate drone take-off and landing points until FS ≤ FS 0 , and obtain FS≤FS 0 The candidate priority set HX at this time is used to screen the target UAV take-off and landing points, thereby improving the consistency of the delivery efficiency and delivery intensity between the target UAV take-off and landing points, and improving the screening accuracy of the target UAV take-off and landing points.

[0117] In a specific implementation, S6000 further includes the following steps:

[0118] S6100, set FS≤FS 0 The M candidate UAV take-off and landing points at that time are determined as M key take-off and landing points.

[0119] S6200, obtain the preset prediction model and the initial UAV take-off and landing point code set A 1 ={A 1 1. A 1 2, ..., A 1 n , ..., A 1 N}、Preset key take-off and landing point code set GJ={GJ1,GJ2,……,GJ μ , ..., GJ M} and preset reference vector YS=(1, 1, ..., 1, 0, 0, ..., 0), where A 1 n A n The corresponding initial take-off and landing point code, GJ μ It refers to the key take-off and landing point code corresponding to the μth key take-off and landing point, μ=1, 2, ..., M, M refers to the total number of key take-off and landing points, the first Щ elements in YS are 1, and the last N-Щ elements are 0, Щ refers to the preset number of target UAV take-off and landing points.

[0120] S6300: Input A, B, Q, P and JS into a preset prediction model to obtain a prediction priority ranking vector PX=(PX1, PX2, ..., PX n , ..., PX N ), where PX n It refers to the initial UAV take-off and landing point with the predicted priority of n.

[0121] S6400, perform vector multiplication on PX and YS to obtain the take-off and landing point reference vector CK 1 =(PX1, PX2, ..., PX σ , ..., PX Щ , 0, 0, ..., 0), where PX σ It refers to the initial UAV take-off and landing point with the prediction priority of σth, PX Щ It refers to the initial UAV take-off and landing point with the predicted priority of the Щth position, σ=1, 2, …, Щ.

[0122] S6500, according to CK 1 , A 1 and GJ, get CK 1The corresponding first take-off and landing point coding reference vector BM 1 =(BM1, BM2, ..., BM σ , ..., BM Щ , 0, 0, ..., 0), where BM σ PX σ The corresponding first take-off and landing point code, BM Щ PX Щ The corresponding first take-off and landing point code.

[0123] S6600, transposed vector of GJ and BM 1 Multiply them to obtain the second take-off and landing point coding reference vector BM 2 .

[0124] S6700, for BM 2 Perform binarization processing to obtain the third take-off and landing point coding reference vector BM 3 .

[0125] S6800, according to BM 3 And the preset encoding threshold YZ, the prediction loss Loss = ρ × (YZ-SUM (BM 3 )), where ρ is the preset loss coefficient, SUM(BM 3 ) means BM 3 The sum of all elements in .

[0126] S6900, update the parameters of the preset prediction model according to Loss until Loss=0, and obtain the target prediction model.

[0127] S61000, input A, B, Q, P and JS into the target prediction model to obtain the target priority ranking vector.

[0128] S61100, determine the first Щ initial UAV take-off and landing points in the target priority sorting vector as the target UAV take-off and landing points.

[0129] As described above, according to A, B, Q, P, JS, JL and the preset initial weight set QZ = {QZ1, QZ2, QZ3, QZ4, QZ5}, the candidate priority set HX = {HX1, HX2, ..., HX n , ..., HX N}, according to HX, M candidate drone take-off and landing points are selected from the N initial drone take-off and landing points, and according to D, E and F, the candidate take-off and landing point prediction score set FZ = {FZ1, FZ2, ..., FZ n , ..., FZ N}, according to FZ, obtain the screening score FS of M candidate drone take-off and landing points, if FS>FS0 , then update QZ, and use the updated QZ to replace the preset initial weight set, and repeat the above steps until FS ≤ FS 0 , FS≤FS 0 The M candidate UAV take-off and landing points at the time are determined as the target UAV take-off and landing points. It can be seen that by measuring the differences between the candidate take-off and landing point prediction scores between the M candidate UAV take-off and landing points, the screening scores FS of the M candidate UAV take-off and landing points are obtained to characterize the screening accuracy of the M candidate UAV take-off and landing points, which is used as the basis for updating the initial weight set and the candidate priority set to screen the target UAV take-off and landing points and provide a setting location for the UAV take-off and landing facilities, thereby achieving the effect of balancing the distribution efficiency and distribution intensity of the UAVs corresponding to each target UAV take-off and landing point, and improving the overall work efficiency of the UAVs.

[0130] Embodiment 2

[0131] This second embodiment provides a complex network model construction system for selecting the location of drone take-off and landing points. Figure 2 As shown, the complex network model construction system for selecting the take-off and landing points of unmanned aerial vehicles includes a processor and a memory storing a computer program, and the memory also stores an initial set of take-off and landing points of unmanned aerial vehicles A = {A1, A2, ..., A n , ..., A N}、Predicted flight route set B={B1,B2,……,B n , ..., B N}、The importance set of take-off and landing point locations H = {H1, H2, ..., H n , ..., H N}, the first traffic flow set C = {C1, C2, ..., C n , ..., C N}, the first building interference level set G = {G1, G2, ..., G n , ..., G N}、Route abnormality degree set D={D1,D2,……,D n , ..., D N}、Route length set E={E1,E2,……,E n ,……,E N}, where A n refers to the nth initial drone take-off and landing point, B n = {B n1 , B n2 , ..., B ni , ..., B nI(n)}, B ni A n The corresponding i-th predicted flight route, H n An The corresponding take-off and landing point location importance, C n A n The corresponding first traffic flow, G n A n The corresponding first building interference level, D n ={D n1 , D n2 , ..., D ni , ..., D nI(n)}, D ni Refers to B ni The corresponding route abnormality degree is used to characterize the degree of damage to the outside world when the UAV encounters a flight abnormality on the corresponding predicted flight route. n ={E n1 , E n2 ,……,E ni ,……,E nI(n)}, E ni Refers to B ni The corresponding route length, n = 1, 2, ..., N, N refers to the total number of initial UAV take-off and landing points, i = 1, 2, ..., I(n), I(n) refers to the total number of predicted flight routes corresponding to the nth initial UAV take-off and landing point.

[0132] When the computer program is executed by a processor, the following steps are performed:

[0133] S1, based on H, C and G, obtain the take-off and landing point weight set Q = {Q1, Q2, ..., Q n , ..., Q N}, where A n The corresponding take-off and landing point weight Q n Meet the following conditions:

[0134] Q n =α1×H n +α2×C n +α3×e^(-G n ), wherein α1 refers to the preset first position weight, α2 refers to the preset first flow weight, and α3 refers to the preset building interference weight.

[0135] The specific values ​​of α1, α2 and α3 can be set by the implementer according to actual conditions.

[0136] In the above, the importance of the take-off and landing point location, the influence of the first traffic flow and the first building interference level on the possibility of setting up drone take-off and landing facilities at the initial drone take-off and landing point are measured, and the take-off and landing point weights corresponding to each initial drone take-off and landing point are obtained, which improves the accuracy of the take-off and landing point weights, thereby improving the accuracy of the construction of the complex network model and the rationality of the site selection of the drone take-off and landing points.

[0137] S2, according to C, obtain the second traffic flow set F corresponding to B = {F1, F2, ..., F n , ..., F N}, where B n The corresponding second traffic flow list F n ={F n1 , F n2 , ..., F ni , ..., F nI(n)}, F ni Equal to B ni The average value of the first traffic flow at the corresponding two initial drone take-off and landing points.

[0138] S3, based on B and H, obtain the route location importance set K corresponding to B = {K1, K2, ..., K n , ..., K N}, where B n The corresponding route location importance list K n ={K n1 , K n2 , ..., K ni , ..., K nI(n)}, K ni Equal to B ni The average value of the importance of the take-off and landing point locations of the two corresponding initial UAV take-off and landing points.

[0139] S4, based on B, D, E, F and K, obtain the route weight set P corresponding to B = {P1, P2, ..., P n , ..., P N}, where P n = {P n1 , P n2 , ..., P ni , ..., P nI(n)}, B ni The corresponding route weight P ni Meet the following conditions:

[0140] P ni =β1×e^(-D ni )+β2×E ni / ((∑ n＝1 N (Σi＝1 I(n) E ni )) / (Σ n＝1 N I(n)))+β3×F ni +β4×K ni , where β

[0141] 1 refers to the preset abnormal weight, β2 refers to the preset route length weight, β3 refers to the preset second traffic weight, and β4 refers to the preset second position weight.

[0142] As mentioned above, the influence of the route abnormality, route length, second traffic flow and route location importance on the possibility of the predicted flight route as the flight path of the UAV is measured, and the route weight corresponding to each predicted flight route is obtained, which improves the accuracy of the route weight, thereby improving the accuracy of the construction of the complex network model and the rationality of the site selection of the UAV take-off and landing points.

[0143] S5, according to A, B, Q and P, the initial UAV take-off and landing point is used as the node, the predicted flight route is used as the edge, the take-off and landing point weight corresponding to the initial UAV take-off and landing point is used as the node value of the corresponding node, and the route weight corresponding to the predicted flight route is used as the edge value of the corresponding edge to obtain the complex network model T, wherein the complex network model T is used to screen the target UAV take-off and landing point from A.

[0144] Among them, the complex network model T is used to screen the target UAV take-off and landing points from A. The target UAV take-off and landing points can be used as the location for setting up UAV take-off and landing facilities to provide support for UAV delivery logistics and other tasks.

[0145] As mentioned above, based on H, C and G, we can obtain the take-off and landing point weight set Q = {Q1, Q2, ..., Q n , ..., Q N}, weigh the influence of the importance of the take-off and landing point location, the first traffic flow and the first building interference degree on the possibility of setting up drone take-off and landing facilities at the initial drone take-off and landing point, obtain the take-off and landing point weight corresponding to each initial drone take-off and landing point, improve the accuracy of the take-off and landing point weight, and obtain the second traffic flow set F corresponding to B according to C = {F1, F2, ..., F n , ..., F N}, to characterize the degree of guarantee of the corresponding predicted flight route for the transportation of goods by drones. According to B and H, the importance set of the route position corresponding to B is obtained K = {K1, K2, ..., K n , ..., K N}, according to B, D, E, F and K, obtain the route weight set P corresponding to B = {P1, P2, ..., P n , ..., P N}, measure the impact of the route abnormality, route length, second traffic flow and route location importance on the possibility of the predicted flight route as the flight path of the UAV, obtain the route weight corresponding to each predicted flight route, and improve the accuracy of the route weight. According to A, B, Q and P, the initial UAV take-off and landing point is taken as the node, the predicted flight route is taken as the edge, the take-off and landing point weight corresponding to the initial UAV take-off and landing point is taken as the node value of the corresponding node, and the route weight corresponding to the predicted flight route is taken as the edge value of the corresponding edge. The complex network model T is obtained and used to screen the target UAV take-off and landing point from A, which improves the construction accuracy of the complex network model T, and then improves the screening accuracy of the target UAV take-off and landing point, that is, improves the rationality of the site selection of the UAV take-off and landing point.

[0146] Embodiment 3

[0147] This embodiment 3 provides a method for generating an initial location for a drone take-off and landing point based on multiple factors, such as Figure 3 As shown, the method for generating the initial site selection of the take-off and landing points of the UAV based on multiple factors includes the following steps:

[0148] S10, obtain Θ original UAV take-off and landing points, several predicted flight routes corresponding to each original UAV take-off and landing point, and the route length, second traffic flow and route abnormality degree corresponding to each predicted flight route, wherein the second traffic flow refers to the average value of the first traffic flow corresponding to the two original UAV take-off and landing points corresponding to the predicted flight route, and the route abnormality degree is used to characterize the degree of damage to the outside world when a UAV encounters a flight abnormality on the corresponding predicted flight route, and Θ is an integer greater than 0.

[0149] In a specific implementation, S10 further includes the following steps:

[0150] S11, obtaining map data corresponding to Θ original drone take-off and landing points and the first traffic flow corresponding to each original drone take-off and landing point.

[0151] S12, input the map data corresponding to Θ original drone take-off and landing points and the first traffic flow corresponding to Θ original drone take-off and landing points into the preset drone route setting model, and obtain several predicted flight routes corresponding to each original drone take-off and landing point.

[0152] Among them, implementers can adopt path planning algorithms, such as A* search algorithm, Dijkstra algorithm, genetic algorithm, etc., on the basis of determining the original UAV take-off and landing points, according to the map data and the first traffic flow corresponding to the original UAV take-off and landing points, quickly generate UAV routes between the original UAV take-off and landing points as predicted flight routes for the site selection task of UAV take-off and landing points.

[0153] Among them, map data may include factors such as terrain data, obstacle data, and weather condition data.

[0154] S20, according to a number of predicted flight routes corresponding to each original UAV take-off and landing point, the route length corresponding to each predicted flight route and the second traffic flow, the number of predicted flight routes, the total route length and the total traffic volume corresponding to each original UAV take-off and landing point are obtained.

[0155] Among them, for any original UAV take-off and landing point, the sum of the route lengths of all predicted flight routes corresponding to the current original UAV take-off and landing point is determined as the total route length corresponding to the current original UAV take-off and landing point, and the sum of the second traffic flows of all predicted flight routes corresponding to the current original UAV take-off and landing point is determined as the total traffic volume corresponding to the current original UAV take-off and landing point, which serves as the basis for screening UAV take-off and landing points.

[0156] S30, according to the predicted number of flight routes, total route length, total traffic volume, preset route number threshold, preset route length threshold and preset traffic threshold corresponding to each original UAV take-off and landing point, select ζ take-off and landing points to be merged from Θ original UAV take-off and landing points, where 0<ζ<Θ.

[0157] Among them, the specific values ​​of the preset route quantity threshold, the preset route length threshold and the preset traffic threshold can be set by the implementer according to actual conditions.

[0158] In a specific implementation, S30 further includes the following steps:

[0159] S31, for any original UAV take-off and landing point, if the number of predicted flight routes corresponding to the current original UAV take-off and landing point is less than the preset route number threshold, the total route length is less than the preset route length threshold, and the total traffic volume is less than the preset flow threshold, then the current original UAV take-off and landing point is determined as the take-off and landing point to be merged.

[0160] S32, traverse Θ original UAV take-off and landing points, and obtain ζ take-off and landing points to be merged.

[0161] Among them, the more predicted flight routes corresponding to the current original UAV take-off and landing point, the longer the total length of the routes, and the larger the total traffic volume, it means that the original UAV take-off and landing point has a lower demand for UAV take-off and landing facilities, and the possibility of setting up UAV take-off and landing facilities at the original UAV take-off and landing point is higher. Conversely, it means that the demand for UAV take-off and landing facilities is higher, and the possibility of setting up UAV take-off and landing facilities at the original UAV take-off and landing point is lower.

[0162] Therefore, for the original drone take-off and landing points with low demand and possibility for setting up drone take-off and landing facilities, they can be used as take-off and landing points to be merged and merged into the take-off and landing points connected to them with high demand and possibility for setting up drone take-off and landing facilities, thereby improving the correlation between the take-off and landing points, and ensuring that the possibility of setting up drone take-off and landing facilities at the merged take-off and landing points is high, and improving the coverage and setting rationality of drone take-off and landing facilities by merging take-off and landing points.

[0163] S40, according to all predicted flight routes corresponding to each take-off and landing point to be merged, obtain n original UAV take-off and landing points connected to each take-off and landing point to be merged, where n is an integer greater than 0.

[0164] S50, for any take-off and landing point to be merged, any original UAV take-off and landing point connected to the current take-off and landing point to be merged is used as the intermediate take-off and landing point, and the route length between the current take-off and landing point to be merged and the other n-1 original UAV take-off and landing points and the intermediate take-off and landing point, as well as the route length between the intermediate take-off and landing point and the other n-1 original UAV take-off and landing points, the route length priority corresponding to the intermediate take-off and landing point is obtained.

[0165] In a specific implementation, S50 further includes the following steps:

[0166] S51, determining the average of the route lengths between the current take-off and landing point to be merged and the other n-1 original UAV take-off and landing points as the first route length.

[0167] S52, determining the average of the route lengths between the middle take-off and landing point and the other n-1 original UAV take-off and landing points as the second route length.

[0168] S53: Determine the sum of the route length between the current take-off and landing point to be merged and the intermediate take-off and landing point and the second route length as the third route length.

[0169] S54, determining the ratio of the first route length to the third route length as the route length priority corresponding to the intermediate take-off and landing point.

[0170] Among them, the first route length can represent the average distance from the current take-off and landing point to be merged directly to the other n-1 original drone take-off and landing points, the second route length can represent the average distance from the current intermediate take-off and landing point to the other n-1 original drone take-off and landing points, and the third route length can represent the average distance from the intermediate take-off and landing point as a transit point, the current take-off and landing point to be merged first to the intermediate take-off and landing point and then to the other n-1 original drone take-off and landing points. If the third route length is less than the corresponding first route length, it means that compared with the current take-off and landing point to be merged directly to the other n-1 original drone take-off and landing points, when the intermediate take-off and landing point is used as the transit point from the current take-off and landing point to the other n-1 original drone take-off and landing points, the drone's flight distance is shorter, which means that the intermediate take-off and landing point has a higher priority than the current take-off and landing point to be merged, and after the take-off and landing point to be merged is merged into the intermediate take-off and landing point, the impact on the drone flight demand from the current take-off and landing point to the other n-1 original drone take-off and landing points is small.

[0171] Correspondingly, the larger the ratio of the first route length to the third route length is, the higher the priority of the route length corresponding to the intermediate take-off and landing point is.

[0172] In the above, according to the route lengths between the current take-off and landing point to be merged and the other n-1 original UAV take-off and landing points and the intermediate take-off and landing points, as well as the route lengths between the intermediate take-off and landing point and the other n-1 original UAV take-off and landing points, the route length priority corresponding to the intermediate take-off and landing point is obtained to characterize the priority of the intermediate take-off and landing point compared to the current take-off and landing point to be merged, and the degree of influence on the UAV flight demand from the current take-off and landing point to be merged to the other n-1 original UAV take-off and landing points after the take-off and landing point to be merged is merged into the intermediate take-off and landing point, which serves as the basis for merging the take-off and landing points to be merged, thereby improving the rationality of the take-off and landing point merger.

[0173] S60, according to the route abnormality degree of the predicted flight route between the current take-off and landing point to be merged and the other n-1 original UAV take-off and landing points and the intermediate take-off and landing point, and the route abnormality degree of the predicted flight route between the intermediate take-off and landing point and the other n-1 original UAV take-off and landing points, obtain the route abnormality degree priority corresponding to the intermediate take-off and landing point.

[0174] In a specific implementation, S60 further includes the following steps:

[0175] S61, determining the average value of the route abnormality degree between the current take-off and landing point to be merged and the other n-1 original UAV take-off and landing points as the first route abnormality degree.

[0176] S62, determining the average value of the route abnormality degree between the middle take-off and landing point and the other n-1 original UAV take-off and landing points as the second route abnormality degree.

[0177] S63: Determine the sum of the route abnormality degree between the current take-off and landing point to be merged and the intermediate take-off and landing point and the second route abnormality degree as a third route abnormality degree.

[0178] S64, determining the ratio of the first route abnormality level to the third route abnormality level as the route abnormality level priority corresponding to the intermediate take-off and landing point.

[0179] Among them, the first route abnormality degree can characterize the average abnormality degree of the drone from the current take-off and landing point to be merged directly to the other η-1 original drone take-off and landing points, the second route abnormality degree can characterize the average abnormality degree of the drone from the current intermediate take-off and landing point directly to the other η-1 original drone take-off and landing points, and the third route abnormality degree can characterize the average abnormality degree of the drone from the current take-off and landing point to be merged to the intermediate take-off and landing point and then to the other η-1 original drone take-off and landing points. If the third route abnormality degree is less than the corresponding first route abnormality degree, it means that compared with the drone directly from the current take-off and landing point to be merged to the other η-1 original drone take-off and landing points, when the intermediate take-off and landing point is used as the transfer point from the current take-off and landing point to be merged to the other η-1 original drone take-off and landing points, the flight abnormality degree of the drone is lower, which can indicate that the intermediate take-off and landing point has a higher priority than the current take-off and landing point to be merged, and after the take-off and landing point to be merged is merged to the intermediate take-off and landing point, the impact on the safety of the drone is small.

[0180] Correspondingly, the greater the ratio of the first route abnormality level to the third route abnormality level, the higher the priority of the route abnormality level corresponding to the intermediate take-off and landing point.

[0181] In the above, according to the route abnormality degree between the current take-off and landing point to be merged and the other n-1 original UAV take-off and landing points and the intermediate take-off and landing point, as well as the route abnormality degree between the intermediate take-off and landing point and the other n-1 original UAV take-off and landing points, the route abnormality degree priority corresponding to the intermediate take-off and landing point is obtained to characterize the priority of the intermediate take-off and landing point compared with the current take-off and landing point to be merged, and the impact on the safety of the UAV after merging the take-off and landing point to be merged into the intermediate take-off and landing point, as the basis for merging the take-off and landing points to be merged, the rationality of the take-off and landing point merger is improved.

[0182] S70, obtaining the merge priority corresponding to the current intermediate take-off and landing point according to the route length priority and the route abnormality priority.

[0183] In a specific implementation, S70 further includes the following steps:

[0184] S71, obtaining a preset route length weight and a preset abnormality degree weight.

[0185] S72, obtaining the merge priority corresponding to the current intermediate take-off and landing point according to the route length priority, the preset route length weight, the route anomaly degree priority and the preset anomaly degree weight.

[0186] Among them, the first product of the route length priority and the preset route length weight, as well as the second product of the route anomaly degree priority and the preset anomaly degree weight are calculated, and the sum of the first product and the second product is determined as the merging priority corresponding to the current intermediate take-off and landing point, representing the priority of merging the current take-off and landing point to be merged into the current intermediate take-off and landing point.

[0187] Among them, the specific values ​​of the preset route length weight and the preset abnormality degree weight can be set by the implementer according to the actual situation.

[0188] As described above, in combination with the route length priority, the preset route length weight, the route anomaly degree priority and the preset anomaly degree weight, the merging priority corresponding to the current intermediate take-off and landing point is obtained to characterize the priority of merging the current take-off and landing point to be merged into the current intermediate take-off and landing point, which serves as the basis for merging the current take-off and landing point to be merged, thereby improving the rationality of the take-off and landing point merging.

[0189] S80, traverse the n original UAV take-off and landing points connected to the current take-off and landing point to be merged, and obtain n merging priorities.

[0190] S90: Merge the current take-off and landing point to be merged with the original UAV take-off and landing point corresponding to the largest merging priority, and determine the merged take-off and landing point as the initial UAV take-off and landing point.

[0191] Among them, the take-off and landing points to be merged, which have low demand for drone take-off and landing facilities and low possibility of setting up drone take-off and landing facilities, are merged into other take-off and landing points, which have high demand for drone take-off and landing facilities and high possibility of setting up drone take-off and landing facilities, and the initial drone take-off and landing points are obtained as the setting locations of drone take-off and landing facilities, thereby generating a batch of initial points that can be used for drone take-off and landing site site selection, so that the setting locations of drone take-off and landing facilities match the take-off and landing point requirements, thereby improving the relative coverage and setting rationality of drone take-off and landing facilities.

[0192] In the above, according to the predicted number of flight routes, total route length and total traffic volume corresponding to each original UAV take-off and landing point obtained, combined with the preset route number threshold, the preset route length threshold and the preset traffic threshold, z take-off and landing points to be merged are screened out from θ original UAV take-off and landing points, and according to all predicted flight routes corresponding to each take-off and landing point to be merged, n original UAV take-off and landing points connected to each take-off and landing point to be merged are obtained. For any take-off and landing point to be merged, any original UAV take-off and landing point connected to the current take-off and landing point to be merged is used as the intermediate take-off and landing point, and according to the route length and route abnormality between the current take-off and landing point to be merged and the other n-1 original UAV take-off and landing points and the intermediate take-off and landing point, as well as the route length and route abnormality between the intermediate take-off and landing point and the other n-1 original UAV take-off and landing points, the route length priority and route abnormality corresponding to the intermediate take-off and landing point are obtained. According to the route length priority and the route abnormality priority, the merging priority corresponding to the current intermediate take-off and landing point is obtained, the current take-off and landing point to be merged is merged with the original drone take-off and landing point corresponding to the largest merging priority, and the merged take-off and landing point is determined as the initial drone take-off and landing point. It can be seen that the take-off and landing points to be merged with a low possibility of setting up drone take-off and landing facilities, whose predicted number of flight routes is less than the preset route number threshold, the total route length is less than the preset route length threshold, and the total traffic volume is less than the preset flow threshold, are merged into other take-off and landing points with a higher possibility of setting up drone take-off and landing facilities, thereby generating a batch of initial points that can be used for drone take-off and landing site selection, so that when drone take-off and landing facilities are set at the initial drone take-off and landing point, they can match the take-off and landing point requirements, thereby improving the relative coverage and setting rationality of drone take-off and landing facilities.

[0193] Embodiment 4

[0194] This fourth embodiment provides a drone take-off and landing point site selection system based on the take-off and landing point candidate priority, such as Figure 4 As shown, the drone take-off and landing point selection system based on the take-off and landing point candidate priority includes a processor and a memory storing a computer program, and the memory also stores a preset prediction model, an initial drone take-off and landing point set A = {A1, A2, ..., A n , ..., A N}、Initial UAV take-off and landing point code set A 1 ={A 1 1. A 1 2, ..., A 1 n , ..., A 1 N}、Predicted flight route set B={B1,B2,……,B n , ..., B N}、The take-off and landing point weight set Q={Q1,Q2,……,Qn , ..., Q N}、Route weight set P={P1,P2,……,P n , ..., P N}、Node betweenness set JS={JS1,JS2,……,JS n , ..., JS N}、Preset key take-off and landing point code set GJ={GJ1,GJ2,……,GJ μ , ..., GJ M} and preset reference vector YS=(1, 1, ..., 1, 0, 0, ..., 0), where A n refers to the nth initial drone take-off and landing point, A 1 n A n The corresponding initial take-off and landing point code, B n = {B n1 , B n2 , ..., B ni , ..., B nI(n)}, B ni A n The corresponding i-th predicted flight route, Q n A n The corresponding take-off and landing point weight, P n = {P n1 , P n2 , ..., P ni , ..., P nI(n)}, P ni Refers to B ni The corresponding route weight, JS n A n The corresponding node betweenness, GJ μ It refers to the key take-off and landing point code corresponding to the preset μth key take-off and landing point, n=1, 2, ..., N, N refers to the total number of initial UAV take-off and landing points, i=1, 2, ..., I(n), I(n) refers to the total number of predicted flight routes corresponding to the nth initial UAV take-off and landing point, μ=1, 2, ..., M, M refers to the total number of preset key take-off and landing points, the first Щ elements in YS are 1, and the last N-Щ elements are 0, Щ refers to the preset number of target UAV take-off and landing points.

[0195] Among them, the initial drone take-off and landing point code can be a code pre-set by the implementer according to the actual situation, which is used to identify the initial drone take-off and landing point, such as 1, 2, 3...

[0196] The preset key take-off and landing point codes may be codes pre-set by implementers according to actual conditions, and are used to identify the key take-off and landing points, such as -1, -2, -3...

[0197] The preset prediction model is used to extract and map the features of the input unordered initial UAV take-off and landing points, predicted flight routes, node betweenness, take-off and landing point weights, and route weight data, and output the predicted priority ranking vector of the ordered initial UAV take-off and landing points.

[0198] The preset reference vector YS = (1, 1, ..., 1, 0, 0, ..., 0), the first N elements in YS are the value 1, and the last N-N elements are the value 0, which is used to process the prediction priority sorting vector and extract the first N elements in the prediction priority sorting vector as the evaluation of the matching degree between the sorting result of the prediction priority sorting vector and the premise condition.

[0199] When the computer program is executed by a processor, the following steps are performed:

[0200] S100, input A, B, Q, P and JS into a preset prediction model, and obtain a prediction priority ranking vector PX=(PX1, PX2, ..., PX n , ..., PX N ), where PX n It refers to the initial UAV take-off and landing point with the predicted priority of n.

[0201] Among them, the higher the priority ranking, the higher the possibility that the corresponding initial UAV take-off and landing point will be used as the target UAV take-off and landing point for setting up UAV take-off and landing facilities.

[0202] In a specific implementation manner, the memory further stores a clustering coefficient set JL={JL1, JL2, . . . , JL n , ..., JL N}, JL n A n The corresponding clustering coefficient, M preset key take-off and landing points are obtained through the following steps:

[0203] S2000: According to A, B, Q, P, JS, JL and the preset initial weight set QZ = {QZ1, QZ2, QZ3, QZ4, QZ5}, a candidate priority set HX = {HX1, HX2, ..., HX n , ..., HX N}, where QZ1 refers to the preset weight of the take-off and landing point weight, QZ2 refers to the preset weight of the route weight, QZ3 refers to the preset weight of the number of routes, QZ4 refers to the preset weight of the node betweenness, QZ5 refers to the preset weight of the clustering coefficient, and A n Corresponding candidate priority HX n Meet the following conditions:

[0204] HX n=QZ1×Q n +QZ2×∑ i＝1 I(n) (P ni / I(n))+QZ3×I(n)+QZ4×JS n +QZ5×JL n .

[0205] S3000, based on HX, selects M candidate UAV take-off and landing points from N initial UAV take-off and landing points.

[0206] S4000: Based on D, E and F, obtain the candidate take-off and landing point prediction score set FZ = {FZ1, FZ2, ..., FZ n , ..., FZ N}, where A n Corresponding candidate take-off and landing point prediction score FZ n Meet the following conditions:

[0207] FZ n =β1×(Σ i＝1 I(n) e^(-D ni ))+β2×(∑ n＝1 N (Σ i＝1 I(n) E ni ))+β3×(∑ n＝1 N (Σ i＝1 I(n) F ni )), where β1

[0208] refers to the preset abnormal weight, β2 refers to the preset route length weight, β3 refers to the preset second flow weight, and e refers to the natural constant.

[0209] S5000: According to FZ, the screening scores FS of M candidate drone take-off and landing points are obtained, where FS meets the following conditions:

[0210] FS=(Σ n＝1 N (FZ n -FZ0) 2 / N, where the candidate take-off and landing point prediction mean FZ0 ​​= Σ n＝1 N (FZ n ) / N.

[0211] S7000, if FS>FS 0 , then update QZ, and use the updated QZ to replace the preset initial weight set, and repeat step S2000 until FS ≤ FS 0, FS≤FS 0 The M candidate UAV take-off and landing points at the time are determined as M key take-off and landing points, FS 0 Refers to the preset screening score threshold.

[0212] S200, perform vector multiplication on PX and YS to obtain the take-off and landing point reference vector CK 1 =(PX1, PX2, ..., PX σ , ..., PX Щ , 0, 0, ..., 0), where PX σ It refers to the initial UAV take-off and landing point with the prediction priority of σth, PX Щ It refers to the initial UAV take-off and landing point with the predicted priority of the Щth position, σ=1, 2, …, Щ.

[0213] Among them, based on the task of screening out the target UAV take-off and landing points, PX and YS are vector-multiplied to obtain the take-off and landing point reference vector CK 1 =(PX1, PX2, ..., PX σ , ..., PX Щ , 0, 0, ..., 0), in order to extract the initial UAV take-off and landing points with the top prediction priority as the basis for measuring the sorting effect of the prediction priority sorting vector PX.

[0214] S300, according to CK 1 , A 1 and GJ, get CK 1 The corresponding first take-off and landing point coding reference vector BM 1 =(BM1, BM2, ..., BM σ , ..., BM Щ , 0, 0, ..., 0), where BM σ PX σ The corresponding first take-off and landing point code, BM Щ PX Щ The corresponding first take-off and landing point code.

[0215] In a specific implementation, S300 further includes the following steps:

[0216] S310, if PX σ If it is a key take-off and landing point, PX σ The corresponding key take-off and landing point code is determined as PX σ The corresponding first take-off and landing point code.

[0217] S320, if PX σ If it is not a key take-off and landing point, PX σ The corresponding initial take-off and landing point code is determined as PX σThe corresponding first take-off and landing point code.

[0218] S330, traverse PX1, PX2, ..., PX σ , ..., PX Щ , get CK 1 The corresponding first take-off and landing point coding reference vector BM 1 =(BM1, BM2, ..., BM σ , ..., BM Щ , 0, 0,…, 0).

[0219] Among them, according to the type of the extracted top 10 initial UAV take-off and landing points, that is, key take-off and landing points or non-key take-off and landing points, the first take-off and landing point code corresponding to each initial UAV take-off and landing point in the top 10 initial UAV take-off and landing points is obtained, so as to judge the matching degree between the top 10 initial UAV take-off and landing points and the prerequisites through coding, so as to provide a data basis for measuring the sorting effect of the prediction priority sorting vector PX.

[0220] In a specific implementation, the key take-off and landing point code is less than 0, and the initial take-off and landing point code is greater than 0.

[0221] As mentioned above, the first initial UAV take-off and landing points are converted into corresponding take-off and landing point codes, which provides a data basis for judging the matching degree between the first initial UAV take-off and landing points and the prerequisites through coding, and then measuring the sorting effect of the prediction priority sorting vector PX.

[0222] S400, transposed vector of GJ and BM 1 Multiply them to obtain the second take-off and landing point coding reference vector BM 2 .

[0223] Among them, the matrix dimension of GJ is 1×M, and the transposed vector of GJ (GJ) T The dimension is M×1, BM 1 The matrix dimension is 1×N, then (GJ) T ×BM 1 The second take-off and landing point coding reference vector BM 2 The matrix dimension is M×N.

[0224] Among them, (GJ) T The codes of each key take-off and landing point are less than 0, BM 1 The first landing point code of the corresponding key landing point is less than 0, BM 1 If the first landing point code of the corresponding non-critical landing point is greater than 0, then BM 1 The first take-off and landing point code of the corresponding key take-off and landing point is (GJ) T The product of the key take-off and landing point codes in is greater than 0, BM 1The corresponding non-critical take-off and landing points in (GJ) T The product of the codes of the key take-off and landing points in is less than 0, so it can be achieved through BM 2 The number of elements greater than 0 in , reverse BM 1 The number of corresponding key take-off and landing points in , which further characterizes the matching degree between the corresponding prediction priority sorting vector PX and the prerequisite.

[0225] For example, BM 2 The number of elements greater than 0 in is M×Amount, so we can determine that BM 1 The corresponding number of key take-off and landing points is Amount.

[0226] S500, for BM 2 Perform binarization processing to obtain the third take-off and landing point coding reference vector BM 3 .

[0227] In a specific implementation, S500 further includes the following steps:

[0228] S510, if BM 2 If the element in the μth row and σth column is greater than 0, then BM 3 The element in the μth row and the σth column is set to 1.

[0229] S520, if BM 2 If the element in the μth row and σth column is less than 0, then BM 3 The elements in the μth row and the σth column are set to 0.

[0230] S530, BM 3 The elements in columns Щ+1 to N are set to 0.

[0231] S530, traverse μ=1, 2, ..., M and σ=1, 2, ..., Щ, and obtain the third take-off and landing point coding reference vector BM 3 .

[0232] Among them, BM 2 Binarization is performed, and elements greater than 0 are set to 1, and elements less than or equal to 0 are set to 0. 3 The data scale is normalized to avoid the influence of the value of each element on the prediction loss, improve the accuracy of the prediction loss, and further improve the accuracy of the judgment of the matching degree between the prediction priority sorting vector PX and the prerequisite.

[0233] S600, according to BM 3 And the preset encoding threshold YZ, the prediction loss Loss = ρ × (YZ-SUM (BM 3 )), where ρ is the preset loss coefficient, SUM(BM3 ) means BM 3 The sum of all elements in .

[0234] Among them, the premise of this embodiment is that the prediction priorities corresponding to the M key take-off and landing points in the final prediction priority sorting vector belong to the top position, SUM(BM 3 ) can characterize BM 3 The number of elements 1 in the BM 1 The number of key take-off and landing points corresponding to , that is, the number of key take-off and landing points in the prediction priority sorting vector PX, and the more the number of key take-off and landing points in the prediction priority sorting vector PX, the higher the degree of matching between the prediction priority sorting vector PX and the prerequisite.

[0235] SUM(BM 3 ) ranges from [0, M 2 ], therefore, based on the prediction loss Loss = ρ × (YZ-SUM (BM 3 )) is used to characterize the matching degree between the prediction priority ranking vector PX and the premise conditions, which serves as the basis for updating the parameters of the prediction model to improve the prediction accuracy of the prediction model.

[0236] In one embodiment, YZ=M 2 .

[0237] In a specific embodiment, p=1000.

[0238] The specific value of ρ can be set by the implementer according to the actual situation, for example, ρ=100, 1000, 10000 and other values, so as to amplify the prediction loss, improve the training effect of the prediction model, and improve the prediction accuracy of the prediction model.

[0239] As mentioned above, the prediction loss is increased to characterize the matching degree between the prediction priority ranking vector PX and the premise, which provides a data basis for updating the parameters of the prediction model to improve the prediction accuracy of the prediction model.

[0240] S700, update the parameters of the preset prediction model according to Loss until Loss=0, and obtain the target prediction model.

[0241] S800, input A, B, Q, P and JS into the target prediction model to obtain the target priority ranking vector.

[0242] S900: Determine the first initial UAV take-off and landing points in the target priority sorting vector as the target UAV take-off and landing points.

[0243] Among them, the target priority sorting vector obtained when Loss = 0 indicates that the number of key take-off and landing points in the predicted priority sorting vector PX is M, which means that the target priority sorting vector matches the prerequisite. Therefore, the first Щ initial UAV take-off and landing points in the target priority sorting vector are determined as the target UAV take-off and landing points, which serve as the setting location of the UAV take-off and landing facilities to provide support for UAV delivery logistics and other tasks.

[0244] As mentioned above, A, B, Q, P and JS are input into the preset prediction model to obtain the prediction priority ranking vector PX = (PX1, PX2, ..., PX n , ..., PX N ), perform vector multiplication on PX and YS to obtain the take-off and landing point reference vector CK 1 =(PX1, PX2, ..., PX σ , ..., PX Щ , 0, 0, ..., 0), according to CK 1 , A 1 and GJ, get CK 1 The corresponding first take-off and landing point coding reference vector BM 1 =(BM1, BM2, ..., BM σ , ..., BM Щ , 0, 0, ..., 0), the transposed vector of GJ and BM 1 Multiply them to obtain the second take-off and landing point coding reference vector BM 2 , for BM 2 Perform binarization processing to obtain the third take-off and landing point coding reference vector BM 3 , according to BM 3 And the preset encoding threshold YZ, the prediction loss Loss = ρ × (YZ-SUM (BM 3 )), update the parameters of the preset prediction model according to Loss until Loss = 0, obtain the target prediction model, input A, B, Q, P and JS into the target prediction model, obtain the target priority sorting vector, and determine the first Щ initial UAV take-off and landing points in the target priority sorting vector as the target UAV take-off and landing points. It can be seen that through BM 3 The number of elements 1 in the , reverse BM 2 The number of elements greater than 0 in , and then infer BM 1 The number of key take-off and landing points corresponding to the preset coding thresholds YZ and BM 3 The number of elements 1 in the formula is used to construct the prediction loss, and the parameters of the prediction model are updated accordingly, which improves the matching degree between the target priority sorting vector and the prerequisites, thereby improving the rationality of the target UAV take-off and landing points.

[0245] Embodiment 5

[0246] This embodiment 5 provides a device for determining the priority of candidate take-off and landing points of unmanned aerial vehicles based on a complex network model, such as Figure 5 As shown, the device for determining the priority of candidate take-off and landing points of unmanned aerial vehicles includes:

[0247] The data acquisition module 51 is used to obtain the initial drone take-off and landing point set A = {A1, A2, ..., A n , ..., A N}、Predicted flight route set B={B1,B2,……,B n , ..., B N}、The take-off and landing point weight set Q={Q1,Q2,……,Q n , ..., Q N}、Route weight set P={P1,P2,……,P n , ..., P N}、Node betweenness set JS={JS1,JS2,……,JS n , ..., JS N}、Clustering coefficient set JL={JL1,JL2,……,JL n , ..., JL N}、Route abnormality degree set D={D1,D2,……,D n , ..., D N}、Route length set E={E1,E2,……,E n ,……,E N} and the second traffic flow set F = {F1, F2, ..., F n , ..., F N}, where A n refers to the nth initial drone take-off and landing point, B n = {B n1 , B n2 , ..., B ni , ..., B nI(n)}, B ni A n The corresponding i-th predicted flight route, Q n A n The corresponding take-off and landing point weight, P n = {P n1 , P n2 , ..., P ni , ..., P nI(n)}, P ni Refers to B ni The corresponding route weight, JS n A n The corresponding node betweenness, JL n An The corresponding clustering coefficient, D n ={D n1 , D n2 , ..., D ni , ..., D nI(n)}, D ni Refers to B ni The corresponding route abnormality degree, E n ={E n1 , E n2 ,……,E ni ,……,E nI(n)}, E ni Refers to B ni The corresponding route length, F n ={F n1 , F n2 , ..., F ni , ..., F nI(n)}, F ni Refers to B ni The corresponding second traffic flow, n=1, 2, ..., N, N refers to the total number of initial drone take-off and landing points, i=1, 2, ..., I(n), I(n) refers to the total number of predicted flight routes corresponding to the nth initial drone take-off and landing point.

[0248] The priority acquisition module 52 is used to acquire a candidate priority set HX = {HX1, HX2, ..., HX3} according to A, B, Q, P, JS, JL and a preset initial weight set QZ = {QZ1, QZ2, QZ3, QZ4, QZ5}. n , ..., HX N}, where QZ1 refers to the preset weight of the take-off and landing point weight, QZ2 refers to the preset weight of the route weight, QZ3 refers to the preset weight of the number of routes, QZ4 refers to the preset weight of the node betweenness, QZ5 refers to the preset weight of the clustering coefficient, and A n Corresponding candidate priority HX n Meet the following conditions:

[0249] HX n =QZ1×Q n +QZ2×∑ i＝1 I(n) (P ni / I(n))+QZ3×I(n)+QZ4×JS n +QZ5×JL n .

[0250] The take-off and landing point screening module 53 is used to screen M candidate drone take-off and landing points from N initial drone take-off and landing points according to HX.

[0251] The first landing and take-off point scoring module 54 is used to obtain the candidate landing and take-off point prediction score set FZ = {FZ1, FZ2, ..., FZ n , ..., FZ N}, where A n Corresponding candidate take-off and landing point prediction score FZ n Meet the following conditions:

[0252] FZ n =β1×(Σ i＝1 I(n) e^(-D ni ))+β2×(∑ n＝1 N (Σ i＝1 I(n) E ni ))+β3×(∑ n＝1 N (Σ i＝1 I(n) F ni )), where β1

[0253] It refers to the preset abnormal weight, β2 refers to the preset route length weight, and β3 refers to the preset second traffic weight.

[0254] The second take-off and landing point scoring module 55 is used to obtain the screening scores FS of M candidate drone take-off and landing points according to FZ, wherein FS meets the following conditions:

[0255] FS=(Σ n＝1 N (FZ n -FZ0) 2 / N, where the candidate take-off and landing point prediction mean FZ0 ​​= Σ n＝1 N (FZ n ) / N.

[0256] Candidate priority determination module 56, for if FS>FS 0 , then update QZ, and use the updated QZ to replace the preset initial weight set, and repeatedly execute the priority acquisition module 52 until FS ≤ FS 0 , get FS≤FS 0 The candidate priority set HX when FS≤FS 0 The candidate priority set HX at this time is used to screen the target UAV take-off and landing point from the N initial UAV take-off and landing points, FS 0 Refers to the preset screening score threshold.

[0257] In a specific implementation, the data acquisition module 51 further includes:

[0258] The first data acquisition submodule is used to obtain the take-off and landing point location importance set H = {H1, H2, ..., H n , ..., H N}, the first traffic flow set C = {C1, C2, ..., C n , ..., C N} and the first building interference level set G = {G1, G2, ..., G n , ..., G N}, where H n A n The corresponding take-off and landing point location importance, C n A n The corresponding first traffic flow, G n A n The corresponding first building interference level.

[0259] The take-off and landing point weight acquisition submodule is used to obtain the take-off and landing point weight set Q = {Q1, Q2, ..., Q n , ..., Q N}, where A n The corresponding take-off and landing point weight Q n Meet the following conditions:

[0260] Q n =α1×H n +α2×C n +α3×e^(-G n ), wherein α1 refers to the preset first position weight, α2 refers to the preset first flow weight, and α3 refers to the preset building interference weight.

[0261] The traffic flow acquisition submodule is used to acquire the second traffic flow set F corresponding to B according to C = {F1, F2, ..., F n , ..., F N}, where B n The corresponding second traffic flow list F n ={F n1 , F n2 , ..., F ni , ..., F nI(n)}, F ni Equal to B ni The average value of the first traffic flow at the corresponding two initial drone take-off and landing points.

[0262] The importance acquisition submodule is used to obtain the route position importance set K corresponding to B according to B and H = {K1, K2, ..., K n , ..., K N}, where B nThe corresponding route location importance list K n ={K n1 , K n2 , ..., K ni , ..., K nI(n)}, K ni Equal to B ni The average value of the importance of the take-off and landing point locations of the two corresponding initial UAV take-off and landing points.

[0263] The route weight acquisition submodule is used to obtain the route weight set P corresponding to B according to B, D, E, F and K = {P1, P2, ..., P n , ..., P N}, where P n = {P n1 , P n2 , ..., P ni , ..., P nI(n)}, B ni The corresponding route weight P ni Meet the following conditions:

[0264] P ni =β1×e^(-D ni )+β2×E ni / ((∑ n＝1 N (Σ i＝1 I(n) E ni )) / (Σ n＝1 N I(n)))+β3×F ni +β4×K ni , where β

[0265] 1 refers to the preset abnormal weight, β2 refers to the preset route length weight, β3 refers to the preset second traffic weight, and β4 refers to the preset second position weight.

[0266] The graph construction submodule is used to obtain the complex network model T based on A, B, Q and P, with the initial UAV take-off and landing points as nodes, the predicted flight routes as edges, the take-off and landing point weights corresponding to the initial UAV take-off and landing points as the node values ​​of the corresponding nodes, and the route weights corresponding to the predicted flight routes as the edge values ​​of the corresponding edges. The complex network model T is used to screen the target UAV take-off and landing points from A.

[0267] The shortest path acquisition submodule is used to obtain the shortest path between any two nodes in T according to T and E.

[0268] The first node betweenness acquisition submodule is used to determine the number of shortest paths passing through the nodes corresponding to the nth initial drone take-off and landing point as the node betweenness JS corresponding to the nth initial drone take-off and landing point n .

[0269] The second node betweenness acquisition submodule is used to traverse n=1, 2, ..., N, and obtain the node betweenness set JS={JS1, JS2, ..., JS n , ..., JS N}.

[0270] The clustering coefficient acquisition submodule is used to obtain the clustering coefficient set JL = {JL1, JL2, ..., JL n , ..., JL N}, where JL n Meet the following conditions:

[0271] JL n =ZS n / (I(n)×(I(n)-1)), where, ZS n It refers to the total number of routes between the I(n) initial UAV take-off and landing points connected to the nth initial UAV take-off and landing point.

[0272] In a specific implementation, the first data acquisition submodule further includes:

[0273] The first data acquisition unit is used to acquire a position coordinate set Z={Z1, Z2, ..., Z n , ..., Z N}, where Z n A n The corresponding position coordinates.

[0274] A position importance acquisition unit is used to acquire the position importance set H corresponding to A according to B and Z, H = {H1, H2, ..., H n , ..., H N}, where A n The corresponding position importance H n Meet the following conditions:

[0275] H n =(2 / ((1+e^(-I(n)))-1))×(e^(-∑ j＝1 N d(Z n , Z j )) / N), where e is the natural index, d(Z n , Z j ) refers to Z n and Z j The distance between.

[0276] In a specific implementation, the first data acquisition submodule further includes:

[0277] The second data acquisition unit is used to acquire the first building density set J 1 ={J 1 1. J 1 2, ..., J 1 n , ..., J 1 N} and the first building height set J 2 ={J 2 1. J 2 2, ..., J 2 n , ..., J 2 N}, where J 1 n A n The first building density within the corresponding first preset range, J 2 n ={J 2 n1 , J 2 n2 , ..., J 2 nu , ..., J 2 nU(n)}, J 2 nu A n The height of the u-th first building in the corresponding first preset range, J 2 nu >J 0 , J 0 It refers to the preset building height threshold, u=1, 2, ..., U(n), U(n) refers to A n The total number of first buildings within the corresponding first preset range.

[0278] The first building interference degree acquisition unit is used to obtain the building interference degree according to J 1 and J 2 , obtain the first building interference level set G = {G1, G2, ..., G n , ..., G N}, where A n The corresponding first building interference level G n Meet the following conditions:

[0279] G n =(2 / ((1+e^(-J 1 n ))-1))×(γ1×(Σu＝1 U(n) J 2 nu ) / U(n)+γ2×max(J 2 n )), wherein max() refers to the maximum value function, γ1 refers to the preset first height weight, and γ2 refers to the preset second height weight.

[0280] In a specific implementation, the data acquisition module 51 further includes:

[0281] The second data acquisition submodule is used to obtain the second building density set R 1 = {R 1 1. R 1 2, ..., R 1 n , ..., R 1 N}, the second building height set R 2 = {R 2 1. R 2 2, ..., R 2 n , ..., R 2 N}、Population density set Y 1 = {Y 1 1. Y 1 2, ..., Y 1 n , ..., Y 1 N} and the population category importance set Y 2 = {Y 2 1. Y 2 2, ..., Y 2 n , ..., Y 2 N}, where R 1 n = {R 1 n1 , R 1 n2 , ..., R 1 ni , ..., R 1 nI(n)}, R 1 ni Refers to B ni The second building density within the corresponding second preset range, R 2 n = {R 2 n1 , R 2 n2, ..., R 2 ni , ..., R 2 nI(n)}, R 2 ni = {R 2 ni1 , R 2 ni2 , ..., R 2 niv , ..., R 2 niV(ni)}, R 2 niv Refers to B ni The height of the vth second building in the corresponding second preset range, Y 1 n = {Y 1 n1 , Y 1 n2 , ..., Y 1 ni , ..., Y 1 nI(n)},Y 1 ni Refers to B ni The corresponding population density within the second preset range, Y 2 n = {Y 2 n1 , Y 2 n2 , ..., Y 2 ni , ..., Y 2 nI(n)},Y 2 ni = {Y 2 ni1 , Y 2 ni2 , ..., Y 2 niw , ..., Y 2 niW(ni)},Y 2 niw Refers to B ni The importance of the category corresponding to the wth population in the corresponding second preset range, v = 1, 2, ..., V (n), V (n) refers to A n The total number of second buildings within the corresponding second preset range, w = 1, 2, ..., W (ni), W (ni) refers to B ni The total population within the corresponding second preset range.

[0282] The route anomaly degree acquisition submodule is used to obtain the route anomaly degree according to R1 , R 2 , Y 1 and Y 2 , we get the route anomaly degree set D = {D1, D2, ..., D n , ..., D N}, where D n ={D n1 , D n2 , ..., D ni , ..., D nI(n)}, B ni The corresponding route abnormality degree D ni Meet the following conditions:

[0283] D ni =δ1×D 1 ni +δ2×D 2 ni , where δ1 refers to the preset building weight, δ2 refers to the preset first population weight,

[0284] D 1 ni =(2 / ((1+e^(-R 1 ni ))-1))×(γ1×(Σ v＝1 V(n) R 2 niv ) / V(n)+γ2×max(R 2 ni )),

[0285] D 2 ni =(2 / ((1+e^(-Y 1 ni ))-1))×(ε1×(Σ w＝1 W(n) Y 2 niv ) / W(n)+ε2×max(Y 2 ni )), ε1 refers to the preset second population weight, and ε2 refers to the preset third population weight.

[0286] In a specific implementation, the take-off and landing point screening module 53 further includes:

[0287] The take-off and landing point screening submodule is used to determine the initial UAV take-off and landing points corresponding to the largest M candidate priorities in HX as candidate UAV take-off and landing points.

[0288] In a specific implementation, the candidate priority determination module 56 further includes:

[0289] The key take-off and landing point determination submodule is used to set FS≤FS 0 The M candidate UAV take-off and landing points at that time are determined as M key take-off and landing points.

[0290] The third data acquisition submodule is used to obtain the preset prediction model and the initial UAV take-off and landing point code set A 1 ={A 1 1. A 1 2, ..., A 1 n , ..., A 1 N}、Preset key take-off and landing point code set GJ={GJ1,GJ2,……,GJ μ , ..., GJ M} and preset reference vector YS=(1, 1, ..., 1, 0, 0, ..., 0), where A 1 n A n The corresponding initial take-off and landing point code, GJ μ It refers to the key take-off and landing point code corresponding to the μth key take-off and landing point, μ=1, 2, ..., M, M refers to the total number of key take-off and landing points, the first Щ elements in YS are 1, and the last N-Щ elements are 0, Щ refers to the preset number of target UAV take-off and landing points.

[0291] The first priority sorting submodule is used to input A, B, Q, P and JS into a preset prediction model to obtain a prediction priority sorting vector PX = (PX1, PX2, ..., PX n , ..., PX N ), where PX n It refers to the initial UAV take-off and landing point with the predicted priority of n.

[0292] The first reference vector acquisition submodule is used to perform vector multiplication on PX and YS to obtain the take-off and landing point reference vector CK 1 =(PX1, PX2, ..., PX σ , ..., PX Щ , 0, 0, ..., 0), where PX σ It refers to the initial UAV take-off and landing point with the prediction priority of σth, PX Щ It refers to the initial UAV take-off and landing point with the predicted priority of the Щth position, σ=1, 2, …, Щ.

[0293] The second reference vector acquisition submodule is used to obtain the reference vector according to CK 1 , A 1 and GJ, get CK 1 The corresponding first take-off and landing point coding reference vector BM 1=(BM1, BM2, ..., BM σ , ..., BM Щ , 0, 0, ..., 0), where BM σ PX σ The corresponding first take-off and landing point code, BM Щ PX Щ The corresponding first take-off and landing point code.

[0294] The third reference vector acquisition submodule is used to obtain the transposed vector of GJ and BM 1 Multiply them to obtain the second take-off and landing point coding reference vector BM 2 .

[0295] The fourth reference vector acquisition submodule is used to obtain BM 2 Perform binarization processing to obtain the third take-off and landing point coding reference vector BM 3 .

[0296] The second reference vector acquisition submodule is used to obtain the reference vector according to BM 3 And the preset encoding threshold YZ, the prediction loss Loss = ρ × (YZ-SUM (BM 3 )), where ρ is the preset loss coefficient, SUM(BM 3 ) means BM 3 The sum of all elements in .

[0297] The parameter updating submodule is used to update the parameters of the preset prediction model according to the Loss until Loss=0, thereby obtaining the target prediction model.

[0298] The second priority sorting submodule is used to input A, B, Q, P and JS into the target prediction model to obtain a target priority sorting vector.

[0299] The target UAV take-off and landing point determination submodule is used to determine the first initial UAV take-off and landing points in the target priority sorting vector as the target UAV take-off and landing points.

[0300] It should be noted that the information interaction, execution process and other contents between the above modules are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0301] Embodiment 6

[0302] Embodiment 6 of the present invention provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the steps:

[0303] S1000, obtain the initial drone take-off and landing point set A = {A1, A2, ..., A n , ..., A N}、Predicted flight route set B={B1,B2,……,B n , ..., B N}、The take-off and landing point weight set Q={Q1,Q2,……,Q n , ..., Q N}、Route weight set P={P1,P2,……,P n , ..., P N}、Node betweenness set JS={JS1,JS2,……,JS n , ..., JS N}、Clustering coefficient set JL={JL1,JL2,……,JL n , ..., JL N}、Route abnormality degree set D={D1,D2,……,D n , ..., D N}、Route length set E={E1,E2,……,E n ,……,E N} and the second traffic flow set F = {F1, F2, ..., F n , ..., F N}, where A n refers to the nth initial drone take-off and landing point, B n = {B n1 , B n2 , ..., B ni , ..., B nI(n)}, B ni A n The corresponding i-th predicted flight route, Q n A n The corresponding take-off and landing point weight, P n = {P n1 , P n2 , ..., P ni , ..., P nI(n)}, P ni Refers to B ni The corresponding route weight, JS n A n The corresponding node betweenness, JL n A n The corresponding clustering coefficient, D n ={D n1 , D n2 , ..., D ni , ..., D nI(n)}, D niRefers to B ni The corresponding route abnormality degree, E n ={E n1 , E n2 ,……,E ni ,……,E nI(n)}, E ni Refers to B ni The corresponding route length, F n ={F n1 , F n2 , ..., F ni , ..., F nI(n)}, F ni Refers to B ni The corresponding second traffic flow, n=1, 2, ..., N, N refers to the total number of initial drone take-off and landing points, i=1, 2, ..., I(n), I(n) refers to the total number of predicted flight routes corresponding to the nth initial drone take-off and landing point.

[0304] S2000: According to A, B, Q, P, JS, JL and the preset initial weight set QZ = {QZ1, QZ2, QZ3, QZ4, QZ5}, a candidate priority set HX = {HX1, HX2, ..., HX n , ..., HX N}, where QZ1 refers to the preset weight of the take-off and landing point weight, QZ2 refers to the preset weight of the route weight, QZ3 refers to the preset weight of the number of routes, QZ4 refers to the preset weight of the node betweenness, QZ5 refers to the preset weight of the clustering coefficient, and A n Corresponding candidate priority HX n Meet the following conditions:

[0305] HX n =QZ1×Q n +QZ2×∑ i＝1 I(n) (P ni / I(n))+QZ3×I(n)+QZ4×JS n +QZ5×JL n .

[0306] S3000, based on HX, selects M candidate UAV take-off and landing points from N initial UAV take-off and landing points.

[0307] S4000: Based on D, E and F, obtain the candidate take-off and landing point prediction score set FZ = {FZ1, FZ2, ..., FZ n , ..., FZ N}, where A n Corresponding candidate take-off and landing point prediction score FZ n Meet the following conditions:

[0308] FZ n =β1×(Σ i＝1 I(n) e^(-D ni ))+β2×(∑ n＝1 N (Σ i＝1 I(n) E ni ))+β3×(∑ n＝1 N (Σ i＝1 I(n) F ni )), where β1

[0309] refers to the preset abnormal weight, β2 refers to the preset route length weight, β3 refers to the preset second flow weight, and e refers to the natural constant.

[0310] S5000: According to FZ, the screening scores FS of M candidate drone take-off and landing points are obtained, where FS meets the following conditions:

[0311] FS=(Σ n＝1 N (FZ n -FZ0) 2 / N, where the candidate take-off and landing point prediction mean FZ0 ​​= Σ n＝1 N (FZ n ) / N.

[0312] S6000, if FS>FS 0 , then update QZ, and use the updated QZ to replace the preset initial weight set, and repeat step S2000 until FS ≤ FS 0 , get FS≤FS 0 The candidate priority set HX when FS≤FS 0 The candidate priority set HX at this time is used to screen the target UAV take-off and landing point from the N initial UAV take-off and landing points, FS 0 Refers to the preset screening score threshold.

[0313] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0314] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0315] Embodiment 7

[0316] A seventh embodiment of the present invention provides an electronic device, which includes a processor and the non-transitory computer-readable storage medium in the sixth embodiment of the present invention.

[0317] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A method for determining the priority of candidate take-off and landing points for unmanned aerial vehicles based on a complex network model, characterized in that: The method for determining the priority of candidate drone take-off and landing points includes: S1000, obtain the initial drone take-off and landing point set A = {A1, A2, ..., A n , ..., A N }、Predicted flight route set B={B1,B2,……,B n , ..., B N }、The take-off and landing point weight set Q={Q1,Q2,……,Q n , ..., Q N }、Route weight set P={P1,P2,……,P n , ..., P N }、Node betweenness set JS={JS1,JS2,……,JS n , ..., JS N }、Clustering coefficient set JL={JL1,JL2,……,JL n , ..., JL N }、Route abnormality degree set D={D1,D2,……,D n , ..., D N }、Route length set E={E1,E2,……,E n ,……,E N } and the second traffic flow set F = {F1, F2, ..., F n , ..., F N }, where A n refers to the nth initial drone take-off and landing point, B n = {B n1 , B n2 , ..., B ni , ..., B nI(n) }, B ni A n The corresponding i-th predicted flight route, Q n A n The corresponding take-off and landing point weight, P n = {P n1 , P n2 , ..., P ni , ..., P nI(n) }, P ni Refers to B ni The corresponding route weight, JS n A n The corresponding node betweenness, JL n A n The corresponding clustering coefficient, D n ={D n1 , D n2 , ..., D ni , ..., D nI(n) }, D ni Refers to B ni The corresponding route abnormality degree, E n ={E n1 , E n2 ,……,E ni ,……,E nI(n) }, E ni Refers to B ni The corresponding route length, F n ={F n1 , F n2 , ..., F ni , ..., F nI(n) }, F ni Refers to B ni The corresponding second traffic flow, n = 1, 2, ..., N, N refers to the total number of initial drone take-off and landing points, i = 1, 2, ..., I (n), I (n) refers to the total number of predicted flight routes corresponding to the nth initial drone take-off and landing point; S2000: According to A, B, Q, P, JS, JL and the preset initial weight set QZ = {QZ1, QZ2, QZ3, QZ4, QZ5}, a candidate priority set HX = {HX1, HX2, ..., HX n , ..., HX N }, where QZ1 refers to the preset weight of the take-off and landing point weight, QZ2 refers to the preset weight of the route weight, QZ3 refers to the preset weight of the number of routes, QZ4 refers to the preset weight of the node betweenness, QZ5 refers to the preset weight of the clustering coefficient, and A n Corresponding candidate priority HX n Meet the following conditions: HX n =QZ1×Q n +QZ2×∑ i＝1 I(n) (P ni / I(n))+QZ3×I(n)+QZ4×JS n +QZ5×JL n ; S3000, based on HX, selects M candidate UAV take-off and landing points from N initial UAV take-off and landing points; S4000: Based on D, E and F, obtain the candidate take-off and landing point prediction score set FZ = {FZ1, FZ2, ..., FZ n , ..., FZ N }, where A n Corresponding candidate take-off and landing point prediction score FZ n Meet the following conditions: FZ n =β1×(Σ i＝1 I(n) e^(-D ni ))+β2×(∑ n＝1 N (S i＝1 I(n) E ni ))+β3×(∑ n＝1 N (S i＝1 I(n) F ni )), among them,b 1 refers to the preset abnormal weight, β2 refers to the preset route length weight, β3 refers to the preset second flow weight, and e refers to the natural constant; S5000: According to FZ, the screening scores FS of M candidate drone take-off and landing points are obtained, where FS meets the following conditions: FS=(Σ n＝1 N (FZ n -FZ0) 2 / N, where the candidate take-off and landing point prediction mean FZ0 ​​= Σ n＝1 N (FZ n ) / N; S6000, if FS>FS 0 , then update QZ, and use the updated QZ to replace the preset initial weight set, and repeat step S2000 until FS ≤ FS 0 , get FS≤FS 0 The candidate priority set HX when FS≤FS 0 The candidate priority set HX at this time is used to screen the target UAV take-off and landing point from the N initial UAV take-off and landing points, FS 0 Refers to the preset screening score threshold.

2. The method for determining the priority of candidate take-off and landing points of unmanned aerial vehicles based on a complex network model according to claim 1 is characterized in that: S1000 also includes the following steps: S1100, obtain the take-off and landing point location importance set H = {H1, H2, ..., H n , ..., H N }, the first traffic flow set C = {C1, C2, ..., C n , ..., C N } and the first building interference level set G = {G1, G2, ..., G n , ..., G N }, where H n A n The corresponding take-off and landing point location importance, C n A n The corresponding first traffic flow, G n A n The corresponding first building interference level; S1200: Based on H, C and G, obtain the take-off and landing point weight set Q = {Q1, Q2, ..., Q n , ..., Q N }, where A n The corresponding take-off and landing point weight Q n Meet the following conditions: Q n =α1×H n +α2×C n +α3×e^(-G n ), wherein α1 refers to a preset first position weight, α2 refers to a preset first flow weight, and α3 refers to a preset building interference weight; S1300, based on C, obtain the second traffic flow set F corresponding to B = {F1, F2, ..., F n , ..., F N }, where B n The corresponding second traffic flow list F n ={F n1 , F n2 , ..., F ni , ..., F nI(n) }, F ni Equal to B ni The average value of the first traffic flow of the two corresponding initial drone take-off and landing points; S1400, based on B and H, obtain the route location importance set K corresponding to B = {K1, K2, ..., K n , ..., K N }, where B n The corresponding route location importance list K n ={K n1 , K n2 , ..., K ni , ..., K nI(n) }, K ni Equal to B ni The average of the importance of the take-off and landing points of the two corresponding initial UAV take-off and landing points; S1500, based on B, D, E, F and K, obtain the route weight set P corresponding to B = {P1, P2, ..., P n , ..., P N }, where P n = {P n1 , P n2 , ..., P ni , ..., P nI(n) }, B ni The corresponding route weight P ni Meet the following conditions: P ni =β1×e^(-D ni )+β2×E ni / ((∑ n＝1 N (S i＝1 I(n) E ni )) / (S n＝1 N I(n)))+β3×F ni +β4×K ni , among them, β1 refers to the preset abnormal weight, β2 refers to the preset route length weight, β3 refers to the preset second flow weight, and β4 refers to the preset second position weight; S1600, according to A, B, Q and P, the initial UAV take-off and landing points are used as nodes, the predicted flight routes are used as edges, the take-off and landing point weights corresponding to the initial UAV take-off and landing points are used as the node values ​​of the corresponding nodes, and the route weights corresponding to the predicted flight routes are used as the edge values ​​of the corresponding edges, and a complex network model T is obtained, wherein the complex network model T is used to screen the target UAV take-off and landing points from A; S1700, based on T and E, obtain the shortest path between any two nodes in T; S1800, determining the number of shortest paths passing through the nodes corresponding to the nth initial drone take-off and landing point as the node betweenness JS corresponding to the nth initial drone take-off and landing point n ; S1900, traverse n = 1, 2, ..., N, obtain the node betweenness set JS = {JS1, JS2, ..., JS n , ..., JS N }; S1910, based on A and B, obtain the clustering coefficient set JL = {JL1, JL2, ..., JL n , ..., JL N }, where JL n Meet the following conditions: JL n =ZS n / (I(n)×(I(n)-1)), where, ZS n It refers to the total number of routes between the I(n) initial UAV take-off and landing points connected to the nth initial UAV take-off and landing point.

3. The method for determining the priority of candidate take-off and landing points of unmanned aerial vehicles based on a complex network model according to claim 2 is characterized in that: S1100 also includes the following steps: S1110, obtain the position coordinate set Z={Z1, Z2, . . . , Z n , ..., Z N }, where Z n A n The corresponding position coordinates; S1120, based on B and Z, obtain the position importance set H corresponding to A = {H1, H2, ..., H n , ..., H N }, where A n The corresponding position importance H n Meet the following conditions: H n =(2 / ((1+e^(-I(n)))-1))×(e^(-∑ j＝1 N d(Z n , Z j )) / N), where e is the natural index, d(Z n , Z j ) refers to Z n and Z j The distance between.

4. The method for determining the priority of candidate take-off and landing points of unmanned aerial vehicles based on a complex network model according to claim 2 is characterized in that: S1100 also includes the following steps: S1130, obtain the first building density set J 1 ={J 1 1. J 1 2, ..., J 1 n , ..., J 1 N } and the first building height set J 2 ={J 2 1. J 2 2, ..., J 2 n , ..., J 2 N }, where J 1 n A n The first building density within the corresponding first preset range, J 2 n ={J 2 n1 , J 2 n2 , ..., J 2 nu , ..., J 2 nU(n) }, J 2 nu A n The height of the u-th first building within the first preset range, J 2 nu >J 0 , J 0 It refers to the preset building height threshold, u=1, 2, ..., U(n), U(n) refers to A n The total number of first buildings within the corresponding first preset range; S1140, according to J 1 and J 2 , obtain the first building interference level set G = {G1, G2, ..., G n , ..., G N }, where A n The corresponding first building interference level G n Meet the following conditions: G n =(2 / ((1+e^(-J 1 n ))-1))×(γ1×(Σ u＝1 U(n) J 2 nu ) / U(n)+γ2×max(J 2 n )), where max() means Take the maximum value function, γ1 refers to the preset first height weight, and γ2 refers to the preset second height weight.

5. The method for determining the priority of candidate take-off and landing points of unmanned aerial vehicles based on a complex network model according to claim 4 is characterized in that: S1000 also includes the following steps: S1920, obtain the second building density set R 1 = {R 1 1. R 1 2, ..., R 1 n , ..., R 1 N }, the second building height set R 2 = {R 2 1. R 2 2, ..., R 2 n , ..., R 2 N }、Population density set Y 1 = {Y 1 1. Y 1 2, ..., Y 1 n , ..., Y 1 N } and the population category importance set Y 2 = {Y 2 1. Y 2 2, ..., Y 2 n , ..., Y 2 N }, where R 1 n = {R 1 n1 , R 1 n2 , ..., R 1 ni , ..., R 1 nI(n) }, R 1 ni Refers to B ni The second building density within the corresponding second preset range, R 2 n = {R 2 n1 , R 2 n2 , ..., R 2 ni , ..., R 2 nI(n) }, R 2 ni = {R 2 ni1 , R 2 ni2 , ..., R 2 niv , ..., R 2 niV(ni) }, R 2 niv Refers to B ni The height of the vth second building within the corresponding second preset range, Y 1 n = {Y 1 n1 , Y 1 n2 , ..., Y 1 ni , ..., Y 1 nI(n) },Y 1 ni Refers to B ni The corresponding population density within the second preset range, Y 2 n = {Y 2 n1 , Y 2 n2 , ..., Y 2 ni , ..., Y 2 nI(n) },Y 2 ni = {Y 2 ni1 , Y 2 ni2 , ..., Y 2 niw , ..., Y 2 niW(ni) },Y 2 niw Refers to B ni The category importance corresponding to the wth population in the corresponding second preset range, v=1, 2, ..., V(n), V(n) refers to A n The total number of second buildings within the corresponding second preset range, w=1, 2, ..., W(ni), W(ni) refers to B ni The corresponding total population within the second preset range; S1930, according to R 1 , R 2 , Y 1 and Y 2 , we get the route anomaly degree set D = {D1, D2, ..., D n , ..., D N }, where D n ={D n1 , D n2 , ..., D ni , ..., D nI(n) }, B ni The corresponding route abnormality degree D ni Meet the following conditions: D ni =δ1×D 1 ni +δ2×D 2 ni , where δ1 refers to the preset building weight, δ2 refers to the preset first population weight, D 1 ni J(2 / ((1+e^(-R 1 ni ))-1))×(γ1×(Σ v＝1 V(n) R 2 niv ) / V(n)+γ2×max(R 2 ni )), D 2 ni =(2 / ((1+e^(-Y 1 ni ))-1))×(ε1×(Σ w＝1 W(n) Y 2 niv ) / W(n)+ε2×max(Y 2 ni )), ε1 refers to the preset second population weight, and ε2 refers to the preset third population weight.

6. The method for determining the priority of candidate take-off and landing points of unmanned aerial vehicles based on a complex network model according to claim 1 is characterized in that: S3000 also includes the following steps: S3100, determine the initial UAV take-off and landing points corresponding to the largest M candidate priorities in HX as candidate UAV take-off and landing points.

7. The method for determining the priority of candidate take-off and landing points of unmanned aerial vehicles based on a complex network model according to claim 1, characterized in that: S6000 also includes the following steps: S6100, set FS≤FS 0 The M candidate UAV take-off and landing points at the time are determined as M key take-off and landing points; S6200, obtain the preset prediction model and the initial UAV take-off and landing point code set A 1 ={A 1 1. A 1 2, ..., A 1 n , ..., A 1 N }、Preset key take-off and landing point code set GJ={GJ1,GJ2,……,GJ μ , ..., GJ M } and preset reference vector YS=(1, 1, ..., 1, 0, 0, ..., 0), where A 1 n A n The corresponding initial take-off and landing point code, GJ μ It refers to the key take-off and landing point code corresponding to the μth key take-off and landing point, μ = 1, 2, ..., M, M refers to the total number of key take-off and landing points, the first Щ elements in YS are value 1, and the last N-Щ elements are value 0, Щ refers to the preset number of take-off and landing points of the target UAV; S6300: Input A, B, Q, P and JS into the preset prediction model to obtain a prediction priority ranking vector PX=(PX1, PX2, ..., PX n , ..., PX N ), where PX n It refers to the initial UAV take-off and landing point with the predicted priority of nth; S6400, perform vector multiplication on PX and YS to obtain the take-off and landing point reference vector CK 1 =(PX1, PX2, ..., PX σ , ..., PX Щ , 0, 0, ..., 0), where PX σ It refers to the initial UAV take-off and landing point with the prediction priority of σth, PX Щ It refers to the initial UAV take-off and landing point with the predicted priority of the Щth position, σ=1, 2, …, Щ; S6500, according to CK 1 , A 1 and GJ, get CK 1 The corresponding first take-off and landing point coding reference vector BM 1 =(BM1, BM2, ..., BM σ , ..., BM Щ , 0, 0, ..., 0), where BM σ PX σ The corresponding first take-off and landing point code, BM Щ PX Щ The corresponding first take-off and landing point code; S6600, transposed vector of GJ and BM 1 Multiply them to obtain the second take-off and landing point coding reference vector BM 2 ; S6700, for BM 2 Perform binarization processing to obtain the third take-off and landing point coding reference vector BM 3 ; S6800, according to BM 3 And the preset encoding threshold YZ, the prediction loss Loss = ρ × (YZ-SUM (BM 3 )), where ρ is the preset loss coefficient, SUM(BM 3 ) means BM 3 The sum of all elements in ; S6900, updating the parameters of the preset prediction model according to Loss until Loss=0, and obtaining the target prediction model; S61000, input A, B, Q, P and JS into the target prediction model to obtain a target priority ranking vector; S61100, determining the first Щ initial UAV take-off and landing points in the target priority sorting vector as the target UAV take-off and landing points.

8. A device for determining the priority of candidate take-off and landing points of unmanned aerial vehicles based on a complex network model, characterized in that: The device for determining the priority of candidate take-off and landing points of unmanned aerial vehicles comprises: The data acquisition module is used to obtain the initial drone take-off and landing point set A = {A1, A2, ..., A n , ..., A N }、Predicted flight route set B={B1,B2,……,B n , ..., B N }、The take-off and landing point weight set Q={Q1,Q2,……,Q n , ..., Q N }、Route weight set P={P1,P2,……,P n , ..., P N }、Node betweenness set JS={JS1,JS2,……,JS n , ..., JS N }、Clustering coefficient set JL={JL1,JL2,……,JL n , ..., JL N }、Route abnormality degree set D={D1,D2,……,D n , ..., D N }、Route length set E={E1,E2,……,E n ,……,E N } and the second traffic flow set F = {F1, F2, ..., F n , ..., F N }, where A n refers to the nth initial drone take-off and landing point, B n = {B n1 , B n2 , ..., B ni , ..., B nI(n) }, B ni A n The corresponding i-th predicted flight route, Q n A n The corresponding take-off and landing point weight, P n = {P n1 , P n2 , ..., P ni , ..., P nI(n) }, P ni Refers to B ni The corresponding route weight, JS n A n The corresponding node betweenness, JL n A n The corresponding clustering coefficient, D n ={D n1 , D n2 , ..., D ni , ..., D nI(n) }, D ni Refers to B ni The corresponding route abnormality degree, E n ={E n1 , E n2 ,……,E ni ,……,E nI(n) }, E ni Refers to B ni The corresponding route length, F n ={F n1 , F n2 , ..., F ni , ..., F nI(n) }, F ni Refers to B ni The corresponding second traffic flow, n = 1, 2, ..., N, N refers to the total number of initial drone take-off and landing points, i = 1, 2, ..., I (n), I (n) refers to the total number of predicted flight routes corresponding to the nth initial drone take-off and landing point; The priority acquisition module is used to obtain the candidate priority set HX = {HX1, HX2, ..., HX5} according to A, B, Q, P, JS, JL and the preset initial weight set QZ = {QZ1, QZ2, QZ3, QZ4, QZ5} n , ..., HX N }, where QZ1 refers to the preset weight of the take-off and landing point weight, QZ2 refers to the preset weight of the route weight, QZ3 refers to the preset weight of the number of routes, QZ4 refers to the preset weight of the node betweenness, QZ5 refers to the preset weight of the clustering coefficient, and A n Corresponding candidate priority HX n Meet the following conditions: HX n =QZ1×Q n +QZ2×∑ i＝1 I(n) (P ni / I(n))+QZ3×I(n)+QZ4×JS n +QZ5×JL n ; The take-off and landing point screening module is used to screen M candidate UAV take-off and landing points from N initial UAV take-off and landing points according to HX; The first landing and take-off point scoring module is used to obtain the candidate landing and take-off point prediction score set FZ = {FZ1, FZ2, ..., FZ n , ..., FZ N }, where A n Corresponding candidate take-off and landing point prediction score FZ n Meet the following conditions: FZ n =β1×(Σ i＝1 I(n) e^(-D ni ))+β2×(∑ n＝1 N (S i＝1 I(n) E ni ))+β3×(∑ n＝1 N (S i＝1 I(n) F ni )), among them,b 1 refers to the preset abnormal weight, β2 refers to the preset route length weight, and β3 refers to the preset second flow weight; The second take-off and landing point scoring module is used to obtain the screening scores FS of M candidate drone take-off and landing points according to FZ, where FS meets the following conditions: FS=(Σ n＝1 N (FZ n -FZ0) 2 / N, where the candidate take-off and landing point prediction mean FZ0 ​​= Σ n＝1 N (FZ n ) / N; Candidate priority determination module, used if FS>FS 0 , then update QZ, and use the updated QZ to replace the preset initial weight set, and repeat the priority acquisition module until FS ≤ FS 0 , get FS≤FS 0 The candidate priority set HX when FS≤FS 0 The candidate priority set HX at this time is used to screen the target UAV take-off and landing point from the N initial UAV take-off and landing points, FS 0 Refers to the preset screening score threshold.

9. A non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the non-transitory computer-readable storage medium, characterized in that: The at least one instruction or the at least one program is loaded and executed by the processor to implement the method for determining the candidate priority of the UAV take-off and landing points based on the complex network model as described in any one of claims 1-7.

10. An electronic device, characterized in that: The invention comprises a processor and the non-transitory computer-readable storage medium as claimed in claim 9.

Images