A method and system for optimizing the layout of unmanned aerial vehicle (UAV) take-off and landing facilities for resilient efficiency.

Abstract

This invention discloses a method and system for optimizing the layout of traffic drone take-off and landing facilities based on resilience and efficiency. Belonging to the field of drone take-off and landing facility layout technology, it addresses the technical problem that existing methods for fixed take-off and landing facility layouts cannot meet the needs of resilient urban transportation construction and cannot provide timely and effective responses to emergencies. The method includes: determining demand point data for drone take-off and landing facilities based on urban road network structure data; determining alternative layout locations for fixed take-off and landing facilities based on the demand point data; constructing a spatial resilience index system for the layout of fixed take-off and landing facilities; constructing a multi-objective optimization model and its constraints for the layout of fixed take-off and landing facilities based on the alternative layout locations and the spatial resilience index system; and solving the multi-objective optimization model using a non-dominated sorting genetic algorithm to obtain an optimized layout scheme for fixed take-off and landing facilities.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) take-off and landing facility layout technology, and in particular to a method and system for optimizing the layout of traffic UAV take-off and landing facilities with optimized resilience and efficiency. Background Technology

[0002] Urban resilient transportation infrastructure is a crucial component of smart city development. It refers to a transportation construction approach that addresses the risks and disruptions to urban road networks caused by natural disasters, social events, and traffic incidents. This involves deploying effective measures throughout the pre-disaster, during-disaster, and post-disaster phases to monitor and prevent transportation risks, manage disaster resilience, and facilitate rapid recovery. Strengthening urban resilient transportation infrastructure can enhance a city's comprehensive disaster prevention, mitigation, and emergency management capabilities.

[0003] However, current site selection processes for fixed take-off and landing facilities for transportation drones typically only consider factors such as the drone's range and the cost of deploying the facilities. They fail to consider how to ensure that transportation drones can handle both routine inspections and emergency support in the event of unforeseen circumstances, and how to respond quickly to emergencies. In other words, they do not consider indicators related to resilient transportation services. Therefore, the current methods for deploying fixed take-off and landing facilities for transportation drones cannot meet the needs of building resilient urban transportation systems and cannot provide timely and effective responses and handling in the event of emergencies. Summary of the Invention

[0004] This invention provides a method and system for optimizing the layout of traffic drone take-off and landing facilities based on resilience and efficiency, to solve the following technical problem: the layout scheme of fixed take-off and landing facilities for traffic drones obtained by existing methods cannot meet the needs of urban resilient transportation construction, and cannot provide timely and effective response and handling in the event of emergencies.

[0005] The embodiments of the present invention adopt the following technical solutions:

[0006] On the one hand, embodiments of the present invention provide a method for optimizing the layout of traffic drone take-off and landing facilities based on resilience and efficiency. The method includes: determining the demand point data of drone take-off and landing facilities based on urban road network structure data.

[0007] Based on the aforementioned demand data, alternative site selection schemes for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs) are determined.

[0008] Construct a spatial resilience index system for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs); wherein the spatial resilience index system includes at least the following spatial resilience indicators: pre-event robustness, pre-event redundancy, and post-event rapidity;

[0009] Based on the aforementioned alternative layout schemes and the aforementioned spatial resilience index system, a multi-objective optimization model for the layout of UAV fixed take-off and landing facilities and its constraints are constructed.

[0010] The multi-objective optimization model is solved using a non-dominated sorting genetic algorithm to obtain an optimized layout scheme for UAV fixed take-off and landing facilities.

[0011] In one feasible implementation, the demand point data for drone take-off and landing facilities is determined based on urban road network structure data, specifically including:

[0012] Acquire urban road network structure data and urban traffic data; wherein, the urban road network structure data includes urban expressway data, arterial road data, and secondary arterial road data; the urban traffic data includes urban POI data, traffic operation status data, and traffic accident data;

[0013] Based on the urban traffic data, the demand point categories for drone take-off and landing facility services are determined, and the latitude and longitude information of all demand points is obtained to obtain demand point data; wherein, the demand point categories include at least: special structure points, traffic facility service points, frequently congested points, and accident-prone points;

[0014] The demand point data and the urban road network structure data are normalized, and the demand point layer and the road network structure layer are connected by the spatial connection method to obtain the spatial distribution layer of the demand points.

[0015] In one feasible implementation, based on the demand point data, alternative site selection schemes for the layout of UAV fixed take-off and landing facilities are determined, specifically including:

[0016] Based on the aforementioned demand point data, establish a demand point set U = {u1, u2, u3, ..., u...} Q};

[0017] Initialize the initial layout point set Unvisited demand point set Φ = U, noise point set Cluster layout candidate point set Cluster partitioning set

[0018] Cluster set And initialize the cluster set with cluster index k = 0;

[0019] Iterate through the set of demand points U, and for any demand point u in it... a Perform neighborhood analysis, and update the noise point set, the cluster layout candidate point set, and the cluster partition set based on the analysis results; after traversal, output the cluster layout candidate point set Θ. ω ={Θω1 ,Θ ω2 ,…,Θ ωk}, the cluster partitioning set Ω={Ω1,Ω2,…,Ω k} and the noise point set Τ={τ1,τ2,…,τ α};

[0020] Select Θ ω The demand points in T are used as candidate points for the layout of fixed take-off and landing facilities for UAVs, resulting in the candidate point set Z = Θ. ω ∪Τ={z1,z2,…,z k+α This serves as an alternative layout option for the fixed take-off and landing facility for the aforementioned UAV.

[0021] In one feasible implementation, the set of demand points U is traversed, and for any demand point u within it... a Perform neighborhood analysis, and update the noise point set, the cluster layout candidate point set, and the cluster partition set based on the analysis results, specifically including:

[0022] Searching for any demand point u in the set of demand points U a The set of sub-demand points covered in the neighborhood radius ε like Then demand point u a Let θ be the initial layout point a And add it to the initial layout point set Θ; if Furthermore, within the ε-neighborhood of any other initial layout point, the demand point u... a If it is a boundary point, then the required point is u. a For noise point τ a And add it to the noise point set T; where,

[0023] The neighborhood radius ε is half of the maximum range of the UAV, and minPts is the minimum number of fixed take-off and landing facilities required for the UAV.

[0024] After traversing the demand point set U, select any initial layout point θ from the initial layout point set Θ. a Add to the cluster layout candidate point set Θ ω And update the cluster index k = k + 1;

[0025] Traverse the set of candidate cluster layout points Θ ω Find the candidate points θ for any cluster layout. ωa The set of all sub-demand points within the ε-neighborhood And add to cluster set Ω k Meanwhile, the initial layout point set Θ = Θ - (Ω) is updated sequentially. k ∩Θ), Unvisited demand point set Φ=Φ-Ω kAnd the cluster partition set Ω;

[0026] When the initial layout point set And the set of demand points was not visited. At this point, the final cluster partition set Ω = {Ω1, Ω2, ..., Ω} is obtained. k};

[0027] When the initial layout point set And the set of demand points was not visited. At that time, the ε-neighborhood of each unvisited demand point is searched sequentially in the set of unvisited demand points Φ to obtain the set of unvisited sub-demand points.

[0028] Get The demand point u with the most unvisited demand points in the middle neighborhood coverage b ∈U, will u b Add to cluster layout candidate point set Θ ω ,u b The set of unvisited sub-demand points covered by the neighborhood of a point Then add it to the cluster set Ω k ;

[0029] Update the set of unvisited demand points Φ = Φ - Ω k k = k + 1; and so on, until... Update the cluster partition set Ω = {Ω1, Ω2, ..., Ω} k}

[0030] In one feasible implementation, a spatial resilience index system for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs) is constructed, specifically including:

[0031] according to Construct a pre-launch robustness calculation formula; where S i Indicates the number of required points covered by fixed take-off and landing facility point i; n represents the total number of fixed take-off and landing facility points.

[0032] according to Construct a pre-emptive redundancy calculation formula; where Q represents the sum of the number of demand points in the four categories;

[0033] according to Construct a formula for calculating post-event speed; where R i The supply and demand ratio for a fixed takeoff and landing facility point i; N i The number of drones deployed at fixed take-off and landing facility point i; d ij d0 is the distance from fixed takeoff and landing facility point i to demand point j; Ω is the service radius of the fixed takeoff and landing facility; i The set of demand points serving fixed take-off and landing facility point i; is the Gaussian distance decay function; H is the spatial accessibility of the UAV fixed take-off and landing facility system, representing the post-event speed.

[0034] In one feasible implementation, based on the proposed layout alternatives and the spatial resilience index system, a multi-objective optimization model for the layout of UAV fixed take-off and landing facilities and its constraints are constructed, specifically including:

[0035] Based on the proposed alternative layout schemes, the constraints of the multi-objective optimization model are constructed. These constraints include at least: integer constraints, constraints on the number of UAVs that a fixed take-off and landing facility can accommodate, upper limits on the demand points for UAV services, constraints on the number of fixed take-off and landing facilities that can be deployed, constraints that a UAV can only be deployed at one fixed take-off and landing facility, constraints on the coverage of demand points by UAVs and fixed take-off and landing facilities, constraints on the relationship between UAVs and fixed take-off and landing facilities, and constraints on the coverage relationship between fixed take-off and landing facilities and demand points.

[0036] Based on the aforementioned constraints, a multi-objective optimization model for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs) is constructed with the optimization objectives of maximizing spatial resilience index and minimizing comprehensive operating cost.

[0037] In one feasible implementation, the integer constraint is:

[0038]

[0039] The number of drones that a fixed take-off and landing facility can accommodate is constrained as follows: in, The number of drones deployed within fixed take-off and landing facility point i; N imax Z represents the maximum number of UAVs that can be accommodated within a fixed take-off and landing facility i; Z represents the set of alternative locations; and G represents the set of UAVs.

[0040] The upper limit constraint on the demand points for drone services is: in, The number of demand points for services provided by drones;

[0041] The number of fixed take-off and landing facilities is constrained as follows: Where δ is the lower limit of the number of fixed take-off and landing facilities; δ+α represents the upper limit of the number of fixed take-off and landing facilities.

[0042] A single drone is deployed at only one fixed take-off and landing facility:

[0043] The constraints on the coverage requirements of drones and fixed take-off and landing facilities are as follows:

[0044]

[0045] The relationship between the drone and the fixed take-off and landing facility is constrained as follows: This indicates that if fixed take-off and landing facility point i is not selected, drones will not be deployed at that point;

[0046] The constraint on the coverage relationship between fixed take-off and landing facilities and demand points is: This means that if fixed take-off and landing facility point i is not selected or demand point j is not within the service radius of fixed take-off and landing facility point i, then i cannot provide service to j.

[0047] In one feasible implementation, based on the aforementioned constraints, a multi-objective optimization model for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs) is constructed with the optimization objectives of maximizing spatial resilience and minimizing overall operating costs. This model specifically includes:

[0048] Based on the aforementioned constraints, the pre-hoc robustness index value C is calculated respectively. s The pre-event redundancy index value E and the post-event rapidity index value H;

[0049] Let y = C s Substituting y = E and y = H into the normalization formula respectively In this process, the normalized ex-ante robustness index value C was obtained. s ', Pre-event redundancy index value E' and Post-event rapidity index value H';

[0050] According to f1 = C s '+E'+H' yields the resilience index function f1;

[0051] according to The comprehensive cost function f2 is obtained; where C g G represents the purchase cost of the drones; cm g Maintenance and management costs for drones; g The service life of the drone (g); C i To determine the construction cost of fixed take-off and landing facility i; cm i For the maintenance and management costs of fixed take-off and landing facilities i; i The service life of fixed take-off and landing facility i;

[0052] Based on maxF=V(f1,-f2), the overall optimization objective function of the multi-objective optimization model is obtained; where V is the effect function.

[0053] In one feasible implementation, the multi-objective optimization model is solved using a non-dominated sorting genetic algorithm to obtain an optimized layout scheme for the UAV fixed take-off and landing facility, specifically including:

[0054] Initialize a set of feasible solutions for the deployment of UAV take-off and landing facilities, as the initial parent population P. t ; and initialize the set of individuals dominated by individual a. The number of solution individuals n that dominate individual a a =0 and non-dominant hierarchical sorting set Where γ = 1;

[0055] Select the initial parent population P t The deployment scheme of UAV take-off and landing facilities corresponding to individual a in the middle is p. ta Calculate p ta The objective function value;

[0056] If p ta At least one of the objective function values ​​is greater than p. tq The corresponding objective function value, and all other values ​​are greater than or equal to p. tq Other objective function values ​​determine p. ta Dominate p tq S a =S a ∪{p tq}; if p tq Dominate p ta Let n a =n a +1;

[0057] If n a =0, then let the non-dominant rank of individual a be 0. p ta Add to the non-dominant hierarchical sorted set P′ γ In, and P′ tγ The n corresponding to individual q dominated by individual a. q Decrease by 1; if n q =0, then let the non-dominant rank of individual q be 0. γ = γ + 1; this process is iterated to obtain the non-dominant level W for each individual. t rank ;

[0058] according to Calculate the crowding distance for each individual. a ; where n f Let b represent the number of objective functions, where b ∈ n. f ;f b (x a+1 ), f b (x a-1 ) are the b-th objective function values ​​for individuals a+1 and a-1, respectively, f b max f b min These are the maximum and minimum values ​​of the b-th objective function, respectively.

[0059] For the initial parent population P tWhen individuals have the same non-dominance level, individuals with a large crowding distance are preferentially selected for crossover and mutation to generate the offspring population Y. t When individuals have unequal non-dominance levels, individuals with lower non-dominance levels are preferentially selected for crossover and mutation to generate offspring population Y. t ;

[0060] By merging the parent and offspring populations, a new generation population R is obtained. t For the new generation population R t Perform non-dominated ranking and crowding distance calculation, and select the top M individuals with the lowest non-dominated rank and the largest crowding distance as the parent population P of the new generation. t+1 Repeat this iterative process until the maximum number of iterations Tmax is reached, then output P. tmax An optimized layout scheme for aggregating fixed take-off and landing facilities for drones.

[0061] On the other hand, embodiments of the present invention also provide a resilience-efficiency optimized layout system for traffic drone take-off and landing facilities, the system comprising:

[0062] The alternative site generation module is used to determine the demand point data and spatial distribution layer of UAV take-off and landing facilities based on urban road network structure data; and to determine the layout alternative site schemes for UAV fixed take-off and landing facilities based on the demand point data.

[0063] A multi-objective optimization model construction module is used to construct a spatial resilience index system for the layout of UAV fixed take-off and landing facilities; wherein, the spatial resilience index system includes at least the following spatial resilience indices: pre-event robustness, pre-event redundancy, and post-event rapidity; based on the layout alternative point schemes and the spatial resilience index system, a multi-objective optimization model for the layout of UAV fixed take-off and landing facilities and its constraints are constructed.

[0064] The layout optimization module is used to solve the multi-objective optimization model using a non-dominated sorting genetic algorithm to obtain the layout optimization scheme for the UAV fixed take-off and landing facility.

[0065] Compared with the prior art, the method and system for optimizing the layout of traffic drone take-off and landing facilities according to the embodiments of the present invention have the following beneficial effects:

[0066] (1) This invention starts from different dimensions such as road attributes, traffic efficiency, and traffic safety. By integrating and processing multi-source heterogeneous data such as urban road network structure, POI, traffic operation status and traffic accidents, it deeply mines the demand point categories served by UAV take-off and landing facilities, establishes a demand system structure based on "special structure points - traffic facility service points - frequently congested points - accident-prone points", obtains the spatial distribution pattern of UAV take-off and landing facility service demand points, and provides an effective reference for the site selection of UAV take-off and landing facilities.

[0067] (2) Starting from the entire disturbance cycle from "before" to "after", this invention constructs a spatial resilience index system for optimizing the layout of UAV fixed take-off and landing facilities, including pre-event robustness, pre-event redundancy and post-event rapidity, and uses the supply and demand matching relationship to realize the quantitative analysis of the above three indicators, so as to determine the resilience service optimization target for the site selection and capacity planning of UAV take-off and landing facilities.

[0068] (3) This invention comprehensively considers the collaborative site selection problem of UAVs and fixed take-off and landing facilities. With the goal of maximizing resilience index and minimizing overall cost, and with constraints such as the capacity of UAV fixed take-off and landing facilities, UAV range, and the number of demand points that UAVs can cover, a multi-objective layout optimization model for UAV fixed take-off and landing facilities is constructed and solved by a non-dominated sorting genetic algorithm, thereby generating a UAV fixed take-off and landing facility deployment scheme. The above model algorithm can provide a reliable new infrastructure construction scheme for improving urban traffic resilience and efficiency. Attached Figure Description

[0069] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0070] Figure 1 A flowchart of a resilience-efficiency optimized layout method for traffic drone take-off and landing facilities is provided in an embodiment of the present invention;

[0071] Figure 2 This is a structural schematic diagram of a traffic drone take-off and landing facility layout system with optimized resilience and efficiency, provided as an embodiment of the present invention. Detailed Implementation

[0072] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0073] This invention provides a method for optimizing the layout of transportation drone take-off and landing facilities based on resilience and efficiency, such as... Figure 1 As shown, the method for optimizing the layout of transportation drone take-off and landing facilities with optimized resilience efficiency specifically includes steps S101-S105:

[0074] S101. Based on urban road network structure data, determine the demand point data for drone take-off and landing facilities.

[0075] Specifically, the data includes urban road network structure data and urban traffic data. The urban road network structure data includes urban expressway data, arterial road data, and secondary arterial road data. The urban traffic data includes urban POI data, traffic operation status data, and traffic accident data.

[0076] Furthermore, based on urban traffic data, the demand points for drone take-off and landing facilities services are categorized, and the latitude and longitude information of all demand points is obtained to obtain demand point data; among them, the demand point categories include at least: special structure points, traffic facility service points, frequently congested points, and accident-prone points.

[0077] As a feasible implementation method, urban road network structure data is acquired, and urban expressways, arterial roads, and secondary arterial roads are filtered out. The data is then imported into ArcGIS software for topology correction. Next, urban POI data, traffic operation status data, and traffic accident data are acquired. The acquired data is cleaned and processed to identify the demand points served by the UAV take-off and landing facilities, including special structure points, traffic facility service points, frequently congested points, and accident-prone points. The latitude and longitude information of all demand points is then obtained using ArcGIS software.

[0078] The POI data cleaning process is as follows: First, duplicate data is deleted. Then, the latitude and longitude information of the POI data is matched with the electronic map. If the deviation between the two is small, the latitude and longitude information is corrected. If the spatial latitude and longitude information of the POI data does not exist in the actual geographical location, the data item is deleted.

[0079] The categories of requirements and the methods for determining them are as follows:

[0080] (1) Special structural points: including tunnels and bridges; tunnel and bridge related data are obtained from urban POI data, and the midpoint of the tunnel or bridge is extracted as special structural points.

[0081] (2) Transportation facility service points: including five categories: railway stations, high-speed rail stations, airports, parking lots, and highway toll stations; relevant data of the above five categories are obtained by filtering from urban POI data, and any one of the entrance points of transportation facilities is extracted as a transportation facility service point.

[0082] (3) Frequently Congested Points: Traffic operation status data of road segments during peak hours over the past three months were obtained through the electronic map API interface, with a statistical time interval of 15 minutes. The road segment traffic status values ​​included: 0 for unknown; 1 for smooth traffic; 2 for slow traffic; 3 for congestion; and 4 for severe congestion. The average traffic status of each road segment during all peak hours over the past three months was calculated. Road segments with an average traffic status greater than or equal to 3 were selected as frequently congested road segments, and the midpoint of the frequently congested road segments was extracted as the frequently congested point.

[0083] (4) Accident-prone areas: Based on urban traffic accident data, the latitude and longitude information of accident locations is extracted, the number of traffic accidents on each road segment is counted, and the accident rate per million kilometers of the road segment is calculated. The calculation formula is as follows: In the formula, T i A represents the accident rate per million kilometers for road segment i; i L represents the number of traffic accidents occurring on road segment i; i This represents the length of road segment i.

[0084] The road sections with the highest accident rates per million kilometers were selected as accident-prone sections, and the midpoint of these sections was designated as accident-prone points.

[0085] Furthermore, ArcGIS software was used to normalize the demand point data and the urban road network structure data, unifying them to WGS-84 coordinates. Then, the demand point layer and the road network structure layer were connected using the spatial connection method to obtain the spatial distribution layer of the demand points.

[0086] S102. Based on demand point data, determine alternative layout options for fixed take-off and landing facilities for unmanned aerial vehicles (UAVs).

[0087] Specifically, firstly, based on the demand point data, a demand point set U = {u1, u2, u3, ..., u...} is established. Q Then initialize the initial layout point set. Unvisited demand point set Φ = U, noise point set Cluster layout candidate point set Cluster partitioning set Cluster set And initialize the cluster set with cluster index k=0.

[0088] Furthermore, iterate through the set of demand points U, and for any demand point u in it... a Perform neighborhood analysis, and based on the analysis results, update the noise point set, the cluster layout candidate point set, and the cluster partition set. The specific process is as follows:

[0089] (1) Search for any demand point u in the demand point set U. a The set of sub-demand points covered in the neighborhood radius ε like Then demand point u a Let θ be the initial layout point a And add it to the initial layout point set Θ; if Furthermore, within the ε-neighborhood of any other initial layout point, the demand point u... a If it is a boundary point, then the required point is u. a For noise point τ a And add the noise point set T; where the neighborhood radius ε is half of the maximum range of the UAV, and minPts is the minimum number of points required for the UAV's fixed take-off and landing facilities.

[0090] (2) After the demand point set U has been traversed, select any initial layout point θ from the initial layout point set Θ. a Add to the cluster layout candidate point set Θ ω And update the cluster number k = k + 1.

[0091] (3) Traverse the cluster layout candidate point set Θ ω Find the candidate points θ for any cluster layout. ωa The set of all sub-demand points within the ε-neighborhood And add to cluster set Ω k Meanwhile, the initial layout point set Θ = Θ - (Ω) is updated sequentially. k ∩Θ), Unvisited demand point set Φ=Φ-Ω k And the cluster partition set Ω.

[0092] (4) Repeat the above steps, when the initial layout point set And the set of demand points was not visited. At this point, the final cluster partition set Ω = {Ω1, Ω2, ..., Ω} is obtained. k}. When the initial layout point set And the set of demand points was not visited. At that time, the ε-neighborhood of each unvisited demand point is searched sequentially in the set of unvisited demand points Φ to obtain the set of unvisited sub-demand points. Get The demand point u with the most unvisited demand points in the middle neighborhood coverage b ∈U, will u b Add to cluster layout candidate point set Θ ω ,u b The set of unvisited sub-demand points covered by the neighborhood of a point Then add it to the cluster set Ω k Update the set of unvisited demand points Φ = Φ - Ω k k = k + 1; and so on, until... Update the cluster partition set Ω = {Ω1, Ω2, ..., Ω} k}

[0093] (5) After the traversal is complete, output the set of candidate cluster layout points Θ. ω ={Θ ω1 ,Θ ω2 ,…,Θ ωk}, the cluster partitioning set Ω={Ω1,Ω2,…,Ω k} and the noise point set Τ={τ1,τ2,…,τ α}. Select Θ ω The demand points in T are used as candidate points for the layout of fixed take-off and landing facilities for UAVs, resulting in the candidate point set Z = Θ. ω ∪Τ={z1,z2,…,z k+α This serves as an alternative layout option for fixed take-off and landing facilities for unmanned aerial vehicles (UAVs).

[0094] S103. Construct a spatial resilience index system for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs).

[0095] Specifically, pre-emptive robustness is calculated using the degree centrality of fixed take-off and landing (RTD) facilities for UAVs. The more demand points within the service area of ​​a fixed RTD facility, the greater its degree centrality, the greater its service capacity, and the stronger its pre-emptive robustness. According to... Construct a pre-launch robustness calculation formula; where S i This represents the number of required points covered by fixed take-off and landing facility point i; n represents the total number of fixed take-off and landing facility points.

[0096] Furthermore, pre-emptive redundancy is calculated using the average redundancy of nodes in the UAV fixed take-off and landing facility system. That is, the more points requiring repeated services from the UAV fixed take-off and landing facility, the greater the average node redundancy, and the stronger the pre-emptive redundancy. According to... Construct a pre-emptive redundancy calculation formula; where Q represents the sum of the number of demand points in the four categories.

[0097] Furthermore, post-event accessibility represents the degree of rapid response of a fixed UAV take-off and landing facility to different demand points. Here, considering the distance decay effect, a Gaussian two-step move search method is used to calculate the spatial accessibility of the fixed UAV take-off and landing facility to different demand points. The greater the spatial accessibility, the higher the facility's post-event response efficiency and the stronger its post-event accessibility. According to...

[0098] Construct a formula for calculating post-event speed; where R i The supply and demand ratio for a fixed takeoff and landing facility point i; N i The number of drones deployed at fixed take-off and landing facility point i; d ij d0 is the distance from fixed takeoff and landing facility point i to demand point j; Ω is the service radius of the fixed takeoff and landing facility;i The set of demand points serving fixed take-off and landing facility point i; is the Gaussian distance decay function; H is the spatial accessibility of the UAV fixed take-off and landing facility system, representing the post-event speed.

[0099] S104. Based on the alternative layout schemes and spatial resilience index system, construct a multi-objective optimization model for the layout of UAV fixed take-off and landing facilities and its constraints.

[0100] Specifically, based on the alternative site layout schemes, constraints are constructed for a multi-objective optimization model. These constraints include at least: integer constraints, constraints on the number of UAVs that can be accommodated at fixed take-off and landing facilities, upper limit constraints on the demand points for UAV services, constraints on the number of fixed take-off and landing facilities, constraints on the coverage of demand points by UAVs and fixed take-off and landing facilities, constraints on the relationship between UAVs and fixed take-off and landing facilities, and constraints on the coverage relationship between fixed take-off and landing facilities and demand points.

[0101] As a feasible implementation method, the integer constraint is:

[0102] The number of drones that a fixed take-off and landing facility can accommodate is constrained as follows: in, The number of drones deployed within fixed take-off and landing facility point i; N imax Z represents the maximum number of UAVs that can be accommodated within a fixed take-off and landing facility i; Z represents the set of alternative locations; and G represents the set of UAVs.

[0103] The upper limit constraint on the demand points for drone services is: in, The number of demand points for services for drones.

[0104] The number of fixed take-off and landing facilities is constrained as follows: Where δ represents the lower limit of the number of fixed take-off and landing facilities; δ+α represents the upper limit of the number of fixed take-off and landing facilities.

[0105] A single drone is deployed at only one fixed take-off and landing facility:

[0106] The constraints on the coverage requirements of drones and fixed take-off and landing facilities are as follows:

[0107] d ij d0 represents the distance from the fixed take-off and landing facility point i to the demand point j; d0 represents the service radius of the fixed take-off and landing facility for drones.

[0108] The relationship between the drone and the fixed take-off and landing facility is constrained as follows: This indicates that if fixed take-off and landing facility point i is not selected, drones will not be deployed at that point;

[0109] The constraint on the coverage relationship between fixed take-off and landing facilities and demand points is: This means that if fixed take-off and landing facility point i is not selected or demand point j is not within the service radius of fixed take-off and landing facility point i, then i cannot provide service to j.

[0110] Furthermore, based on constraints, and with the optimization objectives of maximizing spatial resilience and minimizing overall operating costs, a multi-objective optimization model for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs) is constructed, specifically including:

[0111] Calculate the pre-hoc robustness index values ​​based on the constraints. Pre-emptive redundancy index value and the rapid response index value after the event in,

[0112]

[0113] Then y = C s Substituting y = E and y = H into the normalization formula respectively In this process, the normalized ex-ante robustness index value C was obtained. s ', Pre-event redundancy index value E' and post-event rapidity index value H'; then according to f1=C s Adding E and H gives us the resilience index function f1.

[0114] Further according to The comprehensive cost function f2 is obtained; where C g G represents the purchase cost of the drones; cm g Maintenance and management costs for drones; g The service life of the drone (g); C i To determine the construction cost of fixed take-off and landing facility i; cm i For the maintenance and management costs of fixed take-off and landing facilities i; i The service life of fixed take-off and landing facility i.

[0115] Finally, based on maxF=V(f1,-f2), the overall objective function of the multi-objective optimization model is obtained, thus completing the construction of the multi-objective optimization model; where V is the effect function.

[0116] S105. Using a non-dominated sorting genetic algorithm, solve the multi-objective optimization model to obtain the layout optimization scheme for UAV fixed take-off and landing facilities.

[0117] Specifically, the population size M, the number of iterations t=1, and the maximum number of iterations T are set. maxInitialize a set of feasible solutions for the deployment of UAV take-off and landing facilities, as the initial parent population P. t And initialize the set of individuals dominated by individual a. The number of solution individuals n that dominate individual a a =0 and non-dominant hierarchical sorting set Where γ = 1.

[0118] Furthermore, the initial parent population P was selected. t The deployment scheme of UAV take-off and landing facilities corresponding to individual a in the middle is p. ta Calculate p ta The objective function values ​​include f1 and f2. If p ta At least one of the objective function values ​​is greater than p. tq The corresponding objective function value, and all other values ​​are greater than or equal to p. tq Other objective function values ​​determine p. ta Dominate p tq S a =S a ∪{p tq}; if p tq Dominate p ta Let n a =n a +1.

[0119] In one embodiment, if p ta The objective function value is f a1 and f a2 , p tq The objective function value is f q1 and f q2 If "f a1 >f q1 f a2 >f q2 ", or "f a1 >f q1 f a2 =f q2 ", or "f a1 =f q1 f a2 >f q2 "It can be determined that p" ta Dominate p tq The same applies to the opposite.

[0120] If n a =0, then let the non-dominant rank of individual a be 0. p ta Add to the non-dominant hierarchical sorted set P′ tγ In, and P′ tγThe n corresponding to individual q dominated by individual a. q Decrease by 1; if n q =0, then let the non-dominant rank of individual q be 0. γ = γ + 1; this process is iterated to obtain the non-dominant level W for each individual. t rank .

[0121] Furthermore, according to Calculate the crowding distance for each individual. a ; where n f Let b represent the number of objective functions, where b ∈ n. f ;f b (x a+1 ), f b (x a-1 ) are the b-th objective function values ​​for individuals a+1 and a-1, respectively, f b max f b min These are the maximum and minimum values ​​of the b-th objective function, respectively.

[0122] In one embodiment, when b = 1, then f b (x a+1 f1(x) is the resilience index function value of individual a+1. a+1 Similarly, when b = 2, then f b (x a+1 f2(x) represents the comprehensive cost function value of individual a+1. a+1 ).

[0123] Furthermore, for the initial parent population P t When individuals have the same non-dominance level, individuals with a large crowding distance are preferentially selected for crossover and mutation to generate the offspring population Y. t When individuals have unequal non-dominance levels, individuals with lower non-dominance levels are preferentially selected for crossover and mutation to generate offspring population Y. t .

[0124] Then, the parent population and the offspring population are merged to obtain the new generation population R. t For the new generation population R t Perform non-dominated ranking and crowding distance calculation, and select the top M individuals with the lowest non-dominated rank and the largest crowding distance as the parent population P of the new generation. t+1 Then repeat the above steps iteratively until the maximum number of iterations T is reached. max Output P tmax This collection serves as an optimized layout scheme for fixed take-off and landing facilities for unmanned aerial vehicles (UAVs).

[0125] In addition, embodiments of the present invention also provide a resilience-efficiency optimized layout system for traffic drone take-off and landing facilities, such as... Figure 2 As shown, the resilience and efficiency optimized traffic drone take-off and landing facility layout system 200 specifically includes:

[0126] The alternative site generation module 210 is used to determine the demand point data and spatial distribution layer of the UAV take-off and landing facilities based on the urban road network structure data; and to determine the layout alternative site schemes of the UAV fixed take-off and landing facilities based on the demand point data.

[0127] The multi-objective optimization model construction module 220 is used to construct a spatial resilience index system for the layout of UAV fixed take-off and landing facilities; wherein, the spatial resilience index system includes at least the following spatial resilience indices: pre-event robustness, pre-event redundancy, and post-event speed; based on the layout alternative point scheme and the spatial resilience index system, a multi-objective optimization model for the layout of UAV fixed take-off and landing facilities and its constraints are constructed.

[0128] The layout optimization module 230 is used to solve the multi-objective optimization model using a non-dominated sorting genetic algorithm to obtain the layout optimization scheme of the UAV fixed take-off and landing facility.

[0129] The various embodiments in this invention are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0130] The foregoing has described specific embodiments of the present invention. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0131] The above description is merely an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations can be made to the embodiments of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the embodiments of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing the layout of unmanned aerial vehicle (UAV) take-off and landing facilities for resilient efficiency, characterized in that, The method includes: Based on urban road network structure data, determine the demand point data and spatial distribution layer of drone take-off and landing facilities. Based on the aforementioned demand data, alternative site selection schemes for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs) are determined. Construct a spatial resilience index system for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs); wherein the spatial resilience index system includes at least the following spatial resilience indicators: pre-event robustness, pre-event redundancy, and post-event rapid response; specifically including: according to Construct a pre-emptive robustness calculation formula; among which, Indicates the number of required points covered by fixed take-off and landing facility point i; n represents the total number of fixed take-off and landing facility points. according to Construct a pre-emptive redundancy calculation formula; where, This represents the sum of the number of demand points in the four categories; according to , Construct a formula for rapid post-event calculation; where, The supply and demand ratio for a fixed take-off and landing facility point i; The number of drones deployed at fixed take-off and landing facility point i; To fix the distance from take-off and landing facility point i to demand point j; The service radius of fixed take-off and landing facilities; The set of demand points serving fixed take-off and landing facility point i; It is a Gaussian distance decay function; Spatial accessibility of a fixed take-off and landing facility system for unmanned aerial vehicles (UAVs) represents the aforementioned post-event rapidity; Based on the aforementioned alternative layout schemes and the aforementioned spatial resilience index system, a multi-objective optimization model for the layout of UAV fixed take-off and landing facilities and its constraints are constructed. The multi-objective optimization model is solved using a non-dominated sorting genetic algorithm to obtain an optimized layout scheme for UAV fixed take-off and landing facilities.

2. The method for optimizing the layout of traffic drone take-off and landing facilities based on resilience efficiency according to claim 1, characterized in that, Based on urban road network structure data, the demand points for drone take-off and landing facilities are determined, specifically including: Acquire urban road network structure data and urban traffic data; wherein, the urban road network structure data includes urban expressway data, arterial road data, and secondary arterial road data; the urban traffic data includes urban POI data, traffic operation status data, and traffic accident data; Based on the urban traffic data, the demand point categories for drone take-off and landing facility services are determined, and the latitude and longitude information of all demand points is obtained to obtain demand point data; wherein, the demand point categories include at least: special structure points, traffic facility service points, frequently congested points, and accident-prone points; The demand point data and the urban road network structure data are normalized, and the demand point layer and the road network structure layer are connected by the spatial connection method to obtain the spatial distribution layer of the demand points.

3. The method for optimizing the layout of traffic drone take-off and landing facilities based on resilience efficiency according to claim 1, characterized in that, Based on the aforementioned demand data, alternative site selection schemes for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs) are determined, specifically including: Based on the aforementioned demand point data, a demand point set is established. ; Initialize the initial layout point set Unvisited demand point set noise point set Cluster layout candidate point set Cluster partitioning set Cluster set and the cluster index for initializing the cluster set. ; Iterate through the set of demand points U, and for any demand point within it... Perform neighborhood analysis, and update the noise point set, the cluster layout candidate point set, and the cluster partition set based on the analysis results; After traversal, output the set of candidate cluster layout points. Cluster partitioning set and noise point set ; Select The demand points in T are used as alternative points for the layout of fixed take-off and landing facilities for drones, resulting in a set of alternative points. This serves as an alternative layout option for the fixed take-off and landing facility for the aforementioned UAV.

4. The method for optimizing the layout of traffic drone take-off and landing facilities according to claim 3, characterized in that, Iterate through the set of demand points U, and for any demand point within it... Perform neighborhood analysis, and update the noise point set, the cluster layout candidate point set, and the cluster partition set based on the analysis results, specifically including: Search for any demand point in the demand point set U neighborhood radius The set of sub-demand points covered in the middle ;like Then the demand point Initial layout point and add to the initial layout point set. ;like And at any other initial layout point Within the neighborhood, the demand point If it is a boundary point, then it is a demand point. Noise point And add the noise point set T; wherein, the neighborhood radius Half the maximum range of a drone. Minimum number of fixed take-off and landing points required for drones; In the set of demand points U After the traversal is complete, in the initial set of layout points Select any initial layout point Add to the cluster layout candidate point set And update the cluster index k=k+1; Traverse the set of candidate cluster layout points Find candidate points for arbitrary cluster layouts. of The set of all sub-demand points within the neighborhood and join the cluster set Meanwhile, the initial layout point set is updated sequentially. Unvisited demand point set and cluster partitioning set ; When the initial layout point set And the set of demand points was not visited. At this point, the final cluster partition set is obtained. ; When the initial layout point set And the set of demand points was not visited. At that time, the set of demand points was never visited. Search each unvisited demand point sequentially. Neighborhood, obtain the set of unvisited sub-demand points. ; Get The demand point with the most unvisited demand points in the middle neighborhood coverage ,Will Add to cluster layout candidate point set , The set of unvisited sub-demand points covered by the neighborhood of a point Then add to the cluster set ; Update the set of unvisited requirements k = k + 1; and so on, until... Update the cluster partition set .

5. The method for optimizing the layout of traffic drone take-off and landing facilities according to claim 1, characterized in that, Based on the aforementioned alternative layout schemes and the aforementioned spatial resilience index system, a multi-objective optimization model for the layout of UAV fixed take-off and landing facilities and its constraints are constructed, specifically including: Based on the proposed alternative layout schemes, the constraints of the multi-objective optimization model are constructed. These constraints include at least: integer constraints, constraints on the number of UAVs that a fixed take-off and landing facility can accommodate, upper limits on the demand points for UAV services, constraints on the number of fixed take-off and landing facilities that can be deployed, constraints that a UAV can only be deployed at one fixed take-off and landing facility, constraints on the coverage of demand points by UAVs and fixed take-off and landing facilities, constraints on the relationship between UAVs and fixed take-off and landing facilities, and constraints on the coverage relationship between fixed take-off and landing facilities and demand points. Based on the aforementioned constraints, a multi-objective optimization model for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs) is constructed with the optimization objectives of maximizing spatial resilience index and minimizing comprehensive operating cost.

6. The method for optimizing the layout of traffic drone take-off and landing facilities according to claim 5, characterized in that, Integer constraints are: ； ； ； ； The number of drones that a fixed take-off and landing facility can accommodate is constrained as follows: ;in, The number of drones deployed within the fixed take-off and landing facility point i; Z represents the maximum number of UAVs that can be accommodated within a fixed take-off and landing facility i; Z represents the set of alternative locations; and G represents the set of UAVs. The upper limit constraint on the demand points for drone services is: ;in, The number of demand points for services provided by drones; The number of fixed take-off and landing facilities is constrained as follows: ;in, This is the minimum number of fixed take-off and landing facilities to be deployed; This indicates the maximum number of fixed take-off and landing facilities that can be deployed. A single drone is deployed at only one fixed take-off and landing facility: ; The constraints on the coverage requirements of drones and fixed take-off and landing facilities are as follows: ; The relationship between the drone and the fixed take-off and landing facility is constrained as follows: This indicates that if the landing facility point is fixed. i If a point is not selected, no drones will be deployed there. The constraint on the coverage relationship between fixed take-off and landing facilities and demand points is: This indicates that if the landing facility point is fixed. i Not selected or required j If it is not within the service radius of the fixed take-off and landing facility point i, then i Cannot be j Provide services.

7. A method for optimizing the layout of traffic drone take-off and landing facilities based on resilience efficiency according to claim 6, characterized in that, Based on the aforementioned constraints, and with the optimization objectives of maximizing spatial resilience and minimizing overall operating costs, a multi-objective optimization model for the layout of fixed take-off and landing facilities for unmanned aerial vehicles (UAVs) is constructed, specifically including: Based on the aforementioned constraints, calculate the pre-hoc robustness index values ​​respectively. Pre-emptive redundancy index value E and the rapid response index value after the event H ; Will y= y=E, y=H Substitute into the normalization formula respectively In this process, the normalized ex-ante robustness index value was obtained. Pre-emptive redundancy index value E’ and the rapid response index value after the event H’ ; according to The resilience index function is obtained. ; according to The comprehensive cost function is obtained. ;in, Let G be the purchase cost of the drone g, where G is the collection of drones; Maintenance and management costs for drones (g); The service life of the drone (g); To determine the construction cost of fixed take-off and landing facility i; The maintenance and management costs of fixed take-off and landing facility i; The service life of fixed take-off and landing facility i; according to Thus, the overall objective function of the multi-objective optimization model is obtained; where V is the effect function.

8. A method for optimizing the layout of traffic drone take-off and landing facilities based on resilience efficiency according to claim 1, characterized in that, The multi-objective optimization model is solved using a non-dominated sorting genetic algorithm to obtain an optimized layout scheme for UAV fixed take-off and landing facilities, specifically including: Initialize a set of feasible solutions for the deployment of UAV take-off and landing facilities as the initial parent population. P t ; and initialize the set of individuals dominated by individual a. The number of solution individuals n that dominate individual a a =0 and the set of non-dominant hierarchical ranking. ;in, =1; Selecting the initial parent population P t Deployment scheme for drone take-off and landing facilities corresponding to individual a p ta ,calculate p ta The objective function value; like p ta At least one of the objective function values ​​is greater than 1. p tq The corresponding objective function value, and all other values ​​are greater than or equal to p tq Other objective function values ​​are then determined. p ta Dominate p tq ,make ;like p tq Dominate p ta Let n a =n a +1; If n a =0, then let the non-dominant rank of individual a be 0. ,Will p ta Add to non-dominant hierarchical sorting set In, and will The n corresponding to individual q dominated by individual a. q Decrease by 1; if n q =0, then let the non-dominant rank of individual q be 0. , This iterative process yields the non-dominant level for each individual. ; according to Calculate the crowding distance for each individual. o a ;in, n f Indicates the number of objective functions. ; , These are the b-th objective function values ​​for individuals a+1 and a-1, respectively. , These are the maximum and minimum values ​​of the b-th objective function, respectively. For the initial parent population P t When individuals have the same non-dominance level, individuals with a large crowding distance are preferentially selected for crossover and mutation to generate offspring populations. Y t When individuals have unequal non-dominant levels, individuals with lower non-dominant levels are preferentially selected for crossover and mutation to generate offspring populations. Y t ; By merging the parent and offspring populations, a new generation of population is obtained. R t For the new generation of population R t Non-dominated ranking and crowding distance calculations were performed, and the top M individuals with the lowest non-dominated ranking and the largest crowding distance were selected as the parent population for the next generation. P t+1 Repeat this process iteratively until the maximum number of iterations is reached. T max Output P tmax An optimized layout scheme for aggregating fixed take-off and landing facilities for drones.

9. A resilience-efficiency optimized layout system for traffic drone take-off and landing facilities, employing the resilience-efficiency optimized layout method for traffic drone take-off and landing facilities as described in any one of claims 1-8, characterized in that, The system includes: The alternative site generation module is used to determine the demand point data and spatial distribution layer of UAV take-off and landing facilities based on urban road network structure data; and to determine the layout alternative site schemes for UAV fixed take-off and landing facilities based on the demand point data. A multi-objective optimization model construction module is used to construct a spatial resilience index system for the layout of UAV fixed take-off and landing facilities; wherein, the spatial resilience index system includes at least the following spatial resilience indices: pre-event robustness, pre-event redundancy, and post-event rapidity; based on the layout alternative point schemes and the spatial resilience index system, a multi-objective optimization model for the layout of UAV fixed take-off and landing facilities and its constraints are constructed. The layout optimization module is used to solve the multi-objective optimization model using a non-dominated sorting genetic algorithm to obtain the layout optimization scheme for the UAV fixed take-off and landing facility.

Images