Logistics unmanned aerial vehicle airport site selection method based on multi-source data driving
Through the multi-source data-driven method, the location selection of drones is analyzed, and the hazard risk and distribution efficiency of drones in crowded areas are solved, scientific and reasonable airport location selection and risk assessment are achieved, and drones are improved. Delivery efficiency.
Patent Information
- Application Number
- CN202411002238.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-25
- Publication Date
- 2024-11-26
AI Technical Summary
How to choose drone airports to improve distribution efficiency and reduce the risk of drone hazards in crowded areas, insufficient existing monitoring deployment, and lack effective data-driven methods for drone airport site selection and intelligent take-off and landing point selection.
Using a multi-source data-driven method, the target demand prediction algorithm, entropy weight method is used by collecting and analyzing pedestrian flow, obstacle distribution, traffic conditions and building distribution information, combining high-precision lidar, camera and environmental perception sensor, target demand prediction algorithm, entropy weight method, TOPSIS method and multi-objective optimization algorithm generate drone airport site selection scheme, evaluate flight paths, obstacle avoidance and coordinated control, and select the optimal airport location.
It significantly improves the efficiency of drone distribution, reduces the risk of harm in crowded areas, reduces the dependence on the experience of practitioners, and selects a scientific and reasonable drone airport location.
Smart Images

Figure CN119026767A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of logistics and unmanned aerial vehicle (UAV), and in particular relates to a logistics UAV airport site selection method driven by multi-source data. Background Art
[0002] With the rapid development of science and technology, the gradual popularization of drone technology and the gradual development of computer technology, logistics drones have become a reality. Logistics drones have gone from ideal to reality. Compared with traditional methods, logistics drones can ignore the terrain and deliver more efficiently. In addition, logistics drones can greatly reduce labor costs and are safer because drones are light in weight and are controlled by machines, with less human factors. Logistics drones can ensure transportation efficiency and can operate normally in a variety of weather conditions.
[0003] However, logistics drones also face a series of problems. How to select drone airports to make delivery more efficient, make items arrive faster, and reduce the risk of drones harming people in densely populated areas is the key to the application of this service. Improving the resilience of logistics drone services by considering dynamic factors and improving airport site selection is also the key to the promotion and application of logistics drones. In order to achieve drone airport site selection and intelligent take-off and landing point selection, it is necessary to build a data-driven method that can achieve drone airport site selection. Summary of the invention
[0004] The purpose of the present invention is to provide a method for selecting a logistics drone airport based on multi-source data to address the shortcomings of existing monitoring deployments. By mining the spatial information of the logistics drone, a spatial model is established, and an algorithm is used to generate a logistics drone airport location. The method of the present invention can be used to integrate multi-source data to determine the appropriate drone airport location, significantly improving the efficiency of drone delivery.
[0005] In order to achieve the above object, the present invention adopts the following technical solutions:
[0006] A method for selecting a logistics drone airport based on multi-source data, the method comprising the following steps:
[0007] Step (1): Collect information on pedestrian flow, obstacle distribution, traffic conditions, and building distribution in the area at each time period, obtain real-time images of the regional environment, perform pattern recognition and prediction on the collected data, and determine the initial plan for the location of the drone airport based on user delivery needs;
[0008] Step (2): simulate the operation effect of the drone under the initial schemes of different drone airport locations obtained in step (1), consider the flight path planning, obstacle avoidance algorithm and cooperative control factors between drones, so as to comprehensively evaluate the impact of the initial schemes of different drone airport locations obtained in step (1) on regional logistics efficiency;
[0009] Step (3): Based on the simulation results and data analysis results of step (2), a multi-objective optimization algorithm is used to comprehensively evaluate and compare the initial plans for multiple drone airport site selections. The cost-benefit analysis and environmental impact are evaluated through the airport management system to select the optimal drone airport site selection plan.
[0010] Furthermore, the process of collecting pedestrian flow, obstacle distribution, traffic conditions and building distribution information in the area in each time period and obtaining a real-time image of the regional environment in step (1) includes: using a high-precision laser radar carried by a drone to perform drone measurement, and conducting field measurement to obtain map data, wherein the map data includes terrain elevation, three-dimensional coordinates of buildings, target object characteristics and distance information, and then converting the obtained map data into a digital elevation model and a building information model through point cloud data processing and three-dimensional reconstruction of buildings. Suppose the pedestrian flow is P(t); the building distribution is B(x, y), where (x, y) represents the spatial coordinates; the traffic condition is T(t, l), where t represents time and l represents the road section; at the same time, the camera and environmental perception sensor carried by the drone are used to obtain real-time location information and environmental information, and provide a real-time image of the regional environment for subsequent site selection decisions, which is expressed as I(t).
[0011] Furthermore, the step (1) performs pattern recognition and prediction on the collected data, and the specific process is as follows:
[0012] A. Use the target demand forecasting algorithm to count, classify and summarize the collected data; according to the first exponential smoothing method, we know that y t+1 =ax t +(1-a)y t , x t is the measured value at time t, y t is the predicted value at time t, a is the weighting factor, and its value range is 0≤a≤1;
[0013] B. Use the entropy weight method to process the data obtained in step A separately to obtain the information entropy of each indicator, and then calculate the weight of each indicator; the weight formula is H(X i ) is the information entropy of the i-th index, ω i is the weight of the i-th indicator, and n is the number of evaluation indicators;
[0014] C. Based on the weights obtained in step B, use the TOPSIS method to calculate the risk ranking and obtain the relative closeness between each data object and the demand. i , and then comprehensively evaluate and identify areas with high demand for drone airports; according to the proximity formula is the distance between the (i)th alternative and the negative ideal solution, is the distance between the (i)th alternative and the positive ideal solution;
[0015] D. Perform a series of spatial queries on the GIS model to verify the results obtained in step C. The requirements of the GIS model include pedestrian flow P(t), building distribution B(x, y), traffic conditions T(t, l), and the remaining power E(t) of the drone at time t, thereby predicting the number of drones required.
[0016] Furthermore, in the airport site selection process in step (1), based on the delivery needs of local users, a user behavior analysis method is used to perform a cluster analysis on the user order data to determine the concentrated area of delivery demand. After the concentrated area is known, the alternative solution obtained in claim 3 is used to use the Dijkstra algorithm or the A* algorithm to plan the path of the concentrated area logistics, and the basic path under the initial solution of the drone airport site selection is obtained.
[0017] Furthermore, the step (2) comprises the following steps:
[0018] 2.1. Use the quadcopter UAV power model and Euler equation to calculate the position of the UAV in the air. By collecting the position information of the UAV at different times, it is determined whether the UAV site selection can effectively avoid obstacles. The Euler equation is M = Jω b +ω b ×(Jω b ),ω b is the angular velocity of the machine system, where ω b =[pqr] T , p, q, r are the components of angular velocity on the three axes of the machine system, ψ and θ are both attitude angles, J is the moment of inertia, and its matrix is expressed as J = diag(I x I y I z ), the resultant force generated by the thrust f b =[f x f y f z ], Where m is the mass of the drone and its payload; s is the distance the drone travels to avoid obstacles, which is obtained from the real-time position of the drone;
[0019] Simulate the different solutions obtained in step 1 to avoid obstacles and obtain the time t of different solutions to avoid obstacles. Through simulation, sort the time t of different solutions obtained in step 1 and select 20 different drone airport site selection solutions with the least time.
[0020] 2.2, predict the impact of the drone's height h, speed v and load w on the airport's operation:
[0021] In the 20 different site selection schemes obtained in step 2.1, the power consumption of the drone due to the height h, speed v and load w of the drone is simulated by inputting different drone data, and the influence of different drone flight characteristics on the airport operation effect is predicted by combining the drone power consumption. Assuming the drone power is e, a half-hour simulated flight is performed on the fully charged drone, and the power after simulation is recorded. The simulated power is sorted from most to least, and the top 8 drone airport site selection schemes with the highest power from the 20 different site selection schemes obtained in step 2.1 are selected;
[0022] 2.3, predict the impact of different climate conditions on airport operations:
[0023] For the 8 UAV airport site selection schemes obtained in step 2.2, let time be t. By simulating different schemes of UAVs completing a task under wind speeds of 18m / s, 5m / s, 1m / s, 10mm / h of rainfall, 6mm / h of rainfall, and 1mm / h of rainfall, combined with the airport operation effect, the impact of different climatic conditions on the airport operation effect is predicted. Through simulation, the time of the 8 UAV airport site selection schemes obtained in step 2.2 is sorted, and the 3 different UAV airport site selection schemes with the least time are selected.
[0024] Furthermore, the specific method of step (3) is: real-time monitoring of the operating status and location information of the drones corresponding to the drone airport site obtained in step 2.3, evaluating the cost-benefit analysis and environmental impact, wherein the multi-objective optimization algorithm is integrated into the particle swarm optimization algorithm, and each airport site selection plan is represented as a particle, each particle contains a location attribute, and each particle will update its own position through each iteration, and iterate multiple times until the preset maximum number of iterations is reached or the best position obtained in multiple consecutive iterations has no obvious improvement, and comprehensively consider the construction cost, operation effect and environmental impact of the drone airport. These diversified factors, among the drone airport plans screened out in the above process, further use these different factors to solve the optimal solution set of the comprehensive evaluation function Z(x) corresponding to different plans, and select the optimal drone airport site selection plan from the drone airport site selection obtained in step 2.3.
[0025] Furthermore, the comprehensive evaluation function is established as follows:
[0026] Assume that the construction cost of the selected i-th airport site selection plan is C1(Si), the maintenance cost is C2(Si), and the environmental impact is E(Si). Establish a linear regression equation for the comprehensive evaluation function. By calculating the influence of the independent variables on the dependent variables, calculate the respective weight coefficients. The comprehensive evaluation function Z(Si) can be expressed as:
[0027] Z(S i )=α·C1(S i )+β·C1(S i )+θ·E(S i )
[0028] Where Si represents the i-th airport site selection plan, α, β, θ are weight coefficients, reflecting the relative importance of different factors in the decision-making process. By solving the optimal solution set of Z(Si), the optimal drone airport site selection plan is selected from the drone airport sites obtained in step 2.3.
[0029] Compared with the prior art, the present invention has the following beneficial effects:
[0030] The present invention intelligently integrates multi-source data information, reduces reliance on practitioners' site selection experience, comprehensively considers flight paths and risk assessments, selects scientific and reasonable drone airport locations, improves drone delivery efficiency, and reduces the risk of drone damage in densely populated areas. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] Figure 1 The present invention provides a flow chart of the method. DETAILED DESCRIPTION
[0032] The method of the present invention is further described below in conjunction with a specific embodiment. In this embodiment, the Tianmu Lake Campus of Nanjing University of Aeronautics and Astronautics in 2024 is taken as the regional research object, the location set of the research area is N, and there are 7 locations investigated. The spatial data of each location is shown in Table 1.
[0033] Table 1 Location data of Nanjing University of Aeronautics and Astronautics Tianmu Lake Campus in 2024
[0034] Coordinate number Delivery point name Latitude (x) Longitude (y) 1 Library 1st Floor 31.3711 119.48542 2 First canteen entrance 31.3729 119.48527 3 Dormitory Building A, downstairs 31.3729 119.48419 4 Stadium delivery point 31.3693 119.48769 5 Teaching Building A, 1st Floor Lobby 31.3718 119.48599 6 Entrance No. 1 of the Laboratory Building 31.3718 119.48136 7 Teachers' Apartment Delivery Point 31.3734 119.48503
[0035] like Figure 1 As shown, a method for selecting a logistics drone airport based on multi-source data driving of the present invention comprises the following steps:
[0036] Step (1): Collect pedestrian flow, obstacle distribution, traffic conditions and building distribution information in the area in each time period, and obtain real-time images of the regional environment. Specifically, use the high-precision laser radar carried by the drone to perform drone measurement, and conduct field measurements to obtain map data. The map data includes terrain elevation, three-dimensional coordinates of buildings, target characteristics and distance information. Then, through point cloud data processing and three-dimensional reconstruction of buildings, the obtained map data is converted into a digital elevation model and a building information model. Let the pedestrian flow be P(t); building distribution B(x, y), where (x, y) represents spatial coordinates; traffic conditions are T(t, l), where t represents time and l represents road section; at the same time, use the camera and environmental perception sensor carried by the drone to obtain real-time location information and environmental information, and provide a real-time image of the regional environment for subsequent site selection decisions, represented as I(t). Perform pattern recognition and prediction on the collected data, and determine the initial plan for the drone airport site selection based on user delivery needs. The specific process is as follows:
[0037] A. Use the target demand forecasting algorithm to count, classify and summarize the collected data; according to the first exponential smoothing method, we know that y t+1 =ax t +(1-a)y t , x t is the measured value at time t, y t is the predicted value at time t, a is the weighting factor, and its value range is 0≤a≤1;
[0038] B. Use the entropy weight method to process the data obtained in step A separately to obtain the information entropy of each indicator, and then calculate the weight of each indicator; the weight formula is H(X i ) is the information entropy of the i-th index, ω i is the weight of the i-th indicator, and n is the number of evaluation indicators;
[0039] C. Based on the weights obtained in step B, use the TOPSIS method to calculate the risk ranking and obtain the relative closeness between each data object and the demand. i , and then comprehensively evaluate and identify areas with high demand for drone airports; according to the proximity formula is the distance between the (i)th alternative and the negative ideal solution, is the distance between the (i)th alternative and the positive ideal solution;
[0040] D. Perform a series of spatial queries on the GIS model to verify the results obtained in step C. The requirements of the GIS model include pedestrian flow P(t), building distribution B(x, y), traffic conditions T(t, l), and the remaining power E(t) of the drone at time t, thereby predicting the number of drones required.
[0041] In summary, the drone airports numbered 1-7 all meet the take-off and landing requirements of logistics drones and can store goods for a short period of time. Therefore, they can all be candidate airports. At the same time, based on the distribution needs of local users, the user behavior analysis method is used to perform cluster analysis on user order data to determine the concentrated area of distribution needs. After knowing the concentrated area, the initial airport site selection plan obtained in claim 3 uses the Dijkstra algorithm and A* algorithm, which are optimal path planning algorithms, to plan the path of concentrated area logistics, and obtain the basic path that simulates the simulation steps. The priority of the airport is determined based on the demand for the number of logistics drones.
[0042] The results are shown in Table 2.
[0043] Table 2 Initial site selection of drone airports in 2024 based on demand
[0044]
[0045] Based on the delivery needs of local users, user behavior analysis methods are used to perform cluster analysis on user order data to determine the concentrated areas of delivery demand. After knowing the concentrated areas, the alternative solution obtained in claim 3 is used to use the Dijkstra algorithm or the A* algorithm to plan the path of concentrated regional logistics, and the basic path under the initial solution of drone airport site selection is obtained.
[0046] Step (2): Simulate the operation effect of the UAV under the initial schemes of different UAV airport sites obtained in step (1), taking into account the flight path planning, obstacle avoidance algorithm and the coordinated control factors between UAVs, so as to comprehensively evaluate the impact of the initial schemes of different UAV airport sites obtained in step (1) on the regional logistics efficiency; First, the latitude and longitude data of each candidate in the candidate UAV airport sequence set S1 are input into the advanced simulation experiment software, so that the flight state of the UAV under various environmental conditions can be accurately simulated. In the simulation process, the flight dynamics model is used to simulate the basic flight characteristics of the UAV, such as wind resistance, lift, gravity, etc. At the same time, the wind field model is also introduced to simulate the impact of different wind speeds, wind directions and rainfall intensities on the flight of the UAV. The comprehensive application of these models enables the simulation of the flight state of the UAV under complex weather conditions, providing strong support for the evaluation of the site selection scheme. In addition to weather factors, flight path planning factors are also considered. Using advanced path planning algorithms, the optimal flight path can be planned for the UAV according to the terrain, obstacle distribution and flight requirements. Obstacle avoidance is also an important factor to be considered when evaluating the site selection scheme. Using advanced obstacle recognition and avoidance algorithms, ensure that drones can accurately sense and avoid obstacles during flight to ensure flight safety. In addition, collaborative control between drones is also an important part of evaluating site selection. In a logistics drone system, multiple drones need to share transportation routes and cargo information to achieve collaborative operations. Using advanced communication technology and collaborative control algorithms, ensure that drones can exchange information in real time and complete tasks collaboratively. All of the above factors were taken into consideration when selecting the preferred site. The specific steps include:
[0047] 2.1. Use the quadcopter UAV power model and Euler equation to calculate the position of the UAV in the air. By collecting the position information of the UAV at different times, it is determined whether the UAV site selection can effectively avoid obstacles. The Euler equation is M = Jω b +ω b ×(Jω b ),ω b is the angular velocity of the machine system, where ω b =[pqr] T , p, q, r are the components of angular velocity on the three axes of the machine system, ψ and θ are both attitude angles, J is the moment of inertia, and its matrix is expressed as J = diag(I x I y I z ), the resultant force generated by the thrust f b =[f x f y f z ], Where m is the mass of the drone and its payload; s is the distance the drone travels to avoid obstacles, which is obtained from the real-time position of the drone;
[0048] Simulate the different solutions obtained in step 1 to avoid obstacles and obtain the time t of different solutions to avoid obstacles. Through simulation, sort the time t of different solutions obtained in step 1 and select 20 different drone airport site selection solutions with the least time.
[0049] 2.2, predict the impact of the drone's height h, speed v and load w on the airport's operation effect:
[0050] In the 20 different site selection schemes obtained in step 2.1, the power consumption of the drone due to the height h, speed v and load w of the drone is simulated by inputting different drone data, and the influence of different drone flight characteristics on the airport operation effect is predicted by combining the drone power consumption. Assuming the drone power is e, a half-hour simulated flight is performed on the fully charged drone, and the power after simulation is recorded. The simulated power is sorted from most to least, and the top 8 drone airport site selection schemes with the highest power from the 20 different site selection schemes obtained in step 2.1 are selected;
[0051] 2.3, predict the impact of different climate conditions on airport operations:
[0052] For the 8 UAV airport site selection schemes obtained in step 2.2, let time be t. By simulating different schemes of UAVs completing a task under wind speeds of 18m / s, 5m / s, 1m / s, 10mm / h of rainfall, 6mm / h of rainfall, and 1mm / h of rainfall, combined with the airport operation effect, the impact of different climatic conditions on the airport operation effect is predicted. Through simulation, the time of the 8 UAV airport site selection schemes obtained in step 2.2 is sorted, and the 3 different UAV airport site selection schemes with the least time are selected.
[0053] Then, the site selection plan is comprehensively evaluated based on the cost-benefit analysis and environmental impact assessment; finally, the better site selection plan is selected based on the evaluation results.
[0054] The results are shown in Table 3.
[0055] Table 3 The simulation system simulates the situation of drone airport
[0056] Airport Flight Path Obstacle Avoidance Collaborative Control Drone Airport 1 Short distance Many obstacles No collaborative control Drone Airport 2 Long distance Few obstacles Collaborative control Drone Airport 3 Long distance Few obstacles No collaborative control Drone Airport 4 Short distance Few obstacles No collaborative control Drone Airport 5 Short distance Few obstacles Collaborative control Drone Airport 6 Long distance Many obstacles Collaborative control Drone Airport 7 Short distance Few obstacles Collaborative control Drone Airport 8 Long distance Many obstacles No collaborative control
[0057] Step (3): Based on the simulation results and data analysis results of step (2), a multi-objective optimization algorithm is used to comprehensively evaluate and compare the initial plans for multiple drone airport site selections. The cost-benefit analysis and environmental impact are evaluated through the airport management system to select the optimal drone airport site selection plan. Specifically, the operation status and location information of the drones corresponding to the drone airport site selection obtained in step 2.3 are monitored in real time to evaluate the cost-benefit analysis and environmental impact. The multi-objective optimization algorithm is integrated into the particle swarm optimization algorithm, and each airport site selection plan is represented as a particle. Each particle contains a position attribute. Through each iteration, each particle will update its own position according to the preset maximum number of iterations or the best position obtained in multiple consecutive iterations has no significant improvement. In addition, the construction cost, operation effect and environmental impact of the drone airport are comprehensively considered. Among the drone airport plans selected in the above process, the optimal solution set of the comprehensive evaluation function Z(x) corresponding to different plans is further solved by using these different factors, and the optimal drone airport site selection plan is selected from the drone airport site selection obtained in step 2.3.
[0058] The comprehensive evaluation function is established as follows:
[0059] Assume that the construction cost of the selected i-th airport site selection plan is C1(Si), the maintenance cost is C2(Si), and the environmental impact is E(Si). Establish a linear regression equation for the comprehensive evaluation function. By calculating the influence of the independent variables on the dependent variables, calculate the respective weight coefficients. The comprehensive evaluation function Z(Si) can be expressed as:
[0060] Z(S i )=α·C1(S i )+β·C1(S i )+θ·E(S i )
[0061] Where Si represents the i-th airport site selection plan, α, β, θ are weight coefficients, reflecting the relative importance of different factors in the decision-making process. By solving the optimal solution set of Z(Si), the optimal drone airport site selection plan is selected from the drone airport sites obtained in step 2.3.
[0062] The results are shown in Table 4.
[0063] Table 4. UAV airport site selection decision results
[0064]
[0065] The spatiotemporal location and user data information of the location are effectively utilized and used as decision information for optimizing the site selection method. Especially for the case where the path from the logistics drone airport location to the target location is long and there are many obstacles, the mining and analysis of location and user data can help provide a decision solution for the site selection of the logistics drone airport based on multi-source data.
[0066] The above content is only a further detailed description of the present invention through a specific embodiment, and it cannot be determined that the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the scope of protection of the present invention.
Claims
1. A logistics drone airport site selection method based on multi-source data driven, characterized by: The method comprises the following steps: Step (1): Collect information on pedestrian flow, obstacle distribution, traffic conditions, and building distribution in the area at each time period, obtain real-time images of the regional environment, perform pattern recognition and prediction on the collected data, and determine the initial plan for the location of the drone airport based on user delivery needs; Step (2): simulate the operation effect of the drone under the initial schemes of different drone airport locations obtained in step (1), consider the flight path planning, obstacle avoidance algorithm and cooperative control factors between drones, so as to comprehensively evaluate the impact of the initial schemes of different drone airport locations obtained in step (1) on regional logistics efficiency; Step (3): Based on the simulation results and data analysis results of step (2), a multi-objective optimization algorithm is used to comprehensively evaluate and compare the initial plans for multiple drone airport site selections. The cost-benefit analysis and environmental impact are evaluated through the airport management system to select the optimal drone airport site selection plan.
2. The method for selecting a logistics drone airport based on multi-source data drive according to claim 1 is characterized in that: The process of collecting pedestrian flow, obstacle distribution, traffic conditions and building distribution information in the area in each time period and obtaining a real-time image of the regional environment in step (1) includes: using a high-precision laser radar carried by a drone to perform drone measurement, and performing field measurement to obtain map data, wherein the map data includes terrain elevation, three-dimensional coordinates of buildings, target object characteristics and distance information, and then converting the obtained map data into a digital elevation model and a building information model through point cloud data processing and three-dimensional reconstruction of buildings. Suppose the pedestrian flow is P(t); the building distribution is B(x, y), where (x, y) represents the spatial coordinates; the traffic condition is T(t, l), where t represents time and l represents the road section; at the same time, the camera and environmental perception sensor carried by the drone are used to obtain real-time location information and environmental information, and provide a real-time image of the regional environment for subsequent site selection decisions, which is expressed as I(t).
3. The method for selecting a logistics drone airport based on multi-source data drive according to claim 1 is characterized in that: The step (1) of performing pattern recognition and prediction on the collected data is specifically as follows: A. Use the target demand forecasting algorithm to count, classify and summarize the collected data; according to the first exponential smoothing method, we know that y t+1 =ax t +(1-a)y t , x t is the measured value at time t, y t is the predicted value at time t, a is the weighting factor, and its value range is 0≤a≤1; B. Use the entropy weight method to process the data obtained in step A separately to obtain the information entropy of each indicator, and then calculate the weight of each indicator; the weight formula is H(X i ) is the information entropy of the i-th index, ω i is the weight of the i-th indicator, and n is the number of evaluation indicators; C. Based on the weights obtained in step B, use the TOPSIS method to calculate the risk ranking and obtain the relative closeness between each data object and the demand. i , and then comprehensively evaluate and identify areas with high demand for drone airports; according to the proximity formula is the distance between the (i)th alternative and the negative ideal solution, is the distance between the (i)th alternative and the positive ideal solution; D. Perform a series of spatial queries on the GIS model to verify the results obtained in step C. The requirements of the GIS model include pedestrian flow P(t), building distribution B(x, y), traffic conditions T(t, l), and the remaining power E(t) of the drone at time t, thereby predicting the number of drones required.
4. The method for selecting a logistics drone airport based on multi-source data drive according to claim 3 is characterized in that: In the airport site selection process in step (1), based on the delivery needs of local users, a user behavior analysis method is first used to perform a cluster analysis on the user order data to determine the concentrated area of delivery demand. After the concentrated area is known, the alternative solution obtained in claim 3 is used to use the Dijkstra algorithm or the A* algorithm to plan the path of the concentrated area logistics, and the basic path under the initial solution of the drone airport site selection is obtained.
5. The method for selecting a logistics drone airport based on multi-source data drive according to claim 1 is characterized in that: The step (2) comprises the following steps: 2.
1. Use the quadcopter UAV power model and Euler equation to calculate the position of the UAV in the air. By collecting the position information of the UAV at different times, it is determined whether the UAV site selection can effectively avoid obstacles. The Euler equation is M = Jω b +ω b ×(Jω b ),ω b is the angular velocity of the machine system, where ω b =[pqr] T , p, q, r are the components of angular velocity on the three axes of the machine system, ψ and θ are both attitude angles, J is the moment of inertia, and its matrix is expressed as J = diag(I x I y I z ), the resultant force generated by the thrust Where m is the mass of the drone and its payload; s is the distance the drone travels to avoid obstacles, which is obtained from the real-time position of the drone; Simulate the different solutions obtained in step 1 to avoid obstacles and obtain the time t of different solutions to avoid obstacles. Through simulation, sort the time t of different solutions obtained in step 1 and select 20 different drone airport site selection solutions with the least time. 2.2, predict the impact of the drone's height h, speed v and load w on the airport's operation effect: In the 20 different site selection schemes obtained in step 2.1, the power consumption of the drone due to the height h, speed v and load w of the drone is simulated by inputting different drone data, and the influence of different drone flight characteristics on the airport operation effect is predicted by combining the drone power consumption. Assuming the drone power is e, a half-hour simulated flight is performed on the fully charged drone, and the power after simulation is recorded. The simulated power is sorted from most to least, and the top 8 drone airport site selection schemes with the highest power from the 20 different site selection schemes obtained in step 2.1 are selected; 2.3, predict the impact of different climate conditions on airport operations: For the 8 UAV airport site selection schemes obtained in step 2.2, let time be t. By simulating different schemes of UAVs completing a task under wind speeds of 18m / s, 5m / s, 1m / s, 10mm / h of rainfall, 6mm / h of rainfall, and 1mm / h of rainfall, combined with the airport operation effect, the impact of different climatic conditions on the airport operation effect is predicted. Through simulation, the time of the 8 UAV airport site selection schemes obtained in step 2.2 is sorted, and the 3 different UAV airport site selection schemes with the least time are selected.
6. The method for selecting a logistics drone airport based on multi-source data drive according to claim 1 is characterized in that: The specific method of step (3) is: real-time monitoring of the operating status and location information of the drones corresponding to the drone airport site obtained in step 2.3, evaluating the cost-benefit analysis and environmental impact, wherein the multi-objective optimization algorithm is integrated into the particle swarm optimization algorithm, and each airport site selection plan is represented as a particle, each particle contains a location attribute, and each particle will update its own position through each iteration, and iterate multiple times until the preset maximum number of iterations is reached or the best position obtained in multiple consecutive iterations has no obvious improvement, and comprehensively consider the construction cost, operation effect and environmental impact of the drone airport. These diversified factors, among the drone airport plans screened out in the above process, further use these different factors to solve the optimal solution set of the comprehensive evaluation function Z(x) corresponding to different plans, and select the optimal drone airport site selection plan from the drone airport site selection obtained in step 2.
3.
7. The method for selecting a logistics drone airport based on multi-source data drive according to claim 6 is characterized in that: The comprehensive evaluation function is established as follows: Assume that the construction cost of the selected i-th airport site selection plan is C1(Si), the maintenance cost is C2(Si), and the environmental impact is E(Si). Establish a linear regression equation for the comprehensive evaluation function. By calculating the influence of the independent variables on the dependent variables, calculate the respective weight coefficients. The comprehensive evaluation function Z(Si) can be expressed as: Z(S i )=α·C1(S i )+β·C1(S i )+θ·E(S i ) Where Si represents the i-th airport site selection plan, α, β, θ are weight coefficients, reflecting the relative importance of different factors in the decision-making process. By solving the optimal solution set of Z(Si), the optimal drone airport site selection plan is selected from the drone airport sites obtained in step 2.3.
Citation Information
Patent Citations
UAM site selection evaluation method and unmanned aerial vehicle inspection route planning method
CN116203984A
Unmanned aerial vehicle rapid path planning method and system based on deep learning
CN116449860A
Logistics transportation point network planning method in low-altitude airspace
CN116452105A
Multi-mode traffic simulation evaluation method and system for airport comprehensive transportation hub
CN118278169A
Cited By
Unmanned aerial vehicle scheduling method and device, computer equipment and storage medium
CN119312990A
UAV dispatching method, device, computer equipment and storage medium
CN119312990B
Unmanned aerial vehicle airport multi-target constraint planning and site selection method and system and electronic equipment
CN119397817A
Unmanned aerial vehicle take-off and landing point intelligent site selection method and system for low-altitude logistics
CN120235367A
Unmanned aerial vehicle airport site selection method suitable for large-range power line inspection
CN120409760A
