A method for selecting drone take-off and landing points for urban terminal logistics

Abstract

This invention discloses a method for selecting drone takeoff and landing points for urban terminal logistics, relating to the field of facility site selection technology. The method comprises the following steps: S1. Obtaining building outlines in the area to be sited and constructing a customer buffer zone based on the building outlines; S2. Based on the customer buffer zone, performing noise and safety analyses on drones to construct an overall restricted area; S3. Based on the overall restricted area, constructing an objective function with minimum cost as the optimization goal, an objective function with maximum satisfaction as the optimization goal, and several constraints to obtain a site selection layout result. This invention verifies the applicability of the grid mapping method in complex urban environments and provides new ideas for the layout of drone takeoff and landing points for urban logistics.

Description

Technical Field

[0001] The present invention relates to the technical field of facility site selection, and in particular to a method for selecting take-off and landing points for unmanned aerial vehicle (UAV) for urban terminal logistics. Background Art

[0002] Among the many applications of the low-altitude economy, drone logistics and delivery is an emerging model. Using drones or other low-altitude aircraft to transport goods offers advantages such as timeliness, low cost, and flexibility. Drone logistics and delivery can effectively address the "last mile" challenge, quickly and accurately delivering urgently needed supplies to their destinations, significantly improving logistics efficiency and user experience. Using take-off and landing points as delivery nodes for drones can address the drone's energy consumption and flight range, expand its delivery range, and facilitate dispatch by management personnel.

[0003] Landing and take-off points for drone delivery are primarily categorized into two models: fixed-landing and take-off points for drone delivery and mobile-landing and take-off points for drone delivery. For urban terminal logistics delivery, drone delivery targets are relatively complex, characterized by small batch sizes and high frequency, while landing and take-off points must be intelligent and information-based. Therefore, fixed-landing and take-off points for drone delivery are more suitable for scenarios such as urban terminal logistics and just-in-time delivery. However, existing technologies often oversimplify the application environment of fixed-landing and take-off point layouts, ignoring the impact of noise and safety on satisfaction in real-world scenarios. They also fail to consider the multi-tiered facility model for terminal landing and take-off points to minimize waste of delivery resources. Furthermore, the site selection for fixed-landing and take-off points can be categorized into continuous, discrete, and comprehensive evaluation methods. To reduce environmental exploration and thus computational efficiency, a discrete site selection model will be used as the site selection model.

[0004] Therefore, there is an urgent need for technical improvement, namely, to develop a method for selecting drone take-off and landing points for urban terminal logistics to meet the needs of different customers, complete the demand delivery of take-off and landing points at the end of the city under a series of constraints such as environmental friendliness, and specify the distribution routes of the terminal take-off and landing points to meet the delivery needs within the service range. Summary of the Invention

[0005] In order to solve the above problems, the present invention proposes a method for selecting take-off and landing points for UAVs for urban terminal logistics.

[0006] The technical solution of the present invention is: a method for selecting a take-off and landing point for drones for urban terminal logistics, characterized by comprising the following steps:

[0007] S1. Obtain the building outline of the area to be selected and construct a customer buffer zone based on the building outline;

[0008] S2. Based on the customer buffer zone, conduct noise and safety analysis on the drone and construct the overall restricted area;

[0009] S3. Based on the overall restricted area, construct an objective function with the lowest cost as the optimization goal, an objective function with the highest satisfaction as the optimization goal, and several constraints to obtain the site selection and layout results.

[0010] Furthermore, S1 includes the following sub-steps:

[0011] S11, obtaining the building outline of the area to be selected, and taking the centroid of the building outline as the initial position of the customer point;

[0012] S12, using the DBSCAN method to perform a preliminary screening of the initial positions of the customer points;

[0013] S13. Based on the ratio of the number of drone deliveries to the total number of deliveries in the selected location area, random sampling is performed to perform a secondary screening of the initial customer point locations to obtain the final customer point location distribution;

[0014] S14. Construct a customer buffer zone based on the final customer point location distribution and the noise during the drone take-off and landing process.

[0015] Furthermore, S2 includes the following sub-steps:

[0016] S21. Build a free-field model for the drone based on the client buffer zone;

[0017] S22. determining the sound intensity of the semi-free field according to the noise source of the semi-free field;

[0018] S23. Determine a sound intensity level attenuation model of the semi-free field based on the free field model and the sound intensity of the semi-free field;

[0019] S24. Determine the spherical propagation radius of noise based on the sound intensity level attenuation model in a semi-free field;

[0020] S25. Determine the noise level radius of the take-off and landing points based on the noise spherical propagation radius;

[0021] S26. Determine the horizontal constant force provided by the wind to the UAV within the noise level radius of the take-off and landing point;

[0022] S27, determining the vertical speed and flight altitude of the drone when the drone is falling, and generating the drone falling time;

[0023] S28. Determine the horizontal velocity of the drone in the falling state based on the horizontal constant force provided by the wind to the drone and the falling time of the drone;

[0024] S29. Determine the horizontal displacement radius of the drone based on the horizontal velocity of the drone in the falling state;

[0025] S210: Determine a horizontal impact radius of the UAV based on the horizontal displacement radius of the UAV;

[0026] S211. Determine the environmental buffer zone radius based on the UAV's horizontal impact radius;

[0027] S212. Determine a building buffer zone based on the environmental buffer zone radius, using the actual road radius as the road network buffer zone and public places and no-fly zones as other buffer zones;

[0028] S213. Treat building buffer zones, road network buffer zones and other buffer zones as overall restricted areas.

[0029] Furthermore, in S21, the expression of the free field model is:

[0030] ;

[0031] Where, Δ L Represents the sound intensity level difference between the sound source and the target position, r 1 represents the initial position, r 2 represents the target position, log 10 (·) represents the logarithmic function with base 10;

[0032] In S22, the sound intensity in the semi-free field I The calculation formula is:

[0033] ;

[0034] Where, W represents the sound source power, r Indicates the distance the sound source travels to the sphere;

[0035] In S23, the expression of the sound intensity level attenuation model in a semi-free field is:

[0036] ;

[0037] Where, L I is the sound intensity level in a semi-free field, W 0 represents the reference sound power, r noise represents the spherical propagation radius of the noise, L W represents the sound power level of the drone, lg(·) represents the logarithmic function with base e;

[0038] In S25, the noise level radius of the take-off and landing points rn,x The calculation formula is:

[0039] ;

[0040] Where, r n,y Indicates the vertical radius of noise at the take-off and landing point.

[0041] Furthermore, in S26, the wind provides a horizontal constant force to the drone. F wind The calculation formula is:

[0042] ;

[0043] Where, ρ represents the air density, v w Indicates wind speed, C d represents the drag coefficient, A u represents the drone contact area;

[0044] In S27, the vertical speed of the drone in the falling state v y The calculation formula is:

[0045] ;

[0046] Where, M Indicates the UAV mission payload weight, g represents the acceleration due to gravity, A u,y Represents the contact area in the vertical direction;

[0047] In S28, the horizontal speed of the drone in the falling state v x The calculation formula is:

[0048] ;

[0049] Where, A u,x represents the contact area in the horizontal direction, t represents the drone drop time, tanh(·) represents the hyperbolic tangent function;

[0050] In S29, the horizontal displacement radius of the drone r u , x The calculation formula is:

[0051] ;

[0052] In S210, the horizontal impact radius of the drone r d The calculation formula is:

[0053] ;

[0054] Where, r p represents the average pedestrian radius;

[0055] In S211, the environmental buffer radius r b The calculation formula is:

[0056] ;

[0057] Where, μ 1 represents the noise weight, μ 2 represents the safety weight, r n , x Indicates the noise level radius of the take-off and landing point.

[0058] Furthermore, S3 includes the following sub-steps:

[0059] S31, gridding the area to be selected, performing difference analysis between the gridding result and the overall restricted area to obtain a grid map;

[0060] S32. Based on the grid map, construct an objective function with the minimum cost as the optimization goal;

[0061] S33. Construct an objective function with the highest satisfaction as the optimization goal;

[0062] S34. Construct constraints on the number of take-off and landing points, capacity constraints, take-off and landing point type and layout constraints, customer demand constraints, and service scope constraints;

[0063] S35. Normalizing the objective function with the lowest cost as the optimization goal and the objective function with the highest satisfaction as the optimization goal, respectively, and performing a weighted summation of the normalized objective function with the lowest cost as the optimization goal and the objective function with the highest satisfaction as the optimization goal to obtain a normalized objective function;

[0064] S36. Based on the take-off and landing point quantity constraint, capacity constraint, take-off and landing point type and layout constraint, customer demand constraint, service scope constraint, and normalized objective function, a site selection and layout result is obtained.

[0065] Furthermore, in S32, the expression of the objective function with the minimum cost as the optimization goal is:

[0066] ;

[0067] Where, C represents the total daily economic cost, C f,n represents the average daily fixed cost of the take-off and landing points taking into account depreciation, C v represents the variable cost of the take-off and landing points, y ij It represents the demand quantity allocated to customers at the take-off and landing points, m represents the total number of logistics demand points, n Indicates the total number of alternative take-off and landing points, C L represents the construction cost of the first-level take-off and landing point, C S represents the construction cost of the secondary landing and take-off point, f i Indicates the type of facility, x i Indicates alternative take-off and landing points. c d represents the unit transportation cost, d ij Indicates alternative take-off and landing points i and demand points j The Euclidean distance of i Indicates the sequence number of the alternative take-off and landing point. j Indicates the sequence number of the logistics demand point, t d Indicates the facility depreciation time;

[0068] In S33, the expression of the objective function with the highest satisfaction as the optimization goal is:

[0069] ;

[0070] Where, S Indicates overall customer satisfaction, S ( t ij ) represents the concave-convex time satisfaction function, t ij Indicates that the drone is taking off from an alternative take-off and landing point i To the demand point j The delivery time, D j Indicates demand points j The logistics demand, v d Indicates delivery speed, t L represents the absolute satisfaction threshold, t B represents the absolute dissatisfaction threshold,k i Represents the time sensitivity coefficient.

[0071] Furthermore, in S34, the expression of the take-off and landing point quantity constraint condition is:

[0072] ;

[0073] Where, x i Indicates alternative take-off and landing points. n Indicates the total number of alternative take-off and landing points;

[0074] In S34, the capacity constraint condition is expressed as:

[0075] ;

[0076] Where, y ij It represents the demand quantity allocated to customers at the take-off and landing points, f i Indicates the type of facility, V L,i Indicates the capacity of the first-level take-off and landing point, V S,i Indicates the capacity of the secondary take-off and landing point;

[0077] In S34, the expressions for take-off and landing point types and layout constraints are:

[0078] ;

[0079] Where, I Indicates a set of alternative take-off and landing points;

[0080] In S34, the expression of customer demand constraint is:

[0081] ;

[0082] Where, m represents the total number of logistics demand points, h i Indicate customer needs;

[0083] In S34, the expression of the service range constraint is:

[0084] ;

[0085] Where, d ij Indicates alternative take-off and landing points i and demand points j The Euclidean distance of R L Indicates the service radius of the first-level take-off and landing point,R S Indicates the service radius of the secondary take-off and landing point, J Represents a set of demand points.

[0086] Furthermore, in S35, the calculation formula for normalizing the objective function with the minimum cost as the optimization target is:

[0087] ;

[0088] Where, C represents the total daily economic cost, C n Indicates the reverse target, C (k) It represents the total cost when all the alternative take-off and landing points are established as level 2 take-off and landing points;

[0089] In S35, the calculation formula for normalizing the objective function with the highest satisfaction as the optimization target is:

[0090] ;

[0091] Where, S Express satisfaction, S n Indicates a positive goal, S (k) It indicates the total customer satisfaction when the time satisfaction of each express delivery reaches 1;

[0092] In S35, the calculation formula for weighted summation is:

[0093] ;

[0094] Where, N represents the comprehensive evaluation index, ω 1 represents the economic cost target weight, ω 2 represents the customer satisfaction target weight.

[0095] Furthermore, in S36, based on the constraints on the number of take-off and landing points, the capacity constraints, the type and layout constraints of the take-off and landing points, the customer demand constraints, and the service scope constraints, the Gurobi optimizer is used to solve the normalized objective function to obtain the site selection and layout results.

[0096] The beneficial effects of the present invention are:

[0097] (1) The present invention first analyzes the noise impact and calculates the noise impact range of the drone on the surrounding area during take-off and landing by establishing a noise propagation model. Based on the noise propagation attenuation formula, the noise impact radius is determined, and a noise buffer zone is set on this basis to avoid the layout of take-off and landing points in noise-sensitive areas (such as schools, hospitals and residential areas). Then, based on the safety risk assessment, the possible fall risk of the drone during take-off and landing is analyzed. By establishing a safety assessment model, the impact range of the drone falling is calculated, and on this basis, a safety buffer zone is set to ensure the rationality of the take-off and landing point layout.

[0098] (2) The present invention takes into account the capacity and service range limitations of take-off and landing points, and thus establishes two levels of take-off and landing points, thereby reducing the number of constructions and making better use of the storage space of the take-off and landing points. An integer programming model is developed with construction cost and satisfaction as optimization objectives. Based on this model, the characteristics of the drone and the take-off and landing points are combined, and the Gurobi optimizer is used for solution. Finally, a number of first-level take-off and landing points and second-level take-off and landing points are constructed, completing the full coverage of the demand.

[0099] (3) Compared with the site selection method of the K-means algorithm, the comprehensive cost of the present invention was reduced by 0.8%, which verified the applicability of the grid map method in complex urban environments and provided a new idea for the layout of take-off and landing points for urban logistics drones. A take-off and landing point site selection model was established and solved using an optimizer to obtain the secondary site selection result, that is, the optimal take-off and landing point allocation route. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] Figure 1 A flowchart of the method for selecting drone take-off and landing points for urban terminal logistics;

[0101] Figure 2 Schematic diagram of the drone-takeoff and landing point delivery mode of the present invention;

[0102] Figure 3 A diagram showing the degree of change in the satisfaction curve of the present invention;

[0103] Figure 4 This is a schematic diagram of the urban building and road network distribution in the research scenario of the present invention;

[0104] Figure 5 This is a diagram showing the initial screening results of customer point locations in an embodiment of the present invention;

[0105] Figure 6 This is a diagram showing the secondary screening results of customer point locations in an embodiment of the present invention;

[0106] Figure 7 Schematic diagram of noise radius in an embodiment of the present invention;

[0107] Figure 8This is a schematic diagram of a safety radius in an embodiment of the present invention;

[0108] Figure 9 Establishing a map for a buffer zone of an embodiment of the present invention;

[0109] Figure 10 Establishing a graph for the primary selection grid of an embodiment of the present invention;

[0110] Figure 11 A grid and buffer screening diagram according to an embodiment of the present invention;

[0111] Figure 12 This is the layout diagram of the drone take-off and landing points after solution optimization according to an embodiment of the present invention. DETAILED DESCRIPTION

[0112] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0113] like Figure 1 As shown, the present invention provides a method for selecting a take-off and landing point for UAVs for urban terminal logistics, comprising the following steps:

[0114] S1. Obtain the building outline of the area to be selected and construct a customer buffer zone based on the building outline;

[0115] S2. Based on the customer buffer zone, conduct noise and safety analysis on the drone and construct the overall restricted area;

[0116] S3. Based on the overall restricted area, construct an objective function with the lowest cost as the optimization goal, an objective function with the highest satisfaction as the optimization goal, and several constraints to obtain the site selection and layout results.

[0117] In this embodiment of the present invention, S1 includes the following sub-steps:

[0118] S11, obtaining the building outline of the area to be selected, and taking the centroid of the building outline as the initial position of the customer point;

[0119] S12, using the DBSCAN method to perform a preliminary screening of the initial positions of the customer points;

[0120] S13. Based on the ratio of the number of drone deliveries to the total number of deliveries in the selected location area, random sampling is performed to perform a secondary screening of the initial customer point locations to obtain the final customer point location distribution;

[0121] S14. Construct a customer buffer zone based on the final customer point location distribution and the noise during the drone take-off and landing process.

[0122] In this embodiment of the present invention, S2 includes the following sub-steps:

[0123] S21. Build a free-field model for the drone based on the client buffer zone;

[0124] S22. determining the sound intensity of the semi-free field according to the noise source of the semi-free field;

[0125] S23. Determine a sound intensity level attenuation model of the semi-free field based on the free field model and the sound intensity of the semi-free field;

[0126] S24. Determine the spherical propagation radius of noise based on the sound intensity level attenuation model in a semi-free field;

[0127] S25. Determine the noise level radius of the take-off and landing points based on the noise spherical propagation radius;

[0128] S26. Determine the horizontal constant force provided by the wind to the UAV within the noise level radius of the take-off and landing point;

[0129] S27, determining the vertical speed and flight altitude of the drone when the drone is falling, and generating the drone falling time;

[0130] S28. Determine the horizontal velocity of the drone in the falling state based on the horizontal constant force provided by the wind to the drone and the falling time of the drone;

[0131] S29. Determine the horizontal displacement radius of the drone based on the horizontal velocity of the drone in the falling state;

[0132] S210: Determine a horizontal impact radius of the UAV based on the horizontal displacement radius of the UAV;

[0133] S211. Determine the environmental buffer zone radius based on the UAV's horizontal impact radius;

[0134] S212. Determine a building buffer zone based on the environmental buffer zone radius, using the actual road radius as the road network buffer zone and public places and no-fly zones as other buffer zones;

[0135] S213. Treat building buffer zones, road network buffer zones and other buffer zones as overall restricted areas.

[0136] In the embodiment of the present invention, in S21, the expression of the free field model is:

[0137] ;

[0138] Where, Δ L Represents the sound intensity level difference between the sound source and the target position, r 1 represents the initial position, r 2 represents the target position, log 10 (·) represents the logarithmic function with base 10;

[0139] In S22, the sound intensity in the semi-free field I The calculation formula is:

[0140] ;

[0141] Where, W represents the sound source power, r Indicates the distance the sound source travels to the sphere;

[0142] In S23, the expression of the sound intensity level attenuation model in a semi-free field is:

[0143] ;

[0144] Where, L I is the sound intensity level in a semi-free field, W 0 represents the reference sound power, r noise represents the spherical propagation radius of the noise, L W represents the sound power level of the drone, lg(·) represents the logarithmic function with base e;

[0145] In S25, the noise level radius of the take-off and landing points r n,x The calculation formula is:

[0146] ;

[0147] Where, r n,y Indicates the vertical radius of noise at the take-off and landing point.

[0148] In the embodiment of the present invention, in S26, the wind force is the horizontal constant force provided by the drone. F wind The calculation formula is:

[0149] ;

[0150] Where, ρ represents the air density, v w Indicates wind speed, C d represents the drag coefficient, A u represents the drone contact area;

[0151] In S27, the vertical speed of the drone in the falling state v y The calculation formula is:

[0152] ;

[0153] Where, M Indicates the UAV mission payload weight, g represents the acceleration due to gravity, A u,y Represents the contact area in the vertical direction;

[0154] In S28, the horizontal speed of the drone in the falling state v x The calculation formula is:

[0155] ;

[0156] Where, A u,x represents the contact area in the horizontal direction, t represents the drone drop time, tanh(·) represents the hyperbolic tangent function;

[0157] Assuming there is a constant wind force in the horizontal direction, the wind speed affects the horizontal motion of the drone. Considering the effect of air resistance on the horizontal speed, we can get , a x represents the horizontal acceleration, F f represents the air resistance. Further differentiation yields ; Finally, the horizontal speed of the drone in the falling state is obtained by simplification v x .

[0158] In S29, the horizontal displacement radius of the drone r u , x The calculation formula is:

[0159] ;

[0160] In S210, the horizontal impact radius of the drone r d The calculation formula is:

[0161] ;

[0162] Where, r p represents the average pedestrian radius;

[0163] In S211, the environmental buffer radius r b The calculation formula is:

[0164] ;

[0165] Where, μ 1 represents the noise weight, μ 2 represents the safety weight, r n , x Indicates the noise level radius of the take-off and landing points. In this embodiment of the present invention, S3 includes the following sub-steps:

[0166] S31, gridding the area to be selected, performing difference analysis between the gridding result and the overall restricted area to obtain a grid map;

[0167] S32. Based on the grid map, construct an objective function with the minimum cost as the optimization goal;

[0168] S33. Construct an objective function with the highest satisfaction as the optimization goal;

[0169] S34. Construct constraints on the number of take-off and landing points, capacity constraints, take-off and landing point type and layout constraints, customer demand constraints, and service scope constraints;

[0170] S35. Normalizing the objective function with the lowest cost as the optimization goal and the objective function with the highest satisfaction as the optimization goal, respectively, and performing a weighted summation of the normalized objective function with the lowest cost as the optimization goal and the objective function with the highest satisfaction as the optimization goal to obtain a normalized objective function;

[0171] S36. Based on the take-off and landing point quantity constraint, capacity constraint, take-off and landing point type and layout constraint, customer demand constraint, service scope constraint, and normalized objective function, a site selection and layout result is obtained.

[0172] In an embodiment of the present invention, in S31-S32, a grid map is established. In a real scenario, the site selection space satisfies the three-dimensional spherical area, and the hexagonal grid has a greater filling advantage. Therefore, hexagons are used to fill the site selection area. The grid side length of the initial site selection is determined according to the construction radius of the take-off and landing point, and grid coverage is used for site selection.

[0173] Performing difference analysis on the grid and the buffer zone, the obtained grid graphic has an irregular shape due to the intersection or cutting with the buffer zone boundary. The cut part does not meet the alternative conditions. Therefore, the area elimination method is used to eliminate grids with an area larger or smaller than the hexagonal grid.

[0174] In the embodiment of the present invention, in S32, the expression of the objective function with the lowest cost as the optimization goal is:

[0175] ;

[0176] Where, Crepresents the total daily economic cost, C f,n represents the average daily fixed cost of the take-off and landing points taking into account depreciation, C v represents the variable cost of the take-off and landing points, y ij It represents the demand quantity allocated to customers at the take-off and landing points, m represents the total number of logistics demand points, n Indicates the total number of alternative take-off and landing points, C L represents the construction cost of the first-level take-off and landing point, C S represents the construction cost of the secondary landing and take-off point, f i Indicates the type of facility, x i Indicates alternative take-off and landing points. c d represents the unit transportation cost, d ij Indicates alternative take-off and landing points i and demand points j The Euclidean distance of i Indicates the sequence number of the alternative take-off and landing point. j Indicates the sequence number of the logistics demand point, t d Indicates the facility depreciation time;

[0177] In S33, the expression of the objective function with the highest satisfaction as the optimization goal is:

[0178] ;

[0179] Where, S Indicates overall customer satisfaction, S ( t ij ) represents the concave-convex time satisfaction function, t ij Indicates that the drone is taking off from an alternative take-off and landing point i To the demand point j The delivery time, D j Indicates demand points j The logistics demand, v d Indicates delivery speed, t L represents the absolute satisfaction threshold, t B represents the absolute dissatisfaction threshold, k i Represents the time sensitivity coefficient.

[0180] In the embodiment of the present invention, in S34, the expression of the take-off and landing point quantity constraint condition is:

[0181] ;

[0182] Where, x i Indicates alternative take-off and landing points. n Indicates the total number of alternative take-off and landing points;

[0183] In S34, the capacity constraint condition is expressed as:

[0184] ;

[0185] Where, y ij It represents the demand quantity allocated to customers at the take-off and landing points, f i Indicates the type of facility, V L,i Indicates the capacity of the first-level take-off and landing point, V S,i Indicates the capacity of the secondary take-off and landing point;

[0186] In S34, the expressions for take-off and landing point types and layout constraints are:

[0187] ;

[0188] Where, I Indicates a set of alternative take-off and landing points;

[0189] In S34, the expression of customer demand constraint is:

[0190] ;

[0191] Where, m represents the total number of logistics demand points, h i Indicate customer needs;

[0192] In S34, the expression of the service range constraint is:

[0193] ;

[0194] Where, d ij Indicates alternative take-off and landing points i and demand points j The Euclidean distance of R L Indicates the service radius of the first-level take-off and landing point, R S Indicates the service radius of the secondary take-off and landing point, JRepresents a set of demand points.

[0195] In the embodiment of the present invention, in S35, the calculation formula for normalizing the objective function with the lowest cost as the optimization target is:

[0196] ;

[0197] Where, C represents the total daily economic cost, C n Indicates the reverse target, C (k) It represents the total cost when all the alternative take-off and landing points are established as level 2 take-off and landing points;

[0198] In S35, the calculation formula for normalizing the objective function with the highest satisfaction as the optimization target is:

[0199] ;

[0200] Where, S Express satisfaction, S n Indicates a positive goal, S (k) It indicates the total customer satisfaction when the time satisfaction of each express delivery reaches 1;

[0201] In S35, the calculation formula for weighted summation is:

[0202] ;

[0203] Where, N represents the comprehensive evaluation index, ω 1 represents the economic cost target weight, ω 2 represents the customer satisfaction target weight.

[0204] In an embodiment of the present invention, in S36, based on the take-off and landing point quantity constraint, capacity constraint, take-off and landing point type and layout constraint, customer demand constraint, and service scope constraint, the Gurobi optimizer is used to solve the normalized objective function to obtain the site selection and layout result.

[0205] The present invention will be described below with reference to specific embodiments.

[0206] This paper focuses on the "last mile" logistics distribution, so it selects an area about 8.5km away from the main urban area of ​​a city. 2The plane terrain area is taken as the research scope; to ensure the solving performance of the subsequent solver, the operating environment of the present invention is a 64-bit Windows 10 operating system, the processor is AMD Ryzen 5 5600G, the main frequency is 3.9GHz, and the running memory is 32GB.

[0207] To facilitate quantitative analysis, the embodiments of the present invention are based on the following assumptions: assuming that the distribution of customer logistics demand within a certain period of time is known; assuming that the environment of the study area has not changed during the study, and that there will be no factors affecting the location of take-off and landing points, such as temporary no-fly zones; assuming that the slope of the study area is flat, that is, the impact of terrain on the location of take-off and landing points is not considered; assuming that the service range and capacity of the take-off and landing points are known, and to ensure the rational use of resources, they are set as first- and second-level take-off and landing points; assuming that the frequency of the sound waves generated during the take-off and landing of the drone is constant, that is, the radius of the propagated noise is constant; assuming that the wind speed encountered during the take-off and landing of the drone is constant, that is, the horizontal radius of the drop is constant; assuming that the customer's acceptance of the impact of noise and safety remains unchanged during the study, that is, each customer always has a buffer distance between noise and safety.

[0208] The present invention clarifies the distribution mode, and the application scenario focuses on the logistics distribution scenario in the city center area. The distribution mode adopted in this scenario is that after the user places an order online, the drone delivers the goods from the distribution center to the community take-off and landing point, and then the rider picks up the goods and delivers them to the customer; the site selection actually studies the urban distribution scenario and clarifies the research site selection scope; the building outline of the area is obtained, and the centroid of each outline polygon is extracted as the initial position of the customer point; the different sizes and numbers of buildings in the real scene will affect the distribution of customer points, and the DBSCAN method is used to perform an initial screening of the user point location; the approximate proportion of drones selected for logistics distribution during the peak delivery stage of the day is counted, and the sample function in Python is used for random sampling to perform a secondary screening of the customer location distribution; after clarifying the customer point location distribution, the take-off and landing point layout site selection is carried out; the noise and safety impact generated by the drone take-off and landing process are analyzed, and the comprehensive impact radius of noise and safety impact is clarified. The buffer program in GIS is used to select the building outline layer and input the distance data to establish the take-off and landing point and customer buffer zone.

[0209] After S2, assuming that the UAV is flying during the daytime, we can get L I =60 dB , assuming that the sound intensity level emitted by the drone at 1m is , can be obtained r noise =12.6 m .

[0210] Will r Decompose intor x and r y In two directions, assuming the drone descends to the lowest buffer height h b =0.5 m , the height of the take-off and landing point h v =2.2 m , pedestrian height r p,y =1.7 m .set up r p,y = h b + h v - r p,y , and then use the Euclidean distance formula to get .

[0211] The calculation of the buffer radius requires reference to relevant information, including the size and climate indicators of a small and light UAV. The relevant information parameters are shown in Table 1.

[0212] Table 1

[0213]

[0214] The calculated horizontal drop radius of the drone is r d =12.4 m The weight coefficient in the final environmental buffer zone radius is calculated by combining the influence radius of the two. μ 1=0.4, μ 2=0.6, the final environment buffer radius is 12.46 m , and bring in a buffer zone for the building outline.

[0215] The actual road radius is obtained as the road network buffer, and finally public places, no-fly zones, etc. are added as other buffers as the overall restricted area.

[0216] In S3, a two-level takeoff and landing point location selection model is established. The objective function is first established. Considering both economic cost and customer satisfaction, the economic cost indicator is established, with the lowest cost as the optimization goal; the satisfaction indicator is established, with the highest satisfaction as the optimization goal.

[0217] In S3, constraints are established. The number of established takeoff and landing points cannot exceed the maximum number of alternative points. Capacity constraints: Each customer's demand cannot exceed the capacity of the serviced takeoff and landing point. Type and layout constraints: The facility type is determined only after the facility is selected. Customer demand constraints: Each customer's demand must be met. Service range constraints: Ensure that each customer's assigned facility is within its maximum service distance.

[0218] The value range of relevant variables is: 、 as well as . Finally, multi-objective weighted summation is performed.

[0219] To adapt to the scenarios of this invention and verify the effectiveness of the proposed alternative site selection method, we set the demand quantity for each customer's location to a random number between (0.5 and 2.5) kg, referencing the performance parameters of small and light drones. Based on the payload capacity of small and light drones, this ensured that each customer's demand quantity was within the drone's delivery capacity while also reflecting the differences in demand between different customers. The demand quantity data for each customer is shown in Table 2, and the site selection parameters are shown in Table 3.

[0220] Table 2

[0221]

[0222] Table 3

[0223]

[0224] The Gurobi optimizer is used to solve the model and convert the model into code. Considering the large number of alternative locations in the real environment, a calculation time limit is required. The calculation time is set to 2 hours. The result can be output when the difference between the feasible solution and the boundary value is less than 1%. Finally, the site selection layout result can be obtained.

[0225] Solve to get the total objective function N =0.73690, including 4 level 1 take-off and landing points and 3 level 2 take-off and landing points, customer needs are fully covered, the total average daily economic cost is 477.44 yuan, and the locations of the take-off and landing points are guaranteed to be reasonable and feasible.

[0226] like Figure 2 As shown in the figure, it represents the schematic diagram of the UAV-take-off and landing point distribution mode; Figure 3 As shown in the figure, it represents the degree of change of the satisfaction curve; Figure 4 As shown in, it represents the distribution map of urban buildings and road networks; Figure 5 As shown in , it represents the initial screening result diagram of customer point location; Figure 6 As shown in , it represents the customer point distribution map after secondary screening; Figure 7As shown in the figure, it is a schematic diagram of the noise impact range; Figure 8 As shown in the figure, it represents the safety impact range diagram; Figure 9 As shown in , it represents a mixed buffer map where the building buffer and the road network buffer are superimposed; Figure 10 As shown in the figure, it represents the grid area map generated for the study area; Figure 11 As shown in the figure, it represents the filter map for limiting the buffer zone of the grid area; Figure 12 As shown, it represents the final layout of the logistics drone take-off and landing points.

[0227] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific descriptions and embodiments. Those skilled in the art can make various other specific variations and combinations based on the technical teachings disclosed in the present invention without departing from the essence of the present invention, and such variations and combinations are still within the scope of protection of the present invention.

Claims

1. A method for selecting drone take-off and landing points for urban terminal logistics, characterized in that: The following steps are involved: S1. Obtain the building outline of the area to be selected and construct a customer buffer zone based on the building outline; S2. Based on the customer buffer zone, conduct noise and safety analysis on the drone and construct the overall restricted area; S3. Based on the overall restricted area, construct an objective function with the lowest cost as the optimization goal, an objective function with the highest satisfaction as the optimization goal, and several constraints to obtain the site selection and layout results; The S2 includes the following sub-steps: S21. Build a free-field model for the drone based on the client buffer zone; S22. determining the sound intensity of the semi-free field according to the noise source of the semi-free field; S23. Determine a sound intensity level attenuation model of the semi-free field based on the free field model and the sound intensity of the semi-free field; S24. Determine the spherical propagation radius of noise based on the sound intensity level attenuation model in a semi-free field; S25. Determine the noise level radius of the take-off and landing points based on the noise spherical propagation radius; S26. Determine the horizontal constant force provided by the wind to the UAV within the noise level radius of the take-off and landing point; S27, determining the vertical speed and flight altitude of the drone when the drone is falling, and generating the drone falling time; S28. Determine the horizontal velocity of the drone in the falling state based on the horizontal constant force provided by the wind to the drone and the falling time of the drone; S29. Determine the horizontal displacement radius of the drone based on the horizontal velocity of the drone in the falling state; S210: Determine a horizontal impact radius of the UAV based on the horizontal displacement radius of the UAV; S211. Determine the environmental buffer zone radius based on the UAV's horizontal impact radius; S212. Determine a building buffer zone based on the environmental buffer zone radius, using the actual road radius as the road network buffer zone and public places and no-fly zones as other buffer zones; S213. Build-up buffer zones, road network buffer zones, and other buffer zones are considered as overall restricted areas; The S3 includes the following sub-steps: S31, gridding the area to be selected, performing difference analysis between the gridding result and the overall restricted area to obtain a grid map; S32. Based on the grid map, construct an objective function with the minimum cost as the optimization goal; S33. Construct an objective function with the highest satisfaction as the optimization goal; S34. Construct constraints on the number of take-off and landing points, capacity constraints, take-off and landing point type and layout constraints, customer demand constraints, and service scope constraints; S35. Normalizing the objective function with the lowest cost as the optimization goal and the objective function with the highest satisfaction as the optimization goal, respectively, and performing a weighted summation of the normalized objective function with the lowest cost as the optimization goal and the objective function with the highest satisfaction as the optimization goal to obtain a normalized objective function; S36. Obtaining a site selection and layout result based on the number of take-off and landing points constraints, the capacity constraints, the type and layout constraints of the take-off and landing points, the customer demand constraints, the service scope constraints, and the normalized objective function; In S32, the expression of the objective function with the minimum cost as the optimization goal is: ； Where, C represents the total daily economic cost, C f,n represents the average daily fixed cost of the take-off and landing points taking into account depreciation, C v represents the variable cost of the take-off and landing points, y ij It represents the demand quantity allocated to customers at the take-off and landing points, m represents the total number of logistics demand points, n Indicates the total number of alternative take-off and landing points, C L represents the construction cost of the first-level take-off and landing point, C S represents the construction cost of the secondary landing and take-off point, f i Indicates the type of facility, x i Indicates alternative take-off and landing points. c d represents the unit transportation cost, d ij Indicates alternative take-off and landing points i and demand points j The Euclidean distance of i Indicates the sequence number of the alternative take-off and landing point. j Indicates the sequence number of the logistics demand point, t d Indicates the facility depreciation time; In S33, the expression of the objective function with the highest satisfaction as the optimization goal is: ； Where, S Indicates overall customer satisfaction, S ( t ij ) represents the concave-convex time satisfaction function, t ij Indicates that the drone is taking off from an alternative take-off and landing point i To the demand point j The delivery time, D j Indicates demand points j The logistics demand, v d Indicates the delivery speed, t L represents the absolute satisfaction threshold, t B represents the absolute dissatisfaction threshold; In S34, the expression of the take-off and landing point quantity constraint condition is: ； Where, x i Indicates alternative take-off and landing points. n Indicates the total number of alternative take-off and landing points; In S34, the capacity constraint condition is expressed as: ； Where, y ij It represents the demand quantity allocated to customers at the take-off and landing points, f i Indicates the type of facility, V L,i Indicates the capacity of the first-level take-off and landing point, V S,i Indicates the capacity of the secondary take-off and landing point; In S34, the expression of take-off and landing point type and layout constraint is: ； Where, I Indicates a set of alternative take-off and landing points; In S34, the expression of the customer demand constraint condition is: ； Where, m represents the total number of logistics demand points, h i Indicate customer needs; In S34, the expression of the service range constraint condition is: ； Where, d ij Indicates alternative take-off and landing points i and demand points j The Euclidean distance of R L Indicates the service radius of the first-level take-off and landing point, R S Indicates the service radius of the secondary take-off and landing point, J represents a set of demand points; In S35, the calculation formula for normalizing the objective function with the lowest cost as the optimization target is: ； Where, C represents the total daily economic cost, C n Indicates the reverse target, C (k) It represents the total cost when all the alternative take-off and landing points are established as level 2 take-off and landing points; In S35, the calculation formula for normalizing the objective function with the highest satisfaction as the optimization target is: ； Where, S Express satisfaction, S n Indicates a positive goal, S (k) It indicates the total customer satisfaction when the time satisfaction of each express delivery reaches 1; In S35, the calculation formula for weighted summation is: ； Where, N represents the comprehensive evaluation index, ω 1 represents the economic cost target weight, ω 2 represents the customer satisfaction target weight; In S36, based on the constraints on the number of take-off and landing points, the capacity constraints, the type and layout constraints of the take-off and landing points, the customer demand constraints, and the service scope constraints, the Gurobi optimizer is used to solve the normalized objective function to obtain the site selection and layout results.

2. The method for selecting a take-off and landing point for UAVs for urban terminal logistics according to claim 1 is characterized in that: The S1 includes the following sub-steps: S11, obtaining the building outline of the area to be selected, and taking the centroid of the building outline as the initial position of the customer point; S12, using the DBSCAN method to perform a preliminary screening of the initial positions of the customer points; S13. Based on the ratio of the number of drone deliveries to the total number of deliveries in the selected location area, random sampling is performed to perform a secondary screening of the initial customer point locations to obtain the final customer point location distribution; S14. Construct a customer buffer zone based on the final customer point location distribution and the noise during the drone take-off and landing process.

3. The method for selecting a take-off and landing point for UAVs for urban terminal logistics according to claim 1 is characterized in that: In S21, the expression of the free field model is: ； Where, Δ L Represents the sound intensity level difference between the sound source and the target position, r 1 represents the initial position, r 2 represents the target position, log 10 (·) represents the logarithmic function with base 10; In the S22, the sound intensity of the semi-free field I The calculation formula is: ； Where, W represents the sound source power, r Indicates the distance the sound source travels to the sphere; In S23, the expression of the sound intensity level attenuation model in the semi-free field is: ； Where, L I is the sound intensity level in a semi-free field, W 0 represents the reference sound power, r noise represents the spherical propagation radius of the noise, L W represents the sound power level of the drone, lg(·) represents the logarithmic function with base e; In S25, the noise level radius of the take-off and landing points r n,x The calculation formula is: ； Where, r n,y Indicates the vertical noise radius of the take-off and landing point.

4. The method for selecting a take-off and landing point for UAVs for urban terminal logistics according to claim 1 is characterized in that: In S26, the wind force is the horizontal constant force provided by the drone. F wind The calculation formula is: ； Where, ρ represents the air density, v w Indicates wind speed, C d represents the drag coefficient, A u represents the drone contact area; In the above S27, the vertical speed of the drone in the falling state v y The calculation formula is: ； Where, M Indicates the UAV mission payload weight, g represents the acceleration due to gravity, A u,y Represents the contact area in the vertical direction; In the above S28, the horizontal speed of the drone in the falling state v x The calculation formula is: ； Where, A u,x represents the contact area in the horizontal direction, t represents the drone drop time, tanh(·) represents the hyperbolic tangent function; In S29, the horizontal displacement radius of the UAV r u , x The calculation formula is: ； In S210, the horizontal impact radius of the UAV r d The calculation formula is: ； Where, r p represents the average pedestrian radius; In S211, the radius of the environmental buffer zone r b The calculation formula is: ； Where, μ 1 represents the noise weight, μ 2 represents the safety weight, r n , x Indicates the noise level radius of the take-off and landing point.

Images