Path planning method for low-altitude delivery of express delivery in industrial parks

Abstract

The present application provides a method for low-altitude delivery path planning for park express delivery, including the following: obtaining drone parameters and delivery node information; calculating the relative energy consumption weights between each node; establishing a weight matrix, a transition matrix, and a reduction matrix to obtain a reduction number; calculating the weight lower bound, finding the target path and determining the energy-optimal path; checking whether the driving time of the energy-optimal path exceeds the specified time; updating the weight lower bound; determining the energy-optimal path within the specified time. The method for low-altitude delivery path planning for park express delivery provided in the present application provides a new low-altitude delivery method for the transportation of park express delivery. Compared with the heuristic algorithm that seeks an approximate optimal solution in a large number of nodes for efficiency, the technical solution of the present invention takes into account the characteristics of limited park nodes and clear structure, and can reduce the energy consumption of drone delivery to the greatest extent while keeping the calculation time similar, and is more suitable for the low-altitude delivery system of park express delivery.

Description

Technical Field

[0001] The present application belongs to the technical field of drone path planning, and in particular, relates to a method for low-altitude delivery path planning for park express delivery. Background Art

[0002] With the advancement of drone technology, the application of drones in the field of logistics has gradually matured. Compared with traditional manual delivery methods, drones can fly in a straight line at low altitudes, which not only shortens the distance of the transportation route, but also reduces the impact of traffic congestion on delivery efficiency. High-quality drone low-altitude delivery routes are an important guarantee for the rapid delivery of goods and the reduction of transportation energy consumption. Therefore, planning the optimal route has important practical significance for low-altitude delivery.

[0003] The existing low-altitude delivery path planning methods take into account multiple factors such as energy consumption, distance, unloading volume, and battery replacement between the warehouse and each demand node. They mostly use heuristic algorithms such as the two-stage algorithm for path optimization to find a good delivery path within an acceptable time to minimize the energy consumption of the drone. The two-stage algorithm consists of two stages: Stage 1 uses the WOA design algorithm to solve the lower-level path from the distribution station to the express cabinet; Stage 2 solves the upper-level path from the warehouse to the distribution station. The two-stage algorithm consists of 8 steps in total.

[0004] Step 1: Initialize the whale population, use real number coding, and randomly generate a sequence that traverses all customer access sequences based on load constraints and time constraints, thereby reducing the difficulty of searching for solutions.

[0005] Step 2: Decode and calculate the objective function value of the current initial solution to obtain the current optimal value.

[0006] Step 3: Determine whether the current number of iterations meets the maximum termination condition. If so, output the objective function value of the optimal individual; if not, proceed to step 4.

[0007] Step 4: Through three hunting behaviors, update the positions of all whale individuals and calculate the updated population objective function value; compare with the current optimal value, if the updated result is better, update the current optimal value and all corresponding parameters, otherwise do not update.

[0008] Step 5: Repeat steps 1 to 4 until the number of iterations meets the maximum number of iterations and output the final result.

[0009] Step 6: Input the lower-level delivery plan with parameters such as small drone capacity, number of flights, and range.

[0010] Step 7: Calculate the total delivery demand and time for each distribution station in each time period by analyzing the input quantity, time, arrival and departure time of light drones in each time period, and determine the departure time and number of small drones to be used.

[0011] Step 8: Based on the payload capacity of small drones and the delivery demand and time of distribution stations, small drones are scheduled to generate flight paths in each time period, and the arrival time of each small drone at each upper-level node is output.

[0012] By modifying or adjusting steps 7 and 8, the solution results of different delivery service modes can be output respectively: in the on-demand mode, the small drone arranges the specific flight path of the drone according to the distribution station time window mapped by the subordinate demand point; in the shift service mode, the subordinate demand time is segmented, and multiple drone shifts are calculated according to the demand and time period of the corresponding customer group, and the shift timetable and flight path space-time diagram under the corresponding mode are generated.

[0013] The above-mentioned low-altitude delivery path planning method is mainly aimed at complex scenarios with numerous delivery nodes and unfixed starting and ending points. Due to the large number of parameters and huge amount of calculation, in order to pursue calculation efficiency, heuristic algorithms are often used, relying on evaluation functions or heuristic functions to determine the possible optimal expansion node and expand the path with this node. Due to the small number of nodes and clear structure in the low-altitude delivery scene of the park, the above-mentioned low-altitude delivery path planning method may not be effectively applied to the park. The specific shortcomings are as follows:

[0014] 1. Whether mathematical algorithms or heuristic algorithms are used to plan the low-altitude delivery paths in the park, the solution can be obtained in a shorter calculation time. Compared with mathematical algorithms that can obtain the optimal solution, heuristic algorithms only evaluate the relationship between adjacent nodes and do not consider the impact of node selection on the global situation, so they can only calculate an approximate optimal solution. If a heuristic algorithm is used to plan the low-altitude delivery path for a park with fewer nodes, the path solution time may be slightly shortened, but the energy consumption of transportation cannot be minimized, which is not worth the loss.

[0015] 2. In the current mathematical algorithm for path planning, the weight between two nodes is fixed (i.e., no matter which nodes the drone flies through before, the energy consumption of flying from node A to node B is constant). But in fact, the energy consumption of drones changes dynamically. Every time the drone unloads at a node, the weight of the cargo changes, and the energy consumption of flying to subsequent nodes also changes. The existing mathematical algorithm for path planning cannot calculate the problem of dynamic changes in weights between nodes. Summary of the invention

[0016] In order to solve the problem that the current low-altitude delivery path planning method is applied to the park, there is a mismatch between the algorithm and the actual scenario, such as the heuristic algorithm cannot find the optimal path, the mathematical algorithm cannot consider the dynamic change of node weights, etc. This application provides a method for low-altitude delivery path planning for express delivery in the park, which aims to efficiently find the optimal path for low-altitude delivery of express delivery in the park, and obtain the optimal solution by determining the relative energy consumption weights between nodes and using the branch and bound method. The method includes the following steps:

[0017] Step 1: Obtain the parameters of the drone and the operation node information, mark the serial numbers of each node in the park, and calculate the relative energy consumption weights between each node;

[0018] Step 2: Use the park node number as the matrix row and column node number, and use the matrix column node as the path starting node and the matrix row node as the path ending node to calculate the relative energy consumption cost between each matrix node and establish a weight matrix between each node;

[0019] Step 3: Subtract the minimum value of the row corresponding to each element of the weight matrix to obtain a transition matrix; subtract the minimum value of the column corresponding to each element of the transition matrix to obtain a reduced matrix; add the minimum value of each row of the weight matrix to the minimum value of each column of the transition matrix to obtain the reduced number of the weight matrix;

[0020] Step 4: Taking the starting point as the 0th node, calculate the weight lower bounds of each path and the weight lower bounds of the non-path in turn, and take the path with the lowest weight lower bound as the target path. Repeat the process until all nodes are connected to form a closed path, and take the closed path as the path with the optimal energy consumption.

[0021] Step 5: Calculate the energy cost of the path between every two adjacent nodes in the energy-optimal path, and obtain the battery replacement decision value of the UAV at each node; add the total flight time of the UAV to the total battery replacement time to obtain the driving time of the UAV in the energy-optimal path;

[0022] Step 6: If the UAV's driving time on the energy-optimal path does not exceed the specified time, the path is the optimal path; if it exceeds the specified time, update the lower bound of the weight of the target path, repeat steps 4 to 6, and keep looping until a path that does not exceed the specified time is found. This path is the energy-optimal path within the specified time.

[0023] In a possible implementation, step 1 includes: recording the total number of nodes in the park as u, recording the sequence numbers of the nodes as nodes A, B, C, etc., and calculating the relative energy consumption weights between the nodes. The calculation formula is as follows:

[0024] ;

[0025] In the formula, is the relative energy consumption weight from node a to node b; is the weight of the drone; is the operational weight of node b; is the horizontal distance between the paths of node a and node b; It is the energy consumption per unit distance per unit weight when the UAV is flying horizontally; is the vertical distance the UAV rises at node a; It is the energy consumption per unit distance per unit weight when the UAV rises vertically; is the vertical distance the UAV descends at node b; It is the energy consumption per unit distance per unit weight when the UAV descends vertically;

[0026] Step 2: Establish the weight matrix between each node , the calculation formula is as follows:

[0027] ;

[0028] In the formula, i is the number of rows, i=1, 2, 3...u; j is the number of columns, j=1, 2, 3...u; is the weight matrix The element value of the i-th row and j-th column; is the relative energy consumption cost between the node corresponding to the i-th row and the node corresponding to the j-th column.

[0029] In a possible implementation, step 3 includes: establishing a weight matrix The transition matrix , reduction matrix And calculate the weight matrix The reduction number g is calculated as follows:

[0030] ;

[0031] ;

[0032] ;

[0033] In the formula, is the transition matrix The element value of the i-th row and j-th column; is the weight matrix The smallest element value in row i; is the reduction matrix The element value of the i-th row and j-th column; is the transition matrix The smallest element value in the jth column; g is the weight matrix The reduction number.

[0034] In a possible implementation, step 4 includes:

[0035] The calculation formula for the lower bound of the path weight and the target path determination method are as follows:

[0036] ;

[0037] ;

[0038] ;

[0039] ;

[0040] In the formula, is the starting point of the t-th path, t=0, 1, 2…u; is the end point of the tth path; When a path other than the target path is selected as the t-th path, the path matrix The element value of the i-th row and j-th column in ; ; For Node , In the weight matrix The corresponding number of rows in ; For Node , In the weight matrix The corresponding number of columns in ; is the path matrix The smallest element value in the i-th row; is the path matrix The smallest element value in the jth column; is the weight lower bound of path (t); is the weight lower bound of the target path (t); When a path other than the target path is not selected as the t-th path, the non-path matrix The element value of the i-th row and j-th column in ; is a non-path matrix The minimum value of the elements in the i-th row; is a non-path matrix The minimum value of the elements in the jth column; is the lower bound of the weight of non-path (t);

[0041] like , it means that the weight lower bound of the path has reached the minimum, that is, the path is the target path, and the next cycle can be carried out;

[0042] Step 5: Check whether the driving time of the energy-optimal path exceeds the specified time. The calculation formula for the driving time of the energy-optimal path is as follows:

[0043] ;

[0044] ;

[0045] ;

[0046] ;

[0047] In the formula, q(t) is the tth node of the energy-optimal path; is the energy cost from node q(t) to node q(t+n) in the energy-optimal path, n is a non-negative integer, t+n≤u; is the total weight of the cargo transported from node q(t) to node q(t+1); is the battery replacement decision value of node q(t), 1 means battery replacement, 0 means no battery replacement, If it is unreachable, the default values ​​are q(0)=1 and q(u)=0; m is a non-negative integer, and q(tm) is the closest battery swap node to q(t); The battery capacity of the drone; is the number of battery swaps required for the UAV in the energy-optimal path; is the total driving time of the UAV on the energy-optimal path; is the horizontal flight speed of the UAV; is the vertical ascent speed of the drone; is the vertical descent speed of the drone; The time it takes to replace the battery for the drone.

[0048] In a possible implementation, step 6 includes: if the driving time of the UAV on the energy-optimal path does not exceed the specified time, the path is the optimal path; if it exceeds the specified time, updating the weight lower bound of the target path, repeating steps 4-6, and looping continuously until a path that does not exceed the specified time is found, and the path is the energy-optimal path within the specified time;

[0049] The weight lower bound update formula of the target path is as follows:

[0050] ;

[0051] In the formula, is the lower bound of the target path weight updated before the mth cycle, .

[0052] The above method can consider the weights between nodes and expand all branches of nodes through the branch and bound method, and bound the whole in the branching process to finally obtain the optimal solution. Because the branch and bound method can expand all branches of nodes so that the weights between all nodes can be interconnected, it is not necessary to consider the absolute energy consumption between nodes, but to determine a calculation method for the relative energy consumption weights between nodes, and convert the dynamically changing weights between nodes into fixed weights. Therefore, calculating the relative energy consumption weights between nodes and using the branch and bound method to find the optimal path can effectively overcome the limitations of the current method.

[0053] Beneficial effects: The low-altitude delivery path planning method for park express provided by this application provides a new low-altitude delivery method for the transportation of park express. Compared with the heuristic algorithm that seeks an approximate optimal solution in a large number of nodes for efficiency, the technical solution of the present invention takes into account the characteristics of limited park nodes and clear structure, and can reduce the energy consumption of drone delivery to the greatest extent while keeping the calculation time similar, and is more suitable for the low-altitude delivery system of park express. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0055] Figure 1 It is a flow chart of the low-altitude delivery path planning method for the park express delivery;

[0056] Figure 2 is an example graph of a weight matrix;

[0057] Figure 3 is an example graph of nodes corresponding to the target path;

[0058] Figure 4 is the weight matrix diagram in the first embodiment;

[0059] Figure 5 is a diagram of the transition matrix and the reduction matrix in the first embodiment;

[0060] FIG. 6 (a) to FIG. 6 (f) are diagrams of the path matrix and the non-path matrix in the first embodiment;

[0061] FIG. 7 ( a ) to FIG. 7 ( f ) are diagrams of the path matrix and the non-path matrix during the first cycle in the first embodiment. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be described in more detail below in conjunction with the drawings in the preferred embodiments of the present application. In the drawings, the same or similar reference numerals throughout represent the same or similar parts or parts with the same or similar functions. The described embodiments are part of the embodiments of the present application, not all of the embodiments. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limitations on the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0063] The current low-altitude delivery path planning method applied to the park has the problem of mismatch between the algorithm and the actual scene, such as the heuristic algorithm cannot find the optimal path, and the mathematical algorithm cannot consider the dynamic change of node weight. In order to overcome the limitations of the current low-altitude delivery path planning method in the park, a mathematical algorithm that can consider the dynamic change of the weight of the drone and the weight between the nodes is needed. The branch and bound method can consider the weight between the nodes and expand all the branches of the nodes. In the process of branching, the overall limit is performed to finally obtain the optimal solution. Because the branch and bound method can expand all the branches of the nodes so that the weights between all the nodes can be interconnected, it is not necessary to consider the absolute energy consumption between the nodes, but to determine a method for calculating the relative energy consumption weight between the nodes, and convert the dynamically changing weights between the nodes into fixed weights. Therefore, calculating the relative energy consumption weights between the nodes and using the branch and bound method to find the optimal path can effectively overcome the limitations of the current method.

[0064] Based on this, this application provides a method for low-altitude express delivery path planning in a park. The method is to determine the relative energy consumption weights between nodes and find the optimal path with the least energy consumption within a specified time through the branch and bound method. Figure 1 As shown, the method comprises the following steps:

[0065] Step 1: Obtain the parameters of the drone and the operation node information, record the total number of nodes in the park as u, and record the sequence of each node as node A, B, C, etc., and calculate the relative energy consumption weight between each node. The calculation formula is as follows:

[0066] ;

[0067] In the formula, is the relative energy consumption weight from node a to node b; is the weight of the drone; is the operational weight of node b; is the horizontal distance between the paths of node a and node b; It is the energy consumption per unit distance per unit weight when the UAV is flying horizontally; is the vertical distance the UAV rises at node a; It is the energy consumption per unit distance per unit weight when the UAV rises vertically; is the vertical distance the UAV descends at node b; It is the energy consumption per unit distance per unit weight when the UAV descends vertically;

[0068] Step 2: Establish the weight matrix between each node (See Figure 2 ), the calculation formula is as follows:

[0069] ;

[0070] In the formula, i is the number of rows, i=1, 2, 3...u; j is the number of columns, j=1, 2, 3...u; is the weight matrix The element value of the i-th row and j-th column; is the relative energy consumption cost between the node corresponding to the i-th row and the node corresponding to the j-th column;

[0071] Step 3: Create a weight matrix The transition matrix , reduction matrix And calculate the weight matrix The reduction number g is calculated as follows:

[0072] ;

[0073] ;

[0074] ;

[0075] In the formula, is the transition matrix The element value of the i-th row and j-th column; is the weight matrix The smallest element value in row i; is the reduction matrix The element value of the i-th row and j-th column; is the transition matrix The smallest element value in the jth column; g is the weight matrix The reduction number of

[0076] Step 4: Find the optimal energy consumption path, including the following:

[0077] Taking the starting point as the 0th node, find the path with the lowest weight lower bound as the target path (see Figure 3 ), and the cycle continues until all nodes are connected to form a closed path, which is the energy-optimal path;

[0078] The calculation formula for the lower bound of the path weight and the target path determination method are as follows:

[0079] ;

[0080] ;

[0081] ;

[0082] ;

[0083] In the formula, is the starting point of the t-th path, t=0, 1, 2…u; is the end point of the tth path; When a path other than the target path is selected as the t-th path, the path matrix The element value of the i-th row and j-th column in ; ; For Node , In the weight matrix The corresponding number of rows in ; For Node , In the weight matrix The corresponding number of columns in ; is the path matrix The smallest element value in the i-th row; is the path matrix The smallest element value in the jth column; is the weight lower bound of path (t); is the weight lower bound of the target path (t); When a path other than the target path is not selected as the t-th path, the non-path matrix The element value of the i-th row and j-th column in ; is a non-path matrix The minimum value of the elements in the i-th row; is a non-path matrix The minimum value of the elements in the jth column; is the lower bound of the weight of non-path (t);

[0084] like , it means that the weight lower bound of the path has reached the minimum, that is, the path is the target path, and the next cycle can be carried out;

[0085] Step 5: Check whether the driving time of the energy-optimal path exceeds the specified time. The calculation formula for the driving time of the energy-optimal path is as follows:

[0086] ;

[0087] ;

[0088] ;

[0089] ;

[0090] In the formula, q(t) is the tth node of the energy-optimal path; is the energy cost from node q(t) to node q(t+n) in the energy-optimal path, n is a non-negative integer, t+n≤u; is the total weight of the cargo transported from node q(t) to node q(t+1); is the battery replacement decision value of node q(t), 1 means battery replacement, 0 means no battery replacement, If it is unreachable, the default values ​​are q(0)=1 and q(u)=0; m is a non-negative integer, and q(tm) is the closest battery swap node to q(t); is the battery capacity of the drone. For example, if the drone is equipped with three batteries, Single battery capacity × 3; is the number of battery swaps required for the UAV in the energy-optimal path; is the total driving time of the UAV on the energy-optimal path; is the horizontal flight speed of the UAV; is the vertical ascent speed of the drone; is the vertical descent speed of the drone; The time it takes to replace the battery for the drone.

[0091] Step 6: If the UAV's driving time on the energy-optimal path does not exceed the specified time, the path is the optimal path; if it exceeds the specified time, update the weight lower bound of the target path, repeat steps 4-6, and keep looping until a path that does not exceed the specified time is found. This path is the energy-optimal path within the specified time;

[0092] The weight lower bound update formula of the target path is as follows:

[0093] ;

[0094] In the formula, is the lower bound of the target path weight updated before the mth cycle, . Embodiment 1

[0095] Meituan's fourth-generation drone was used to carry out a four-node low-altitude delivery (1 delivery center and 3 delivery nodes). The drone started from node A (delivery center), passed through nodes B, C, and D (delivery nodes), and finally returned to node A. The drone weighs 7kg, with a maximum cargo weight of 2.5kg. The drone is equipped with 2 12s ternary lithium batteries, each with a capacity of 1250mah. The drone takes 3 minutes to replace the battery. The energy consumption per unit weight per unit distance of the drone's ascent, horizontal, and descent is , , Both , the UAV's ascending, horizontal, and descending flight speeds , , The operation weight of nodes A, B, C and D are 5m / s, 15m / s and 3m / s respectively. , , , The rising and falling distances of each node are 0kg, 2kg, 0.2kg, and 0.3kg respectively. , The horizontal distance between nodes is 50m. , , , , , , requiring that the commissioning time should not exceed 60 minutes.

[0096] According to the formula:

[0097] ;

[0098] ;

[0099] Calculate the relative energy consumption weights between nodes and establish a weight matrix (See Figure 4 ).

[0100] According to the formula:

[0101] ;

[0102] ;

[0103] ;

[0104] See also Figure 5 , establish the weight matrix The transition matrix , reduction matrix And calculate the weight matrix The reduction number g=5307.9.

[0105] According to the formula:

[0106] ;

[0107] ;

[0108] ;

[0109] ;

[0110] As shown in Figure 6 (a) to Figure 6 (f), the path matrix and the non-path matrix are established, A is taken as the 0th node, and the weight lower bound of path (1) is calculated to obtain , , , ,because , so we choose AB as the target path (1); calculate the weight lower bound of path (2) and get , ,because , so BC is selected as the target path (2), and the optimal energy consumption path is ABCDA.

[0111] According to the formula:

[0112] ;

[0113] ;

[0114] ;

[0115] ;

[0116] Calculate the travel time of path ABCDA and get , , , , Because in the calculation Obtained in the process ,so , , the travel time of this route exceeds the specified time, and another route must be found.

[0117] According to the formula:

[0118] ;

[0119] 7 (a) to 7 (f), the first cycle of searching for the energy-consumption-approximate optimal path is performed, and the lower bound of the target path weight is updated to obtain , , , .

[0120] Establish the path matrix and the non-path matrix, take A as the 0th node, calculate the weight lower bound of path (1), and obtain , , , ,because , so we select AD as the target path (1); calculate the weight lower bound of path (2) and get , ,because , so BA is selected as the target path (2), and the optimal energy consumption path is ADCBA.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for low-altitude delivery path planning for express delivery in a park, characterized in that: The steps include: Step 1: Obtain the parameters of the drone and the operation node information, mark the serial numbers of each node in the park, and calculate the relative energy consumption weights between each node; Step 2: Use the park node number as the matrix row and column node number, and use the matrix column node as the path starting node and the matrix row node as the path ending node to calculate the relative energy consumption cost between each matrix node and establish a weight matrix between each node; Step 3: Subtract the minimum value of the row corresponding to each element of the weight matrix to obtain a transition matrix; Subtract the minimum value of the column corresponding to each element of the transition matrix to obtain the reduced matrix; Add the minimum value of each row of the weight matrix and the minimum value of each column of the transition matrix to obtain the reduction number of the weight matrix; Step 4: Taking the starting point as the 0th node, calculate the weight lower bounds of each path and the weight lower bounds of the non-path in turn, and take the path with the lowest weight lower bound as the target path. Repeat the process until all nodes are connected to form a closed path, and take the closed path as the path with the optimal energy consumption. Step 5: Calculate the energy cost of the path between every two adjacent nodes in the energy-optimal path, and obtain the battery replacement decision value of the UAV at each node; add the total flight time of the UAV to the total battery replacement time to obtain the driving time of the UAV in the energy-optimal path; Step 6: If the UAV's driving time on the energy-optimal path does not exceed the specified time, the path is the optimal path; if it exceeds the specified time, update the weight lower bound of the target path, repeat steps 4 to 6, and keep looping until a path that does not exceed the specified time is found. This path is the energy-optimal path within the specified time; The step 1 includes: recording the total number of nodes in the park as u, recording the sequence numbers of the nodes as nodes A, B, C, etc., and calculating the relative energy consumption weights between the nodes. The calculation formula is as follows: ； In the formula, is the relative energy consumption weight from node a to node b; is the weight of the drone; is the operational weight of node b; is the horizontal distance between the paths of node a and node b; It is the energy consumption per unit distance per unit weight when the UAV is flying horizontally; is the vertical distance the UAV rises at node a; It is the energy consumption per unit distance per unit weight when the UAV rises vertically; is the vertical distance the UAV descends at node b; It is the energy consumption per unit distance per unit weight when the UAV descends vertically.

2. The method for low-altitude delivery path planning for park express delivery according to claim 1 is characterized in that: The step 2 includes: establishing a weight matrix between each node , the calculation formula is as follows: ； In the formula, i is the number of rows, i=1, 2, 3...u; j is the number of columns, j=1, 2, 3...u; is the weight matrix The element value of the i-th row and j-th column; is the relative energy consumption cost between the node corresponding to the i-th row and the node corresponding to the j-th column.

3. The method for low-altitude delivery path planning for park express delivery according to claim 2 is characterized in that: The step 3 includes: establishing a weight matrix The transition matrix , reduction matrix And calculate the weight matrix The reduction number g is calculated as follows: ； ； ； In the formula, is the transition matrix The element value of the i-th row and j-th column; is the weight matrix The smallest element value in row i; is the reduction matrix The element value of the i-th row and j-th column; is the transition matrix The smallest element value in the jth column; g is the weight matrix The reduction number.

4. The method for low-altitude delivery path planning for park express delivery according to claim 3 is characterized in that: The step 4 comprises: The calculation formula for the lower bound of the path weight and the target path determination method are as follows: ； ； ； ； In the formula, is the starting point of the t-th path, t=0, 1, 2…u; is the end point of the tth path; When a path other than the target path is selected as the t-th path, the path matrix The element value of the i-th row and j-th column in ; ; , for nodes In the weight matrix The corresponding number of rows in ; For Node In the weight matrix The corresponding number of columns in ; is the path matrix The smallest element value in the i-th row; is the path matrix The smallest element value in the jth column; is the weight lower bound of path (t); is the weight lower bound of the target path (t); When a path other than the target path is not selected as the t-th path, the non-path matrix The element value of the i-th row and j-th column in ; is a non-path matrix The minimum value of the elements in the i-th row; is a non-path matrix The minimum value of the elements in the jth column; is the lower bound of the weight of non-path (t); like , it means that the weight lower bound of the path has reached the minimum, that is, the path is the target path, and the next cycle can be carried out.

5. The method for low-altitude delivery path planning for park express delivery according to claim 4 is characterized in that: The step 5 includes: the calculation formula of the energy consumption optimal path travel time is as follows: ； ； ； ； In the formula, q(t) is the tth node of the energy-optimal path; is the energy cost from node q(t) to node q(t+n) in the energy-optimal path, n is a non-negative integer, t+n≤u; is the total weight of the cargo transported from node q(t) to node q(t+1); is the battery replacement decision value of node q(t), 1 means battery replacement, 0 means no battery replacement, If it is unreachable, the default values ​​are q(0)=1 and q(u)=0; m is a non-negative integer, and q(tm) is the closest battery swap node to q(t); The battery capacity of the drone; is the number of battery replacements required by the UAV in the energy-optimal path; T is the total driving time of the UAV in the energy-optimal path; is the horizontal flight speed of the UAV; is the vertical ascent speed of the drone; is the vertical descent speed of the drone; The time it takes to replace the battery for the drone.

6. The method for low-altitude delivery path planning for park express delivery according to claim 5 is characterized in that: The step 6 includes: if the driving time of the UAV on the energy-optimal path does not exceed the specified time, the path is the optimal path; if it exceeds the specified time, the weight lower bound of the target path is updated, and steps 4-6 are repeated, and the cycle continues until a path that does not exceed the specified time is found, and the path is the energy-optimal path within the specified time; The weight lower bound update formula of the target path is as follows: ； In the formula, is the lower bound of the target path weight updated before the mth cycle, .

Images