Multi-uav path efficient reconstruction method for complex terrain and emergent task

Abstract

To solve the technical problems that the existing path re-planning algorithm cannot realize task re-allocation and path re-planning for dynamic burst tasks, and the existing task re-allocation algorithm is difficult to realize dynamic re-allocation of tasks in real time when facing dynamic burst tasks, and the processing capacity is insufficient when facing complex terrain, the present application provides a kind of multi-unmanned aerial vehicle path efficient reconstruction method for complex terrain and burst task, considering a large number of task demand constraints, fuel constraints, complex elevation terrain and burst task factors, for the existing unmanned aerial vehicle task sequence, a kind of low altitude real-time local path re-planning for reasonable evaluation and sorting of burst task value, optimal unmanned aerial vehicle selection, considering threat degree and digital terrain representation is designed to realize efficient path reconstruction;When inserting the burst task, the optimal insertion point strategy adopted by the present application reduces the repeated calculation process of the insertion task point cost, effectively reduces the time complexity of the calculation, and guarantees the real-time index of the algorithm.

Description

TECHNICAL FIELD

[0001] The present application relates to a multi-UAV path reconstruction method. BACKGROUND

[0002] Unmanned helicopters have important application value in performing tasks such as reconnaissance and surveillance, target tracking, attack on the ground, transportation of materials, evacuation of the wounded, and the like, due to their unique vertical take-off and landing, hovering, and flexible maneuvering characteristics. When applied in low-altitude environments, multi-unmanned helicopters need to plan a safe path for each unmanned helicopter from a starting point to a target point under conditions such as low-altitude complex digital elevation terrain (DEM) environment and netted radar detection, and the path needs to meet the fuel constraint of the unmanned helicopter. However, in the process of performing a series of tasks, the unmanned helicopter may encounter unexpected task requirements. How to efficiently complete real-time path reconstruction according to the path planned for the existing tasks will greatly improve the flexibility and survivability of multi-unmanned helicopters in handling tasks.

[0003] Unlike traditional path re-planning algorithms for coping with dynamic changes in the environment, path reconstruction algorithms emphasize the task redistribution and path re-planning process for dynamic unexpected tasks. The main problems that need to be solved by the path reconstruction algorithm include: (1) under the constraints of fuel and task requirements, coordinating the task allocation of multiple unmanned helicopters to ensure that the fuel of the unmanned helicopter can support it to meet the flight task, and multiple unmanned helicopters can almost simultaneously complete all tasks; (2) identifying the optimal insertion position of the unexpected task in the existing task sequence of the selected unmanned helicopter; (3) considering finding a locally optimal re-planned path on the digital elevation map.

[0004] Therefore, when multiple unmanned helicopters perform path reconstruction for unexpected tasks, not only do they need to consider the problem of finding a locally safe path for a single unmanned helicopter under complex terrain and fire threat conditions, but they also need to solve the problem of task allocation for multiple unmanned helicopters under fuel constraints when facing unexpected tasks, and consider the time coordination problem between multiple unmanned helicopters to ensure that multiple unmanned helicopters can almost simultaneously perform all tasks in the shortest possible time.

[0005] At present, the distributed real-time task reassignment algorithm for multiple unmanned helicopters in the event of tasks includes market auction algorithm, consensus-based binding algorithm (CBBA), heuristic performance influence algorithm (PI), Markov decision process (MDP), etc. Among them, the market auction algorithm is difficult to reasonably estimate the bidding cost when bidding for large-scale tasks, and is insufficient to consider the task demand constraints of dynamic environment and the fuel constraints of unmanned helicopters, and ignores the influence of complex terrain and sudden tasks on path reconstruction, resulting in difficulty in obtaining real-time high-quality bidding scheme. The CBBA algorithm and the PI algorithm both need to repeatedly calculate the cost of inserting a new task into the original task sequence, and more calculation is required in task conflict resolution calculation, resulting in low calculation efficiency and poor real-time performance, which is not suitable for dynamic reassignment process. The MDP algorithm is difficult to handle high-dimensional space and complex environment in the case of a large number of unmanned helicopters and task numbers, and usually needs to use deep reinforcement learning (DRL) method for repeated training, which is difficult to guarantee the training quality, and has insufficient processing capacity for sudden threats and tasks. SUMMARY

[0006] In order to solve the technical problems that the existing path re-planning algorithm cannot realize task reassignment and path re-planning for dynamic sudden tasks, and the existing task reassignment algorithm is difficult to realize real-time dynamic reassignment of tasks in the face of dynamic sudden tasks, and has insufficient processing capacity in the face of complex terrain, the present application provides a multi-unmanned aerial vehicle path efficient reconstruction method for complex terrain and sudden tasks. The present application can realize real-time, high-quality low-altitude task reassignment and path re-planning process in the face of a large number of sudden task demands and complex terrain, while meeting the task demand constraints and fuel constraints.

[0007] The technical solutions adopted by the present application are as follows:

[0008] The multi-unmanned aerial vehicle path efficient reconstruction method for complex terrain and sudden tasks has the following steps:

[0009] Step 1: value evaluation is performed on each sudden task point in the sudden task point set, and the value of each sudden task point is obtained;

[0010] Step 2: the values are arranged in descending order to obtain a value set;

[0011] Step 3: for the sudden task point Ta * corresponding to the first value in the current value set, the improved A* algorithm is used to calculate the optimal path list P i corresponding to each unmanned helicopter i if the sudden task point Ta * is inserted into the task sequence T i currently executed by each unmanned helicopter i i* and the cumulative distance change amount Δd(P i ); i respectively takes 1, 2, …, N; N is the total number of unmanned aerial vehicles performing tasks at the initial moment; the cost function of the improved A* algorithm is designed based on a digital elevation map, containing longitude, latitude and height information, and the three-dimensional node search of the original A* algorithm is converted into two-dimensional node search;

[0012] Step 4: screening out the unmanned aerial vehicle set F that still satisfies the fuel constraint after inserting the emergency task point Ta * ;

[0013] F = {i | Δd(P i ) ≤ u i δ i -d(P i )};

[0014] Step 5: judging whether the unmanned aerial vehicle set F is empty, if yes, the current emergency task point Ta * is not executed, and step 8 is entered; if no, step 6 is entered;

[0015] Step 6: screening out the unmanned aerial vehicle i * from the unmanned aerial vehicle set F, which satisfies that the flight time difference between any two of the N unmanned aerial vehicles is as small as possible after inserting the emergency task point Ta * into the task sequence of the unmanned aerial vehicle i * ;

[0016] Step 7: arranging the current emergency task point Ta * to the unmanned aerial vehicle i * , and updating the path list and the shortest flight length of the task sequence currently to be executed by the unmanned aerial vehicle i * ;

[0017] Step 8: deleting the first value in the current value set;

[0018] Step 9: judging whether the current value set is empty, if yes, outputting the path list of the N unmanned aerial vehicles, and ending the flow; if no, returning to step 3.

[0019] Further, the value of each emergency task point in step 1 is calculated according to the following formula:

[0020]

[0021] In the formula,

[0022] val k represents the value of the emergency task point Ta k ; k = 1, 2, …, |S|; |S| is the total number of emergency tasks;

[0023] The function J k For the sudden mission point Ta k Task requirements;

[0024] m i For the quality of drone i;

[0025] k i Let i be the unit mass contribution of UAV i to any task;

[0026] X ik Let the decision variable be of Boolean type. If drone i has already selected the emergency mission point Ta... k Then X ik =1, otherwise X ik =0;

[0027] ρ i (Ta k ) indicates the sudden mission point Ta k The shortest distance to all currently accepted mission points of drone i. Indicates task point Location and sudden mission point Ta k The straight-line distance of the location on the digital elevation map; In the current task sequence being executed by drone i The j-th task.

[0028] Furthermore, the descending order principle in step 2 is: prioritize sorting by Val. k The first element in the list Sort the elements in descending order of size. If the first element is the same as the first element, sort by the second element. Sort the values ​​in descending order.

[0029] Furthermore, the cost function of the improved A* algorithm described in step 3 is:

[0030] F n =ω G G n +ω H H n +ω I I n +ω c C n ;

[0031]

[0032] H n =k lon |lon n -lon Ta |+k lat |latn -lat Ta |;

[0033]

[0034]

[0035] wherein,

[0036] G n ,H n ,I n ,C n are the cumulative map distance cost, the heuristic cost, the height cost and the threat cost of the improved A* algorithm at node n, respectively, whose initial values are all 0;

[0037] ω G ,ω H ,ω I ,ω C are the weight coefficients of G n ,H n ,I n ,C n ;

[0038] G neighbor(n) is the cumulative map distance cost of the improved A* algorithm at the neighbor node of node n;

[0039] k lon and k lat represent the map resolution along the longitude and latitude directions, respectively;

[0040] lon n and lat n are the longitude and latitude coordinates of the current node n, respectively; lon Ta and lat Ta are the longitude and latitude coordinates of the target point Ta, respectively;

[0041] hei n is the terrain height of the current node n, hei min is the lowest elevation of the entire digital elevation terrain, hei max is the highest elevation of the entire digital elevation terrain, and α hei is the equivalent elevation coefficient,

[0042] C0is the intensity of the threat source, h(n, radar) is the straight-line distance from the current node n to the threat source, and R(radar) represents the influence range of the threat source.

[0043] Further, step 3 is specifically:

[0044] Step 3.1: Initialize the index i = 1 of the drones to be traversed;

[0045] Step 3.2: Traverse the current task sequence being executed by drone i Calculate the sudden task point Ta using the following formula. * Optimal insertion position j * :

[0046]

[0047] In the formula, When considering factors such as complex terrain and threat level, the mission point With the sudden mission point Ta * optimal distance between Compared to distance The percentage increase;

[0048] These represent the sudden mission point Ta, considering both complex terrain and threat level factors. * With mission points optimal distance between Compared to distance The percentage increase;

[0049] and They are respectively and Prior estimates;

[0050] This represents the k-th task point in the task sequence of UAV i on the digital elevation map. Location and sudden mission point Ta * The straight-line distance between the positions;

[0051] Ta represents an emergency task point on a digital elevation map. * The location and the (k+1)th task point in the task sequence of drone i The straight-line distance between the positions;

[0052] In the first iteration, i.e. when i = 1, Adopted and The same prior estimation method is used for calculation; when i>1 The value has already been calculated using the improved A* algorithm in the previous task sequence and is a known value, so it can be called directly.

[0053] k = L i hour, This is a hypothetical task point;

[0054] Step 3.3: Calculate the sudden task point Ta * Insert the task sequence currently being executed by drone i The optimal path list after that and cumulative distance change △d(P) i );

[0055] Step 3.3.1: Calculate the current optimal insertion position j using the improved A* algorithm. * And sudden mission point Ta * Optimal path between optimal distance Sudden mission point Ta * and position j * The optimal path between +1 optimal distance And the current optimal insertion position j * Its next task point j * The optimal distance between +1

[0056] Step 3.3.2: Calculate the optimal insertion position j for drone i. * If a sudden task point is inserted at the location * The cumulative distance change △d(P) i ):

[0057]

[0058] In the formula, when the sudden task point Ta * It is inserted at the last task point. After that, and All are 0;

[0059] Step 3.3.3: Calculate the optimal insertion position j at the current position. * If the task sequence currently being executed by drone i is given Insert sudden task point Ta * Then, the optimal path list for drone i. The calculation formula is as follows:

[0060]

[0061] In the formula, when the sudden task point Ta * It is inserted at the last task point. After that, and All are empty sets;

[0062] Step 3.4: judging whether all the UAVs have been traversed, if not, i←i+1, returning to step 3.1; if yes, entering step 4.

[0063] Further, in step 6, the UAV i * is determined by the following formula:

[0064]

[0065] wherein,

[0066] is the average flight speed of the UAV i;

[0067] is the average flight speed of the UAV j;

[0068] d(P i ) is the path length of the original path list of the UAV i before the tentative insertion of the emergent task point Ta * .

[0069] d(P j ) is the path length of the original path list of the UAV j before the tentative insertion of the emergent task point Ta * .

[0070] Further, step 7 is specifically: inserting the emergent task point Ta * into the path list of the task sequence currently executed by the UAV i * to obtain a new path list , updating the new path list currently required to be executed by the UAV i * , inserting the emergent task point Ta * into the path list of the task sequence currently executed by the UAV i * to obtain a new shortest flight length , and updating the shortest flight length of the new path list currently required to be executed by the UAV i * .

[0071] The present application has the following advantages:

[0072] ​​1. The present application comprehensively considers a large number of task demand constraints, fuel constraints, complex elevation terrain and unexpected tasks, and designs an optimal unmanned aerial vehicle selection, a low-altitude real-time local path replanning considering threat degree and digital terrain representation for reasonably evaluating and sorting the value of unexpected tasks to achieve an efficient path reconstruction method, solves the distribution of multiple unmanned aerial vehicles for a large number of task order demands, and solves the problems that it is difficult to balance the calculation efficiency and task allocation quality under the constraints of complex terrain and threat degree, fuel constraints and task constraints, and improves the survival ability and task execution efficiency of multiple unmanned aerial vehicles in the process of executing complex tasks when facing unexpected conditions.

[0073] 2. Compared with the traditional market auction algorithm and the CBBA algorithm, the optimal insertion point strategy proposed in the present application can well reduce the repeated calculation process of the insertion task point cost, effectively reduce the time complexity of calculation, improve the calculation efficiency, and ensure the real-time index of the algorithm.

[0074] 3. In processing new tasks, the PI algorithm is difficult to accurately evaluate the influence of complex terrain, unexpected task demand and other factors on the performance of the unmanned aerial vehicle, and the calculation amount is larger, which leads to low processing efficiency of large-scale unmanned aerial vehicle cluster and a large number of tasks; and the present application fully considers the influence of complex terrain factors, threat degree, fuel constraints, task demand constraints and the like on path reconstruction, can effectively improve the solving efficiency on the basis of considering the solving quality.

[0075] 4. Compared with the MDP method, the present application has the advantages of simple operation, adaptability to more unmanned aerial vehicles and larger task demand, and easy expansion of more task constraints.

[0076] 5. Traditional algorithms such as elastic band and D* algorithm are difficult to process path rapid replanning problems in unstructured environments such as digital elevation terrain (DEM), and such algorithms rarely consider the influence of threat degree, fuel constraints, task demand and the like on path replanning results; the original A* algorithm has the following shortcomings: 1) only applicable to structured terrain and part of obstacle terrain, not applicable to digital elevation terrain; 2) the original A* algorithm has a large space for node expansion when processing 3D path planning, which will reduce the calculation efficiency. The present application improves the cost function of the original A* algorithm to adapt to the optimal path search of digital elevation terrain, thereby reducing the three-dimensional grid search to two-dimensional search. Since the improved cost function proposed in the present application contains latitude and longitude information and height information, the three-dimensional node search of the original A* algorithm can be converted into a two-dimensional node search problem, improving the calculation efficiency. In addition, the improved cost function of the present application is tailor-made for digital elevation maps, and experiments have verified that it has high search efficiency on commonly used digital elevation maps.

[0077] 6. In the process of calculating the best insertion position of a burst task point into the existing task sequence of a UAV by using the formula, the present application uses the priori estimation to calculate and Thus, it is not necessary to call the improved A* algorithm for searching and solving multiple times, and the path searching time can be effectively reduced when inserting a large-scale burst task, and the real-time performance of the algorithm is improved.

[0078] 7. The method of the present application is not only applicable to unmanned helicopters, but also applicable to other types of unmanned aircrafts for low-altitude (near-ground) operations, such as flapping-wing unmanned aircrafts, etc. BRIEF DESCRIPTION OF DRAWINGS

[0079] Figure 1 is a general flowchart of the present application.

[0080] Figure 2 is a schematic diagram of the position of the inserted burst task point, (a) is a schematic diagram of inserting the task point into the middle of the task sequence, and (b) is a schematic diagram of inserting the task point into the tail of the task sequence.

[0081] Figure 3 is a schematic diagram of the position relationship of adjacent nodes.

[0082] Figure 4 is a schematic diagram of node cost calculation.

[0083] Figure 5 is a diagram of the coordinates of the UAV and the burst tasks after sorting by value, in which the red circle is the UAV, and the black plus sign is the burst task point.

[0084] Figure 6 is a process diagram of the time sequence path reconstruction of three UAVs in the embodiment of the present application when processing burst tasks, (a), (b), …, (i) are respectively the time sequence path reconstruction results when processing burst tasks 1, 2, …, 9.

[0085] Figure 7 is a result diagram of three UAVs processing 10 burst tasks in the embodiment of the present application, (a) is a top view, and (b) is a side view.

[0086] Figure 8 is a time curve of three UAVs processing 10 different burst tasks in the embodiment of the present application. DETAILED DESCRIPTION

[0087] The present application will be further described in detail below in combination with the drawings and embodiments.

[0088] Suppose there are N UAVs, and the task sequence currently executed by UAV i is are respectively the coordinates of the 1st, 2nd, …, L i th task points to be executed by UAV i, coordinates of the jth waypoint to be executed by the ith UAV, including longitude coordinate lon j , latitude coordinate lat j , and height coordinate hei j , L i is the current number of tasks in the task sequence T i of the ith UAV, i is the index of the UAV, i respectively takes 1, 2, …, N; the path list of the task sequence T i currently executed by the ith UAV is represents the set of path points containing longitude, latitude, and height coordinates of the ith UAV when executing the task from coordinate point to , represents the straight-line length of the path , the path length of P i is Perform a discrete rasterization operation on the digital elevation map (DEM), and the straight-line distance between any two points x i =(lon i , lat i , hei i ) and x j =(lon j , lat j , hei j ) is h(x i , x j ), and the calculation formula is k lon and k lat represent the map resolution along the longitude and latitude directions, respectively.

[0089] As shown in Figure 1 , when facing complex terrain and unexpected tasks, the present application provides a method for efficient path reconstruction for the N UAVs, which comprises the following steps:

[0090] Step 1: value evaluation is performed on each unexpected task point Ta k in the unexpected task point set Task, to obtain the value val k of each unexpected task point Ta k ;

[0091] Suppose the unexpected task point set Task={Ta1, Ta2, …, Ta |S|}, each unexpected task point Ta k ={lon k , lat k , hei k} in the unexpected task point set Task, k=1, 2, …, |S|, lonk lat k 、hei k These are the sudden mission points Ta k The longitude, latitude, and altitude coordinates of each emergency mission point Ta k value k All calculations are performed using the following formula:

[0092]

[0093] In formula (1), the function J k For the sudden mission point Ta k The mission requirement, taking transport helicopters as an example, can be understood as the weight of supplies to be transported; taking attack helicopters as an example, the mission requirement can be understood as the amount of ammunition needed to carry out an attack mission at a certain mission point; N is the total number of drones; m i For the mass of drone i; k i Let X be the unit mass contribution of UAV i to any task; ik Let the decision variable be of Boolean type. If drone i has already selected the emergency mission point Ta... k Then X ik =1, otherwise X ik =0, this decision variable is used to mark whether drone i performs the emergency mission point Ta. k ;ρ i (Ta k ) represents the sudden mission point Ta k The shortest distance to all currently accepted task points of drone i is calculated using the following formula:

[0094]

[0095] Step 2: For each sudden task point Ta obtained in Step 1 k value k Sort the values ​​in descending order to obtain the value set.

[0096] Sort the value of each sudden task point in the Task set in descending order to obtain the value set Value = (Val1, Val2, ..., Val |S| The sorting rule is: prioritize according to formula (1) Val. k The first element in the list: Sort the elements in descending order of size. If the first element has the same value as the first element (which is possible), then sort by the second element. Sort them in descending order of size. The val calculated by formula (1) is... kIn essence, it's a list containing two elements, both representing the importance of the task. The difference lies in the fact that the first element is more important than the second. For each emerging task point, a list like this is calculated, and the first element of the value of each emerging task point is compared. The emerging task point with the larger first element is considered more important. If the first elements are the same, the second element of the value of each emerging task point is compared; in this case, the task with the larger second element is considered more important.

[0097] Step 3: For the sudden task point Ta corresponding to the first item Val1 in the current value set Value. * Iterate through the drones and calculate the task sequence currently being executed by each drone i. Executing emergency missions at Ta * After the insertion operation, the changes in the optimal path and cumulative distance for each UAV are recorded.

[0098] Insert sudden task point Ta * Position diagram as shown Figure 2 As shown.

[0099] The index i = 1 of the drones to be traversed is initialized. The specific execution process is as follows:

[0100] Step 3.1: Begin locating drones targeting sudden mission points. * Optimal insertion position j * .

[0101] Step 3.2: Traverse the current task sequence being executed by drone i Get the sudden mission point Ta * Optimal insertion position j * The calculation formula is as follows:

[0102]

[0103] In formula (3), and These represent mission points considering both complex terrain and threat level factors. With the sudden mission point Ta * optimal distance between Compared to task points on digital elevation maps With the sudden mission point Ta * The straight-line distance between The relative percentage increase, and the sudden mission point Ta * With mission points optimal distance between Compared to the sudden task points on the digital elevation map, Ta * With mission points The straight-line distance between The relative percentage increase can be approximated as being related only to the terrain and threats at the current location, and not to the specific mission point coordinates; and The value is determined based on the digital elevation map used in the specific implementation of the method of this invention, and is determined through multiple experiments (an experiment using the ratio of the optimal distance between two task points actually obtained by the improved A* algorithm to their straight-line distance). and Let T be the task sequence currently being executed by drone i. i The coordinates of the k-th and (k+1)-th task points, where k takes the values ​​1, 2, ..., L. i -1; This represents a sudden mission point Ta considering complex terrain and threat environments. * and mission points The optimal (shortest) distance between them; This represents a sudden mission point Ta considering complex terrain and threat environments. * and mission points The optimal (shortest) distance between them; and They are respectively and The prior estimate is only used in step 3.2 to obtain the optimal insertion point position j. * ; This represents the k-th task point in the task sequence of UAV i on the digital elevation map. Location and sudden mission point Ta * The straight-line distance between the positions; Ta represents an emergency task point on a digital elevation map. * The location and the (k+1)th task point in the task sequence of drone i The straight-line distance between the positions; in the first iteration, i.e., when i = 1, Prior estimation is also used; when i>1 The value was already calculated using the improved A* algorithm in the previous task sequence and is a known value, so it can be directly called. Additionally, in formula (3), when k takes the value L... i At that time, the following will occur: Figure 2 -(b) shows the mission sequence T of UAV i i In the case of a sudden task inserted at the end, in order to ensure that formula (3) applies to 1≤k≤L i The universality of time, as specified here Here This is a hypothetical task point (which does not actually exist).

[0104] Step 3.3: Calculate the sudden task point Ta * Insert the task sequence currently being executed by drone i The optimal path list P i * and cumulative distance change △d(P) i );

[0105] Step 3.3.1: Calculate the current optimal insertion position j using the improved A* algorithm. * And sudden mission point Ta * Optimal path between optimal distance Sudden mission point Ta * and position j * The optimal path between +1 optimal distance And the current optimal insertion position j * Its next task point j * The optimal distance between +1

[0106] The cost function of node n estimated by the improved A* algorithm is as follows:

[0107] F n =ω G G n +ω H H n +ω I I n +ω c C n (4)

[0108] In formula (4), G n H n ,I n C n These represent the cumulative map distance cost, heuristic cost, altitude cost, and threat cost of the improved A* algorithm at node n, respectively, all with an initial value of 0. The calculation diagrams for these four costs are shown below. Figure 4 As shown, n-1 represents the previous neighbor of the expanded node n, and so on, with n+1 representing the previous neighbor of the expanded node n+2. Alternatively, it can be described as expanding node n-1 to obtain neighbor n, then expanding node n to obtain neighbor n+1, and then expanding node n+1 to obtain neighbor n+2; ω G ,ω H ,ω I ,ω CThese are the weighting coefficients for the cumulative map distance cost, heuristic cost, altitude cost, and threat cost, respectively, given manually. Because G n H n ,I n C n Since the units of measurement are different, the sum of the weighting coefficients does not necessarily have to be 1; generally speaking, to improve the success rate of finding the shortest path, ω should be increased. G However, this will increase the pathfinding time; to reduce the pathfinding time, the coefficient ω of the heuristic cost term should be increased. H However, this might cause the algorithm to struggle to find the global shortest path and easily get trapped in a local shortest path. To increase the influence of the mountain's height, the weight coefficient ω should be increased. I However, this can cause the final path to easily get stuck in mountain peaks and valleys, increasing pathfinding time; to increase the impact of threat sources, the weight coefficient ω should be increased. C However, this may lead to additional, circuitous path overhead, making it difficult to guarantee that the final path is the shortest and most optimal. In practical applications, the appropriate factor should be selected based on the requirements and the advantages and disadvantages of each weighting coefficient.

[0109] The cumulative map distance cost G between node n and its neighboring nodes neighbor(n) neighbor(n) The iterative calculation formula is:

[0110]

[0111] Where r(n,neighbor(n)) represents the positional relationship between node n and its neighboring node neighbor(n); the positional relationship between node neighbor(n) and node n is as follows: Figure 3 As shown, 2 and 6 represent moving left and right along the longitude direction, respectively; 4 and 8 represent moving upward and downward along the latitude direction, respectively; and 1, 3, 5, and 7 represent moving along the southwest 45-degree angle, northwest 45-degree angle, northeast 45-degree angle, and southeast 45-degree angle, respectively. G n G represents the cumulative map distance cost at node n. n The value at the initial node is 0; k lon and k lat These represent the map resolution along the longitude and latitude directions, respectively.

[0112] The heuristic cost H at the current node n n The calculation formula is:

[0113] H n =k lon |lon n -lon Ta |+k lat |lat n-lat Ta | (6)

[0114] Among them, lon n and lat n These are the longitude and latitude coordinates of the current node n, respectively; lon Ta and lat Ta These are the longitude and latitude coordinates of the target point Ta, respectively.

[0115] The height cost I at the current node n n The calculation formula is:

[0116]

[0117] Among them, hei n Let hei be the terrain height of the current node n. min The lowest elevation in the entire digital elevation terrain, hei max α represents the highest elevation in the entire digital elevation topography system. hei The equivalent altitude coefficient is taken by default.

[0118] The threat cost C at the current node n n The calculation formula is:

[0119]

[0120] Wherein, when the threat source is radar, C0 is the intensity of the radar source, h(n,radar) is the straight-line distance from the current node n to the radar, and R(radar) represents the detection range of the radar.

[0121] Step 3.3.2: Calculate the optimal insertion position j for drone i. * If a sudden task point is inserted at the location * The cumulative distance change △d(P) i The calculation formula is as follows:

[0122]

[0123] in, and It is obtained through iterative search using the improved A* algorithm in step 3.3.1; Representative task point and mission points The actual shortest (optimal) path between them is calculated in step 3.1.1 using the improved A* algorithm; when the sudden task point Ta... * It is inserted at the last task point. After that, due to the mission point It does not exist, therefore at this time, in formula (9) and All are 0.

[0124] Step 3.3.3: Calculate the optimal insertion position j at the current position. * If the task sequence currently being executed by drone i is given Insert sudden task point Ta * Then, the optimal path list for drone i. The calculation formula is as follows:

[0125]

[0126] In the formula, The current optimal insertion position j * And sudden mission point Ta * The optimal path between them is determined by configuring the starting point coordinates in step 3.3.1. End point coordinates Ta * Seek; For the sudden mission point Ta * and position j * The optimal path between +1 is determined in step 3.3.1 by configuring the starting coordinates Ta. * End point coordinates Seek; For the task point and the current best insertion position j * The optimal path between them is a known path and does not require reconstruction; For position j * +1 and mission points The optimal path between them is a known path and does not require reconstruction; when a sudden task point Ta... * It is inserted at the last task point. After that, due to the mission point It does not exist, therefore at this time, in formula (10) and All are empty sets.

[0127] Step 3.4: Determine if all drones have been traversed. If not, let i ← i + 1 and return to step 3.1; if yes, proceed to step 4.

[0128] Step 4: Select the drones from all the drones that have been traversed to insert the emergency mission point Ta. * The set F of drones that subsequently satisfy the fuel constraint:

[0129] F={i|△d(P i )≤u i δ i -d(P i (11)

[0130] In the formula, i is the index of the UAV that satisfies the fuel constraint; δ i The flight distance of the UAV per unit fuel consumption; u i d(P) represents the maximum fuel capacity of drone i; i For drone i, tentatively inserting itself into the sudden mission point Ta * Previous original path list Path length, in, Representative path The sum of the lengths of the line segments connecting all path points in a given sequence can be described as: in, This represents drone i executing from coordinate point arrive A set of waypoints containing longitude, latitude, and altitude coordinates during the task.

[0131] Step 5: Determine whether the set F of drones that meet the fuel constraint is empty. If F = φ, it means that the sudden task does not meet the fuel constraint for any drones, and the sudden task will not be executed. Proceed to step 8; otherwise, proceed to step 6.

[0132] Step 6: Select the drone from the set F of drones that meets the fuel constraint that can best guarantee the timing synchronization of all drones, and use it as the final emergency mission point Ta. * Assigned drones i * The calculation formula is as follows:

[0133]

[0134] In the formula, i * Represented as a sudden mission point Ta * Index of the assigned drones; Let i be the average flight speed of the drone. Let d(P) be the average flight speed of the UAV j; i For drone i, tentatively inserting itself into the sudden mission point Ta * Previous original path list The path length, d(P) j For example, the drone j tentatively inserts itself into the sudden mission point Ta. * Previous original path list The path length.

[0135] The most suitable drone i can be selected using equation (12). * To execute emergency mission point Ta * This makes the sudden task point Ta * Insert drone i* After the task sequence, for a drone swarm, the difference in flight time between any two drones should be as small as possible, that is, the difference between the two drones with the largest flight time difference should be as small as possible.

[0136] Step 7: Set the emergency mission point Ta * Assigned to drones i * Update drone i * The current list of paths and the shortest flight length for the task sequence to be executed;

[0137] For sudden mission points Ta * In finding the optimal drone i * After determining its optimal insertion position, the sudden task point Ta will be... * Assigned to drones i * ,renew (Imminent emergency mission point Ta) * Insert drone i * The optimal path list is obtained from the path list of the currently executing task sequence. Updated to drone i * (List of new paths to be executed) (Imminent emergency mission point Ta) * Insert drone i * The new shortest flight length is obtained from the path list of the currently executing task sequence. Updated to drone i * (The shortest flight length of the current list of new paths to be executed).

[0138] Step 8: Update the value set;

[0139] Pop the first item Val1 from the value set Value to get Value←Value-Val1 (that is, delete the first item Val1 in the current value set, update the value set, and take the original second item value as the first item value in the current value set).

[0140] Step 9: Determine if the current value set Value is empty. If Value = φ, output the path list P of the N drones that need to perform the task at the initial moment. i If i = 1, 2, ... N, the process ends; otherwise, return to step 3.

[0141] Example:

[0142] Preparation: Perform a discretization rasterization operation on the digital elevation map (DEM) to facilitate subsequent path replanning operations using the improved A* algorithm. For example, discretize the map within the range of 128.00°E–128.32°E and 44.68°N–45.00°N into a 100×100 raster map, with a resolution k along the longitude direction. lon The value is 352m, and the resolution along the latitudinal direction is k. lat The distance is 362m. The number of UAVs performing the mission, N, is 5, and the number of unexpected missions to be performed is 10. The location coordinates of the unexpected missions are shown in Table 1 below. Assume the average flight speed of UAV i (i = 1, 2... N) is... Both are 25 m / s, and the mass is m. i All weigh 5092 kg, have a maximum range of 482 km, and a maximum endurance of 3 hours and 9 minutes. The amount of fuel u carried by each UAV i (i = 1, 2... N) is... i Both are 1500kg, so the flight distance δ per unit of fuel consumption is... i Both values ​​are 0.3213 km / kg. The location of the threat source is (128.3°E, 45°N). The threat source is radar, and the threat intensity C0 is 15000m. 4 .

[0143] Table 1 Latitude and longitude coordinates of the emergency task location

[0144]

[0145] The specific process of this embodiment is as follows:

[0146] Step 1: Calculate the value (val) of each emergency task point. k The calculation formula is:

[0147] Step 2: Sort the values ​​of each emergency task point in descending order to obtain the value set Value = (Val1, Val2, ..., Val... 10 The priority order of the emergency task point coordinates after sorting in descending order is as follows: Figure 5 As shown in the figure, the red circles represent drones, and the black plus signs represent emergency mission points. The smaller the number corresponding to the black plus sign, the higher the priority of the mission.

[0148] Step 3: For the sudden task point Ta corresponding to the first item Val1 in the value set Value. * Iterate through the drones and calculate the task sequence for each drone i. Executing emergency missions at Ta *After the insertion operation, the changes in the optimal path and cumulative distance for each UAV are recorded.

[0149] Initialize the drone index i = 1 for subsequent traversals. The specific execution process is as follows:

[0150] Step 3.1: Begin locating drones targeting sudden mission points. * Optimal insertion position j * .

[0151] Step 3.2: Traverse the task sequence T currently being executed by drone i. i Get the current emergency task point Ta * In the task sequence The optimal insertion position j * The calculation formula is as follows:

[0152]

[0153] In the formula, This represents the mission point Ta considering complex terrain and threat environments. * and The optimal distance between them; This represents the mission point Ta considering complex terrain and threat environments. * and The optimal distance between them; This represents a mission point considering complex terrain and threat environments. and The optimal distance between them.

[0154] The following formula is used for calculation. and

[0155]

[0156] in, Let be the longitude, latitude, and altitude coordinates of UAV i at the k-th mission point, respectively; Let be the longitude, latitude, and altitude coordinates of UAV i at the (k+1)th mission point, respectively; They are respectively emergency missions Ta * The longitude, latitude, and altitude coordinates; determined through multiple experiments based on the digital elevation map used in this embodiment. The value ranges from 0.1 to 0.3; here we take... This is used to roughly estimate the relative increase in the optimal distance between two task points relative to the straight-line distance between the two task points on the digital elevation map, taking into account factors such as complex terrain and threat level.

[0157] Step 3.3: Calculate the sudden task point Ta * Insert drone i's task sequence The optimal path list after that and the resulting cumulative distance change △d(P) i );

[0158] Step 3.3.1: Insert at the optimal position j * Starting from point Ta, the sudden mission point * Using the improved A* algorithm as the endpoint, the optimal insertion position j is calculated. * And sudden mission point Ta * Optimal path between optimal distance And sudden mission points Ta * and position j * The optimal path between +1 optimal distance

[0159] The following is to obtain and The specific process will be explained using the following example:

[0160] Sub-step 1): Construct an OPEN table to store nodes to be expanded, and a CLOSED table to store explored nodes. To avoid duplication with the previous UAV set label F, Fu is used to represent the cost function F. Here, the cost function Fu is calculated for each node using the aforementioned formula (4). The initial value of this cost function is set to 0, initializing the task node to be executed by UAV i. ( This indicates that the j-th action to be performed by drone i * The coordinates of each task point are included in the OPEN table, and the CLOSED table is empty.

[0161] Sub-step 2): Pop the node n with the smallest cost function Fu value from the OPEN table, and simultaneously add node n to the CLOSED table, then execute sub-step 3); if the CLOSED table has an extended endpoint task Ta... * Then proceed to sub-step 5).

[0162] Sub-step 3): Calculate the relative positions of the eight neighboring nodes (neighbor(n)) of node n using the following formula (e.g., relative positions are as follows). Figure 3 The corresponding cost (as shown):

[0163] Fu neighbor(n) =ω G G neighbor(n) +ω H H neighbor(n) +ω II neighbor(n) +ω c C neighbor(n) (16)

[0164] In the formula, G neighbor(n) H neighbor(n) ,I neighbor(n) C neighbor(n) The numbers represent the cumulative map distance cost, heuristic cost, height cost, and threat cost of the improved A* algorithm when dealing with the neighboring nodes (n) of node n. A schematic diagram illustrating the calculation of these four costs is shown below. Figure 4 As shown; ω G ,ω H ,ω I ,ω C These are the weighting coefficients for cumulative map distance cost, heuristic cost, altitude cost, and threat cost, respectively. Here, we take the weighting coefficient ω. G =0.4, ω H =0.5, ω I =10, ω c =10.

[0165] The cumulative map distance cost G between node n and its neighbor nodes (neighbor(n)) neighbor(n) The calculation formula is:

[0166]

[0167] Where r(neighbor(n),n) represents the positional relationship between node n and its neighboring node neighbor(n). Taking 2 and 6 respectively represents moving left and right along the longitude direction, taking 4 and 8 respectively represents moving up and down along the latitude direction, and taking 1, 3, 5, and 7 respectively represents moving along the southwest 45-degree angle direction, the northwest 45-degree angle direction, the northeast 45-degree angle direction, and the southeast 45-degree angle direction.

[0168] The heuristic cost H of the neighbor(n) of node n neighbor(n) The calculation formula is:

[0169] H neighbor(n) =k lon |lon neighbor(n) -lon Ta |+k lat |lat neighbor(n) -lat Ta | (18)

[0170] Among them, lon neighbor(n) and lat neighbor(n) Let lon be the longitude and latitude coordinates of the neighbor node neighbor(n). Taand lat Ta Here are the longitude and latitude coordinates of the target point Ta.

[0171] The height cost I of the neighboring nodes neighbor(n) of node n neighbor(n) The calculation formula is:

[0172]

[0173] Among them, hei neighbor(n) Let hei be the terrain height of the neighboring node neighbor(n). min The lowest elevation in the entire digital elevation terrain, hei max α represents the highest elevation in the entire digital elevation topography system. hei The equivalent altitude coefficient is taken by default.

[0174] The threat cost C of neighbor(n) of node n neighbor(n) The calculation formula is:

[0175]

[0176] This embodiment considers a radar threat scenario, where the threat intensity C0 represents the strength of the radar source. Generally, the closer the UAV is to the radar source, the higher the threat level, and the threat intensity is inversely proportional to the fourth power of the straight-line distance from the radar source. Beyond a certain distance, the radar can be considered to have no effect on the UAV. h(neighbor(n), radar) represents the straight-line distance from the neighboring node neighbor(n) to the radar, and R(radar) represents the radar's detection range.

[0177] Sub-step 4): If the current neighbor node (neighbor(n)) already exists in the OPEN list, or the current neighbor node (neighbor(n)) already exists in the CLOSED list, record the cost function value of the neighbor node (neighbor(n)) in the OPEN and CLOSED lists as follows: At this point, the cost function value Fu of the neighboring node, calculated using equation (16) above, will be used. neighbor(n) and Compare, if a neighboring node satisfies Then replace the cost function value of the neighbor node neighbor(n) with Fu. neighbor(n) Proceed to sub-step 2).

[0178] Sub-step 5): From the sudden mission point Ta * Start backtracking, searching backwards for the preceding nodes to obtain the optimal path. And the optimal cost function value, i.e. the optimal distance, is calculated.

[0179] Optimal path and optimal distance The method of obtaining and and The method for obtaining them is the same, so it will not be repeated here.

[0180] Step 3.3.2: Obtaining the optimal distance and Based on this, calculate if drone i is at the current optimal insertion position j * Insert sudden mission point Ta at the location * The cumulative distance change Δd(P) that will occur later i The calculation formula is as follows:

[0181]

[0182] Step 3.3.3: Calculate the optimal insertion position j. * The current task sequence being executed by drone i Insert sudden task point Ta * Then, the optimal path list data for drone i. The calculation formula is as follows:

[0183]

[0184] Step 3.4: Determine if all drones have been traversed. If not, let i ← i + 1 and return to step 3.1; if yes, proceed to step 4.

[0185] Step 4: Select the point Ta from all drones that have been traversed to insert the emergency mission. * The set F of drones that subsequently satisfy the fuel constraint:

[0186] F={i|△d(P i )≤u i δ i -d(P i )} (twenty three)

[0187] In the formula, u i δ i The total range of the drone under the current fuel level is approximately 482 km, d(P i Let ) represent the drone i on the current path P. i The total distance traveled.

[0188] Step 5: Determine if the set F of drones satisfying the fuel constraint is empty. If F = φ, it means that no drone can execute the sudden task point Ta. * If yes, proceed to step 8; otherwise, proceed to step 6.

[0189] Step 6: Select the drone from the set F of drones that meets the fuel constraint that can best guarantee the timing synchronization of all drones, and use it as the final emergency mission point Ta. * Assigned drones i * The calculation formula is as follows:

[0190]

[0191] In the formula, i * For the sudden mission point Ta * Index of the assigned drones; Let i be the average flight speed of the drone. Let d(P) be the average flight speed of the UAV j; i For drone i, tentatively inserting itself into the sudden mission point Ta * Previous original path list The path length, d(P) j For example, the drone j tentatively inserts itself into the sudden mission point Ta. * Previous original path list The path length.

[0192] Step 7: Set the emergency mission point Ta * Assigned to drones i * ,renew

[0193] Step 8: Update the value set Value, pop the first item Val1 from the value set Value to get Value←Value-Val1.

[0194] Step 9: Check if the current value set Value is empty. If Value = φ, the algorithm ends and outputs the path list P of all N drones. i If i = 1, 2...N; otherwise, return to step 3.

[0195] Through steps 1-8 above, using emergency tasks 1-10 to sequentially insert tasks into the existing task and path sequences of UAV 1-3, the dynamic path reconstruction results obtained step by step are as follows: Figure 6 As shown in the image (red circles represent drones, and black plus signs represent emergency mission points), the reconstruction path for the last emergency mission 10 is shown below. Figure 7In Figure (a) (red circles represent drones, and black plus signs represent sudden task points), it can be seen that when 10 sudden tasks occur simultaneously, the algorithm calculates their value in descending order based on attributes such as importance and location, assigning them as sudden tasks 1, 2, ..., 10. Then, it performs a tentative insertion operation on the task sequence for all drones in this order. Each drone evaluates the value of the task, calculating the optimal drone and its optimal task point interval that satisfy the constraints, which is used for subsequent insertion operations. It is worth noting that, from... Figure 6 Figure (c) to Figure 6 In the process shown in Figure (d), UAV No. 2 efficiently reconstructed the original task path when handling the task sequence insertion operation for sudden task 4. It is worth noting that this process took less than 3 seconds, demonstrating high timeliness.

[0196] The top and side views of the final path reconstruction result are as follows: Figure 7 As shown, the time curve for each burst task insertion is as follows: Figure 8 As shown. By Figure 7 and Figure 8 It can be seen that, for 10 unexpected tasks, with 3 drones performing the tasks, the average total time for solving the optimal drone, the best insertion position, and the reconstructed path for each unexpected task is about 3 seconds. Furthermore, the final reconstructed paths for all drones also exhibit good coordination while satisfying constraints such as terrain, fuel, and time.

[0197] The flight times for the three drones are shown in Table 2.

[0198] Table 2. Flight times for the three drones

[0199]

[0200] As shown in Table 2, the total flight time of the three drones is around 100 seconds. The difference in total flight time between drone 1 and drone 2 is the largest, at 19 seconds, while the difference between drone 1 and drone 3 is the smallest, at approximately 8 seconds. The difference in the maximum total flight time accounts for approximately 16.3% of the longest flight time of the drones. This indicates that for a large number of drones performing sudden tasks, the drones, while ensuring optimal local reconstruction paths and high computational efficiency, have achieved as much coordination as possible in terms of flight time among the multiple drones. This coordination results in the shortest total task completion time of 116.3695 seconds for all drones.

Claims

1. A method for efficient path reconstruction of multiple unmanned aerial vehicles (UAVs) in complex terrain and sudden missions, characterized in that, Includes the following steps: Step 1: Evaluate the value of each emergency task point in the set of emergency task points to obtain the value of each emergency task point; Step 2: Sort the values ​​in descending order to obtain the value set; Step 3: For the sudden task point Ta corresponding to the first value in the current value set. * Based on the improved A* algorithm, the task sequence T currently being executed by each UAV i is calculated. i Insert sudden task point Ta * Then, the optimal path list P corresponding to each UAV i is... i * and cumulative distance change △d(P) i ); i takes values ​​of 1, 2, ..., N; N is the total number of UAVs executing the task at the initial time; the cost function of the improved A* algorithm is designed based on digital elevation maps and includes longitude, latitude and altitude information, transforming the three-dimensional node search of the original A* algorithm into a two-dimensional node search; Step 4: Filter out the points where sudden tasks can be inserted. * The set of drones F that still meet the fuel constraint: F={i|△d(P i )≤u i δ i -d(P i )}; Step 5: Determine if the drone set F is empty. If so, the current emergency mission point Ta... * If not, proceed to step 8; otherwise, proceed to step 6. Step 6: Select drone i from the drone set F. * To meet the needs of sudden task points Ta * Insert drone i * After the task sequence, the flight time difference between any two of the N drones should be as small as possible; Step 7: Set the current emergency task point Ta * Assigned to drones i * Update drone i * The current list of paths and the shortest flight length for the task sequence to be executed; Step 8: Delete the first value from the current value set; Step 9: Determine if the current value set is empty. If it is, output the path list of N drones, and the process ends. Otherwise, return to step 3.

2. The method for efficient multi-UAV path reconstruction for complex terrain and sudden missions as described in claim 1, characterized in that, The value of each emergency task point in Step 1 is calculated according to the following formula: In the formula, val k Indicates the sudden mission point Ta k The value; k = 1, 2... |S|; |S| is the total number of unexpected tasks; function J k For the sudden mission point Ta k Task requirements; m i For the quality of drone i; k i Let i be the unit mass contribution of UAV i to any task; X ik For Boolean decision variables, if drone i has already selected the emergency mission point Ta... k Then X ik =1, otherwise X ik =0; ρ i (Ta k ) indicates the sudden mission point Ta k The shortest distance to all currently accepted mission points of drone i. Indicates task point Location and sudden mission point Ta k The straight-line distance of the location on the digital elevation map; In the current task sequence being executed by drone i The j-th task.

3. The efficient multi-UAV path reconstructing method for complex terrain and sudden missions according to claim 2, characterized in that, The descending order principle in step 2 is: prioritize sorting by Val. k The first element in the list Sort the elements in descending order of size. If the first element is the same as the first element, sort by the second element. Sort the values ​​in descending order.

4. The efficient multi-UAV path reconstructing method for complex terrain and sudden missions according to claim 3, characterized in that, The cost function of the improved A* algorithm described in step 3 is: F n =ω G G n +oh H H n +oh I I n +oh c C n ; H n =k lon |lon n -lon Ta |+k lat |years n -years Ta |; In the formula, G n H n ,I n C n These represent the cumulative map distance cost, heuristic cost, height cost, and threat cost of the improved A* algorithm at node n, with initial values ​​of 0 for each. ω G ,ω H ,ω I ,ω C G n H n ,I n C n Weighting coefficients; G neighbor(n) The cumulative map distance cost at the neighboring nodes of node n for the improved A* algorithm; k lon and k lat These represent the map resolution along the longitude and latitude directions, respectively; lon n and lat n These are the longitude and latitude coordinates of the current node n, respectively; lon Ta and lat Ta These are the longitude and latitude coordinates of the target point Ta, respectively. hei n Let hei be the terrain height of the current node n. min The lowest elevation in the entire digital elevation terrain, hei max α represents the highest elevation in the entire digital elevation topography system. hei This is the equivalent altitude coefficient. C0 represents the strength of the threat source, h(n,radar) represents the straight-line distance from the current node n to the threat source, and R(radar) represents the range of influence of the threat source.

5. The efficient multi-UAV path reconstructing method for complex terrain and sudden missions according to claim 4, characterized in that, Step 3 specifically involves: Step 3.1: Initialize the index i = 1 of the drones to be traversed; Step 3.2: Traverse the current task sequence being executed by drone i Calculate the sudden task point Ta using the following formula. * Optimal insertion position j * : In the formula, When considering factors such as complex terrain and threat level, the mission point With the sudden mission point Ta * optimal distance between Compared to distance The percentage increase; These represent the sudden mission point Ta, considering both complex terrain and threat level factors. * With mission points optimal distance between Compared to distance The percentage increase; and They are respectively and Prior estimates; This represents the k-th task point in the task sequence of UAV i on the digital elevation map. Location and sudden mission point Ta * The straight-line distance between the positions; Ta represents an emergency task point on a digital elevation map. * The location and the (k+1)th task point in the task sequence of drone i The straight-line distance between the positions; In the first iteration, i.e. when i = 1, Adopted and The same prior estimation method is used for calculation; when i>1 The value has already been calculated using the improved A* algorithm in the previous task sequence and is a known value, so it can be called directly. k = L i hour, This is a hypothetical task point; Step 3.3: Calculate the sudden task point Ta * Insert the task sequence currently being executed by drone i The optimal path list P i * and cumulative distance change △d(P) i ); Step 3.3.1: Calculate the current optimal insertion position j using the improved A* algorithm. * And sudden mission point Ta * Optimal path between optimal distance Sudden mission point Ta * and position j * The optimal path between +1 optimal distance And the current optimal insertion position j * Its next task point j * The optimal distance between +1 Step 3.3.2: Calculate the optimal insertion position j for drone i. * If a sudden task point is inserted at the location * The cumulative distance change △d(P) i ): In the formula, when the sudden task point Ta * It is inserted at the last task point. After that, and All are 0; Step 3.3.3: Calculate the optimal insertion position j at the current position. * If the task sequence currently being executed by drone i is given Insert sudden task point Ta * Then, the optimal path list P for drone i. i * The calculation formula is as follows: In the formula, when the sudden task point Ta * It is inserted at the last task point. After that, and All are empty sets; Step 3.4: Determine if all drones have been traversed. If not, let i ← i + 1 and return to step 3.1; if yes, proceed to step 4.

6. The efficient multi-UAV path reconstructing method for complex terrain and sudden missions according to claim 5, characterized in that, In step 6, the drone i * Determine using the following formula: In the formula, Let i be the average flight speed of the drone. Let j be the average flight speed of the drone. d(P i For drone i, tentatively inserting itself into the sudden mission point Ta * Previous original path list Path length; d(P j For example, the drone j tentatively inserts itself into the sudden mission point Ta. * Previous original path list The path length.

7. The efficient multi-UAV path reconstructing method for complex terrain and sudden missions according to claim 6, characterized in that, Step 7 specifically involves: setting up the emergency task point Ta * Insert drone i * The new path list is obtained after the path list of the currently executing task sequence. Updated to drone i * The current list of new paths to be executed will include the emergency task point Ta. * Insert drone i * The new shortest flight length is obtained from the path list of the currently executing task sequence. Updated to drone i * The shortest flight length of the list of new paths that need to be executed.

Images