Unmanned aerial vehicle path planning method for multi-task cooperation

Abstract

The invention discloses an unmanned aerial vehicle path planning method for multi-task cooperation, and the method comprises the following steps: 1, carrying out the grouping of a plurality of tasks according to the attributes of the tasks, and dividing the tasks into a task group and a single task; 2, performing unmanned aerial vehicle distribution on the task group and the single task according to the attributes of the unmanned aerial vehicles; 3, performing global path planning on a single task; step 4, performing path sharing and collaborative path optimization on the same task group; and 5, adjusting the path of the unmanned aerial vehicle by adopting a local dynamic path in the flight process. Advanced methods in multiple fields such as a distribution mechanism, path planning and energy management are fused, and the unmanned aerial vehicle multi-task cooperation path planning method suitable for the complex environment is provided. The scheme has relatively high innovativeness, and can effectively solve the problem of efficient task cooperation of the unmanned aerial vehicle under the conditions of limited energy and limited time.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicles, and in particular relates to a unmanned aerial vehicle path planning method oriented to multi-task collaboration. Background Art

[0002] With the rapid development of drone technology, its application scope has gradually expanded to logistics distribution, disaster relief, environmental monitoring and intelligent inspection. In these application scenarios, drones not only need to perform multiple tasks efficiently, but also must avoid obstacles in complex and dynamic environments and complete tasks under limited energy and time constraints. However, since drones face problems such as path conflicts, energy constraints, time constraints and dynamic changes of obstacles during flight, how to reasonably plan multi-task collaborative paths to improve task completion efficiency has become a key challenge in current research.

[0003] The essence of the multi-task cooperative path planning problem of UAVs is a typical combinatorial optimization problem, which not only needs to solve the complexity of task allocation and path planning, but also needs to consider the dynamic obstacle avoidance and energy management of UAVs when performing tasks. In task allocation, the energy and task requirements of UAVs must be reasonably matched to ensure that UAVs can complete tasks efficiently without failing midway due to insufficient energy. At the same time, when multiple UAVs perform tasks at the same time, they will face the risk of path conflicts. Path planning must not only minimize energy consumption, but also dynamically adjust the path to avoid collisions with other UAVs or dynamic obstacles.

[0004] Most existing path planning algorithms focus on the task execution of a single UAV, and there is little research on global path optimization and task grouping when multiple UAVs perform tasks in a coordinated manner. In addition, the dynamic adjustment and optimization of multi-task paths of UAVs in complex environments is also a key and difficult point in the current technological development. Summary of the invention

[0005] In view of this, the object of the present invention is to propose a multi-task collaborative UAV path planning method, comprising the following steps:

[0006] Step 1: Group multiple tasks according to task attributes and divide them into task groups and single tasks;

[0007] Step 2, assign drones to task groups and single tasks according to the attributes of drones;

[0008] Step 3: Perform global path planning for a single task;

[0009] Step 4: perform path sharing and collaborative path optimization for the same task group;

[0010] Step 5: Use local dynamic path to adjust the drone path during flight.

[0011] Specifically, multiple tasks are grouped according to task attributes, including grouping tasks according to criteria such as the geographical location of the tasks, time windows, etc., and tasks with similar locations and overlapping time windows are divided into the same group, so that the UAV can efficiently complete the tasks through collaborative optimization. The task division criteria include spatial proximity and time window overlap. Spatial proximity means that the geographical distance between tasks is close, which is conducive to the UAV completing multiple tasks through one flight. Time window overlap means that the time requirements of the tasks are similar, ensuring that the UAV can perform multiple tasks in the same period of time. The tasks are grouped by a clustering algorithm, and each task group G i Represents a set of tasks with geographical and temporal proximity.

[0012] Specifically, the allocation of drones to task groups and ungrouped tasks is performed using an auction mechanism based on the attributes of the drones. According to the capabilities of the drones and the attributes of the tasks or task groups, suitable drones are matched to perform specific tasks or task groups. The allocation of drones is based on the following factors:

[0013] Energy limitation: Based on the energy consumption estimation, ensure that the UAV has enough energy to complete all tasks in the task group;

[0014] Mission load: Consider the load capacity of the drone. If the mission load in the mission group is heavy, choose a drone with strong load capacity.

[0015] Task allocation formula: Use auction mechanism to allocate tasks and select the most suitable drone to perform the task group. For task group G i The task allocation formula is:

[0016]

[0017] Among them, U j It is a drone j, T k It is Task Group G i In the task, Cost(U j ,T k ) is a drone j Execute Task T k The total cost includes energy consumption and time cost.

[0018] Furthermore, the energy consumption estimation includes the following steps:

[0019] The energy consumption includes flight energy consumption and mission execution energy consumption;

[0020] The flight energy consumption calculation formula is: E fly =P fly ·tfly , where P fly Represents flight power in watts, estimated by the flight speed, load, and flight altitude of the drone; t fly Indicates the flight time in seconds, calculated by the flight distance and flight speed;

[0021] The energy consumption of mission execution depends on the type of mission, mission time and the actions that the drone needs to perform during the mission: E task =P task ·t task , P task is the power during task execution, which is related to the task type; t task It is the time for task execution, which is determined by the complexity and execution time of the task;

[0022] The energy consumption is: C d is the drag coefficient of the drone; A is the cross-sectional area of ​​the drone; ρ is the air density; v is the flight speed, d is the distance the drone flies; m is the total mass of the drone and its payload; g is the acceleration due to gravity; η prop is the propeller efficiency of the drone.

[0023] Specifically, the global path planning for a single task includes the following steps:

[0024] Get the starting point p s and the end point p e ;

[0025] Assume that the obstacles in the environment are a polygon set O = {o1, o2, ..., o m}, each obstacle o i By its vertex set V i = {v i1 ,v i2 ,…,v ik}express;

[0026] Initial path search, assuming there are no obstacles, the drone will follow the path from the starting point p s To the end point p e The straight line is defined as: L = λ(p s ,p e );

[0027] Obstacle detection: Check whether path L intersects with any obstacle in obstacle set O. If there is no intersection, path L is the optimal path and no further planning is required; if there is an intersection, proceed to the next step;

[0028] Find guidance points and identify intersecting obstacles: is the set of all obstacles intersecting with path L;

[0029] Calculate the feasible vertices of the obstacle: For each intersecting obstacle o i ∈O intersect , extract the vertex set V of the obstacle i , and define the feasible vertex set V subopt :V subopt ={v∈V i |λ(p s ,v) and O intersect No intersection}, that is, vertex v is the same as the starting point p s The points where the line does not intersect any obstacles;

[0030] Select the optimal guidance point, for each feasible vertex V of the obstacle subopt , select an optimal guidance point p g , so that from the starting point p s to p g The path and from p g To the end point p e The paths do not intersect with obstacles. The selection criteria of the optimal guidance point can be the point with the shortest distance to the target point: d(v,p e ), where d(v,p e ) is the vertex v to the end point p e The Euclidean distance of

[0031] Path update, find the optimal guidance point p g After that, the global path is divided into two segments: L1 = λ(p s ,p g ) and L2=λ(p g ,p e ), path L1 is the path from the starting point to the guidance point, and path L2 is the path from the guidance point to the end point;

[0032] Detect whether the updated path intersects with other obstacles. Check whether paths L1 and L2 intersect with other obstacles. If so, continue to select new guidance points until the path does not intersect with any obstacles.

[0033] When the final generated path L=L1∪L2∪…∪L m When it does not intersect with any obstacles, the path is output as the optimal path, and m represents the number of obstacles.

[0034] Furthermore, the path sharing and collaborative path optimization for the same task group includes the following steps:

[0035] For tasks in the same task group, a path sharing mechanism is adopted. The principle of path sharing is that the UAV can maximize the coverage of the task points in the task group during the flight process to form a coherent path;

[0036] The goal of path sharing is to minimize the total path length:

[0037]

[0038] Among them, L total is the total path length between task points in the task group, p i is the position of task point i, d(p i ,p i+1 ) is the task point p i and p i+1 The distance between them, n is the number of task points in the task group;

[0039] Based on the path sharing in the task group, the optimization algorithm of the traveling salesman problem is used to plan the optimal path of the UAV;

[0040] When the UAVs are executing a task group, they use local dynamic path adjustment to ensure that they avoid collisions with other UAVs or dynamic obstacles during collaborative flight.

[0041] Preferably, the method of using a local dynamic path to adjust the path of the drone during the flight process includes the following:

[0042] Step 1: Use the Markov decision process to predict the future movement trajectory of the obstacle;

[0043] State definition: Define the current state of the obstacle as That is, the position of the obstacle at time t speed and acceleration

[0044] State transition model: The future state of the obstacle is represented by a probability model. Assuming that the state of the obstacle at time t+Δt is S(t+Δt), its state transition probability P(S(t+Δt)|S(t)) is predicted by the following model:

[0045]

[0046] Where f is the state transfer function, which predicts the future position and state based on the current velocity, acceleration and time step Δt;

[0047] Step 2: Obtain the probability trajectory of the obstacle through multiple iterations of the state transition model to predict the possible location of the obstacle in the future;

[0048] Step 3: Collision risk assessment: Based on the probability distribution of the obstacle’s motion trajectory, the drone assesses the risk of collision with the dynamic obstacle.

[0049] Assume that the current position of the drone is p u (t), calculate the drone position p at the future time t+Δt according to the drone’s speed and planned path u (t+Δt), the collision risk is expressed as the probability that the obstacle and the drone are in the same spatial position:

[0050]

[0051] Quantify the collision risk by calculating the overlap probability between the drone and the obstacle in the future time period;

[0052] Safety distance detection: set a minimum safety distance d safe When the distance between the drone and the obstacle is less than this value, it is considered that there is a collision risk. The collision judgment formula is: (t+Δt))≤d safe ;

[0053] Step 4: Local path adjustment. When the system detects a potential collision risk, the drone will dynamically adjust the current local path to avoid obstacles. The adjustment steps are as follows:

[0054] Path deviation calculation: Based on the probability trajectory of the obstacle and the collision risk assessment, the direction and distance that the drone needs to deviate are calculated; the direction of the path deviation is the direction away from the obstacle, and the deviation Δp u To meet the minimum distance for safety distance;

[0055] Path update: The new position of the drone is: p u (t+Δt)=p u (t+Δt)+Δp u

[0056] Step 5: Real-time dynamic adjustment and feedback: During the flight, the drone continuously monitors environmental changes and the movement trajectory of obstacles, updates the collision risk assessment model in real time, and recalculates the local path when a significant change in the movement trajectory of an obstacle is detected;

[0057] Repeat steps 3 and 4 in each time step Δt to ensure that the UAV can fly safely in a dynamic environment.

[0058] The beneficial effects of the present invention are as follows: The present invention takes into account the allocation mechanism of comprehensive energy and time constraints: Based on the traditional allocation mechanism, this scheme adds energy and time constraints to ensure that the task allocation of drones not only optimizes the task completion time, but also takes into account its energy limitations; a path planning algorithm based on a multi-guidance point strategy is proposed, which can quickly handle obstacle avoidance problems in large-scale environments and significantly reduce the amount of calculation; combined with a real-time feedback mechanism, the system can adaptively adjust task allocation and path planning to ensure that the drone can still efficiently complete the task in an emergency situation, and proposes a collaborative optimization strategy for tasks and paths. Through task grouping and path sharing, conflicts between drones are reduced and the overall efficiency of the task is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 Overall flow chart of the video space conversion method. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0062] This embodiment proposes a multi-task collaborative UAV path planning method, including the following steps:

[0063] Step 1: Group multiple tasks according to task attributes and divide them into task groups and single tasks;

[0064] Step 2, assign drones to task groups and single tasks according to the attributes of drones;

[0065] Step 3: Perform global path planning for a single task;

[0066] Step 4: perform path sharing and collaborative path optimization for the same task group;

[0067] Step 5: Use local dynamic path to adjust the drone path during flight.

[0068] Specifically, multiple tasks are grouped according to task attributes, including grouping tasks according to criteria such as the geographical location of the tasks, time windows, etc., and tasks with similar locations and overlapping time windows are divided into the same group, so that the UAV can efficiently complete the tasks through collaborative optimization. The task division criteria include spatial proximity and time window overlap. Spatial proximity means that the geographical distance between tasks is close, which is conducive to the UAV completing multiple tasks through one flight. Time window overlap means that the time requirements of the tasks are similar, ensuring that the UAV can perform multiple tasks in the same period of time. The tasks are grouped by a clustering algorithm, and each task group G i Represents a set of tasks with geographical and temporal proximity.

[0069] Specifically, the allocation of drones to task groups and ungrouped tasks is performed using an auction mechanism based on the attributes of the drones. According to the capabilities of the drones and the attributes of the tasks or task groups, suitable drones are matched to perform specific tasks or task groups. The allocation of drones is based on the following factors:

[0070] Energy limitation: Based on the energy consumption estimation, ensure that the UAV has enough energy to complete all tasks in the task group;

[0071] Mission load: Consider the load capacity of the drone. If the mission load in the mission group is heavy, choose a drone with strong load capacity.

[0072] Task allocation formula: Use auction mechanism to allocate tasks and select the most suitable drone to perform the task group. For task group G i The task allocation formula is:

[0073]

[0074] Among them, U j It is a drone j, T k It is Task Group G i In the task, Cost(U j ,T k ) is a drone j Execute Task T k The total cost includes energy consumption and time cost.

[0075] Furthermore, the energy consumption estimation includes the following steps:

[0076] The energy consumption includes flight energy consumption and mission execution energy consumption;

[0077] The flight energy consumption calculation formula is: E fly =P fly ·t fly , where Pfly Represents flight power in watts, estimated by the flight speed, load, and flight altitude of the drone; t fly Indicates the flight time in seconds, calculated by the flight distance and flight speed;

[0078] The energy consumption of mission execution depends on the type of mission, mission time and the actions that the drone needs to perform during the mission: E task =P task ·t task , P task is the power during task execution, which is related to the task type; t task It is the time for task execution, which is determined by the complexity and execution time of the task;

[0079] The energy consumption is: C d is the drag coefficient of the drone; A is the cross-sectional area of ​​the drone; ρ is the air density; v is the flight speed, d is the distance the drone flies; m is the total mass of the drone and its payload; g is the acceleration due to gravity; η prop is the propeller efficiency of the drone.

[0080] Estimate flight power based on the speed, load, air resistance, etc. of the drone. Estimate flight time based on path length and speed. Estimate the energy consumption of the task based on the task type and execution time. Add the flight energy consumption to the task energy consumption to get the total energy consumption of the drone to perform the task. Through the above energy consumption estimation calculation process, the system can dynamically calculate the energy cost of each drone during the task allocation process to ensure that the drone can efficiently complete the task within its energy range.

[0081] Specifically, the global path planning for a single task includes the following steps:

[0082] Get the starting point p s and the end point p e ;

[0083] Assume that the obstacles in the environment are a polygon set O = {o1, o2, ..., o m}, each obstacle o i By its vertex set V i = {v i1 ,v i2 ,…,v ik}express;

[0084] Initial path search, assuming there are no obstacles, the drone will follow the path from the starting point p s To the end point p e The straight line is defined as: L = λ(p s,p e );

[0085] Obstacle detection: Check whether path L intersects with any obstacle in obstacle set O. If there is no intersection, path L is the optimal path and no further planning is required; if there is an intersection, proceed to the next step;

[0086] Find guidance points and identify intersecting obstacles: is the set of all obstacles intersecting with path L;

[0087] Calculate the feasible vertices of the obstacle: For each intersecting obstacle o i ∈O intersect , extract the vertex set V of the obstacle i , and define the feasible vertex set V subopt :V subopt ={v∈V i |λ(p s ,v) and O intersect No intersection}, that is, vertex v is the same as the starting point p s The points where the line does not intersect any obstacles;

[0088] Select the optimal guidance point, for each feasible vertex V of the obstacle subopt , select an optimal guidance point p g , so that from the starting point p s to p g The path and from p g To the end point p e The paths do not intersect with obstacles. The selection criteria of the optimal guidance point can be the point with the shortest distance to the target point: d(v,p e ), where d(v,p e ) is the vertex v to the end point p e The Euclidean distance of

[0089] Path update, find the optimal guidance point p g After that, the global path is divided into two segments: L1 = λ(p s ,p g ) and L2=λ(p g ,p e ), path L1 is the path from the starting point to the guidance point, and path L2 is the path from the guidance point to the end point;

[0090] Detect whether the updated path intersects with other obstacles. Check whether paths L1 and L2 intersect with other obstacles. If so, continue to select new guidance points until the path does not intersect with any obstacles.

[0091] When the final generated path L=L1∪L2∪…∪L m When it does not intersect with any obstacles, the path is output as the optimal path, and m represents the number of obstacles.

[0092] Furthermore, the path sharing and collaborative path optimization for the same task group includes the following steps:

[0093] For tasks in the same task group, a path sharing mechanism is adopted. The principle of path sharing is that the UAV can maximize the coverage of the task points in the task group during the flight process to form a coherent path;

[0094] The goal of path sharing is to minimize the total path length:

[0095]

[0096] Among them, L total is the total path length between task points in the task group, p i is the position of task point i, d(p i ,p i+1 ) is the task point p i and p i+1 The distance between them, n is the number of task points in the task group;

[0097] Based on the path sharing in the task group, the optimization algorithm of the traveling salesman problem is used to plan the optimal path of the UAV;

[0098] When the UAVs are executing a task group, they use local dynamic path adjustment to ensure that they avoid collisions with other UAVs or dynamic obstacles during collaborative flight.

[0099] Preferably, the method of using a local dynamic path to adjust the path of the drone during the flight process includes the following:

[0100] Step 1: Use the Markov decision process to predict the future movement trajectory of the obstacle;

[0101] State definition: Define the current state of the obstacle as That is, the position of the obstacle at time t speed and acceleration

[0102] State transition model: The future state of the obstacle is represented by a probability model. Assuming that the state of the obstacle at time t+Δt is S(t+Δt), its state transition probability P(S(t+Δt)|S(t)) is predicted by the following model:

[0103]

[0104] Where f is the state transfer function, which predicts the future position and state based on the current velocity, acceleration and time step Δt;

[0105] Step 2: Obtain the probability trajectory of the obstacle through multiple iterations of the state transition model to predict the possible location of the obstacle in the future;

[0106] Step 3: Collision risk assessment: Based on the probability distribution of the obstacle’s motion trajectory, the drone assesses the risk of collision with the dynamic obstacle.

[0107] Assume that the current position of the drone is p u (t), calculate the drone position p at the future time t+Δt according to the drone’s speed and planned path u (t+Δt), the collision risk is expressed as the probability that the obstacle and the drone are in the same spatial position:

[0108]

[0109] Quantify the collision risk by calculating the overlap probability between the drone and the obstacle in the future time period;

[0110] Safety distance detection: set a minimum safety distance d safe When the distance between the drone and the obstacle is less than this value, it is considered that there is a collision risk. The collision judgment formula is: d(p u (t+Δt),

[0111] Step 4: Local path adjustment. When the system detects a potential collision risk, the drone will dynamically adjust the current local path to avoid obstacles. The adjustment steps are as follows:

[0112] Path deviation calculation: Based on the probability trajectory of the obstacle and the collision risk assessment, the direction and distance that the drone needs to deviate are calculated; the direction of the path deviation is the direction away from the obstacle, and the deviation Δp u To meet the minimum distance for safety distance;

[0113] Path update: The new position of the drone is: p u (t+Δt)=p u (t+Δt)+Δp u

[0114] Step 5: Real-time dynamic adjustment and feedback: During the flight, the drone continuously monitors environmental changes and the movement trajectory of obstacles, updates the collision risk assessment model in real time, and recalculates the local path when a significant change in the movement trajectory of an obstacle is detected;

[0115] Repeat steps 3 and 4 in each time step Δt to ensure that the UAV can fly safely in a dynamic environment.

[0116] The advantages and beneficial effects of the present invention are as follows: the technical solution integrates advanced methods in multiple fields such as allocation mechanism, path planning and energy management, and proposes a multi-task collaborative path planning method for UAVs in complex environments. The solution is highly innovative and can effectively solve the problem of efficient task collaboration of UAVs under limited energy and time constraints.

[0117] As used herein, the word "preferred" is intended to be used as an example, instance, or illustration. Any aspect or design described herein as "preferred" is not necessarily to be construed as being more advantageous than other aspects or designs. On the contrary, the use of the word "preferred" is intended to present concepts in a specific way. The term "or" as used in this application is intended to mean an inclusive "or" rather than an exclusive "or". That is, unless otherwise specified or clear from the context, "X uses A or B" means any one of the naturally included permutations. That is, if X uses A; X uses B; or X uses both A and B, then "X uses A or B" is satisfied in any of the foregoing examples.

[0118] Moreover, although the present disclosure has been shown and described with respect to one or implementations, those skilled in the art will think of equivalent variations and modifications based on the reading and understanding of this specification and the accompanying drawings. The present disclosure includes all such modifications and variations, and is limited only by the scope of the appended claims. In particular, with respect to the various functions performed by the above-mentioned components (such as elements, etc.), the terms used to describe such components are intended to correspond to any component (unless otherwise indicated) that performs the specified function of the component (such as it is functionally equivalent), even if the structure is not equivalent to the disclosed structure of the function in the exemplary implementation of the present disclosure shown herein. In addition, although the specific features of the present disclosure have been disclosed with respect to only one of several implementations, such features can be combined with one or other features of other implementations that may be desired and advantageous for a given or specific application. Moreover, insofar as the terms "including", "having", "containing" or their variations are used in specific embodiments or claims, such terms are intended to be included in a manner similar to the term "comprising".

[0119] The functional units in the embodiments of the present invention may be integrated into a processing module, or each unit may exist physically separately, or multiple or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium. The above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc. The above-mentioned devices or systems may execute the storage method in the corresponding method embodiment.

[0120] To sum up, the above embodiment is an implementation mode of the present invention, but the implementation mode of the present invention is not limited by the embodiment, and any other changes, modifications, substitutions, combinations, and simplifications that deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A multi-task collaborative UAV path planning method, characterized in that: The steps include: Step 1: Group multiple tasks according to task attributes and divide them into task groups and single tasks; Step 2, assign drones to task groups and single tasks according to the attributes of drones; Step 3: Perform global path planning for a single task; Step 4: perform path sharing and collaborative path optimization for the same task group; Step 5: Use local dynamic path to adjust the drone path during flight.

2. A multi-task collaborative UAV path planning method according to claim 1, characterized in that: Multiple tasks are grouped according to task attributes, including grouping tasks according to criteria such as the geographical location of the task, time window, etc., and tasks with similar locations and overlapping time windows are divided into the same group, so that the UAV can efficiently complete the tasks through collaborative optimization. The task division criteria include spatial proximity and time window overlap. Spatial proximity means that the geographical distance between tasks is close, which is conducive to the UAV completing multiple tasks through one flight. Time window overlap means that the time requirements of the tasks are similar, ensuring that the UAV can perform multiple tasks in the same period of time. The tasks are grouped by a clustering algorithm, and each task group G i Represents a set of tasks with geographical and temporal proximity.

3. The multi-task collaborative UAV path planning method according to claim 2, characterized in that: The allocation of drones to task groups and ungrouped tasks based on the attributes of the drones is carried out using an auction mechanism. Based on the capabilities of the drones and the attributes of the tasks or task groups, suitable drones are matched to perform specific tasks or task groups. The allocation of drones is based on the following factors: Energy limitation: Based on the energy consumption estimation, ensure that the UAV has enough energy to complete all tasks within the task group; Mission load: Consider the load capacity of the drone. If the mission load in the mission group is heavy, choose a drone with strong load capacity. Task allocation formula: Use auction mechanism to allocate tasks and select the most suitable drone to perform the task group. i The task allocation formula is: Among them, U j It is a drone j, T k It is Task Group G i The tasks in Cost(U j ,T k ) is a drone j Execute Task T k The total cost includes energy consumption and time cost.

4. The multi-task collaborative UAV path planning method according to claim 3, characterized in that: The energy consumption estimation comprises the following steps: The energy consumption includes flight energy consumption and mission execution energy consumption; The flight energy consumption calculation formula is: E fly =P fly ·t fly , where P fly Represents flight power in watts, estimated by the flight speed, load, and flight altitude of the drone; t fly Indicates the flight time in seconds, calculated by the flight distance and flight speed; The energy consumption of mission execution depends on the type of mission, mission time and the actions that the drone needs to perform during the mission: E task =P task ·t task , P task is the power during task execution, which is related to the task type; t task It is the time for task execution, which is determined by the complexity and execution time of the task; The energy consumption is: C d is the drag coefficient of the drone; A is the cross-sectional area of ​​the drone; ρ is the air density; v is the flight speed, d is the distance the drone flies; m is the total mass of the drone and its payload; g is the acceleration due to gravity; η prop is the propeller efficiency of the drone.

5. The multi-task collaborative UAV path planning method according to claim 4, characterized in that: The global path planning for a single task includes the following steps: Get the starting point p s and the end point p e ; Assume that the obstacles in the environment are a polygon set O = {o1, o2, ..., o m }, each obstacle o i By its vertex set V i = {v i1 ,v i2 ,…,v ik }express; Initial path search, assuming there are no obstacles, the drone will follow the path from the starting point p s To the end point p e The straight line is defined as: L = λ(p s ,p e ); Obstacle detection: Check whether path L intersects with any obstacle in obstacle set O. If there is no intersection, path L is the optimal path and no further planning is required; if there is an intersection, proceed to the next step; Find guidance points and identify intersecting obstacles: is the set of all obstacles intersecting with path L; Calculate the feasible vertices of the obstacle: For each intersecting obstacle o i ∈O intersect , extract the vertex set V of the obstacle i , and define the feasible vertex set V subopt :V subopt ={v∈V i |λ(p s ,v) and O intersect No intersection}, that is, vertex v is the same as the starting point p s The points where the line does not intersect any obstacles; Select the optimal guidance point, for each feasible vertex V of the obstacle subopt , select an optimal guidance point p g , so that from the starting point p s to p g The path and from p g To the end point p e The paths do not intersect with obstacles. The selection criteria of the optimal guidance point can be the point with the shortest distance to the target point: Among them, d(v,p e ) is the vertex v to the end point p e The Euclidean distance of Path update, find the optimal guidance point p g After that, the global path is divided into two segments: L1 = λ(p s ,p g ) and L2=λ(p g ,p e ), path L1 is the path from the starting point to the guidance point, and path L2 is the path from the guidance point to the end point; Detect whether the updated path intersects with other obstacles. Check whether paths L1 and L2 intersect with other obstacles. If so, continue to select new guidance points until the path does not intersect with any obstacles. When the final generated path L=L1∪L2∪…∪L m When it does not intersect with any obstacles, the path is output as the optimal path, and m represents the number of obstacles.

6. A multi-task collaborative UAV path planning method according to claim 5, characterized in that: The path sharing and collaborative path optimization for the same task group includes the following steps: For tasks in the same task group, a path sharing mechanism is adopted. The principle of path sharing is that the UAV can maximize the coverage of the task points in the task group during the flight process to form a coherent path; The goal of path sharing is to minimize the total path length: Among them, L total is the total path length between task points in the task group, p i is the position of task point i, d(p i ,p i+1 ) is the task point p i and p i+1 The distance between them, n is the number of task points in the task group; Based on the path sharing in the task group, the optimization algorithm of the traveling salesman problem is used to plan the optimal path of the UAV; When the UAVs are executing a task group, they use local dynamic path adjustment to ensure that they avoid collisions with other UAVs or dynamic obstacles during collaborative flight.

7. A multi-task collaborative UAV path planning method according to claim 6, characterized in that: The local dynamic path is used to adjust the drone path during the flight process, including the following: Step 1: Use the Markov decision process to predict the future movement trajectory of the obstacle; State definition: Define the current state of the obstacle as That is, the position of the obstacle at time t speed and acceleration State transition model: The future state of the obstacle is represented by a probability model. Assuming that the state of the obstacle at time t+Δt is S(t+Δt), its state transition probability P(S(t+Δt)|S(t)) is predicted by the following model: Where f is the state transfer function, which predicts the future position and state based on the current velocity, acceleration and time step Δt; Step 2: Obtain the probability trajectory of the obstacle through multiple iterations of the state transition model to predict the possible location of the obstacle in the future; Step 3: Collision risk assessment: Based on the probability distribution of the obstacle’s motion trajectory, the drone assesses the risk of collision with the dynamic obstacle. Assume that the current position of the drone is p u (t), calculate the drone position p at the future time t+Δt according to the drone’s speed and planned path u (t+Δt), the collision risk is expressed as the probability that the obstacle and the drone are in the same spatial position: Quantify the collision risk by calculating the overlap probability between the drone and the obstacle in the future time period; Safety distance detection: set a minimum safety distance d safe When the distance between the drone and the obstacle is less than this value, it is considered that there is a collision risk. The collision judgment formula is: d(p u (t+Δt), Step 4: Local path adjustment. When the system detects a potential collision risk, the drone will dynamically adjust the current local path to avoid obstacles. The adjustment steps are as follows: Path deviation calculation: Based on the probability trajectory of the obstacle and the collision risk assessment, the direction and distance that the drone needs to deviate are calculated; the direction of the path deviation is the direction away from the obstacle, and the deviation Δp u To meet the minimum distance for safety distance; Path update: The new position of the drone is: p u (t+Δt)=p u (t+Δt)+Δp u Step 5: Real-time dynamic adjustment and feedback: During the flight, the drone continuously monitors environmental changes and the movement trajectory of obstacles, updates the collision risk assessment model in real time, and recalculates the local path when a significant change in the movement trajectory of an obstacle is detected; Repeat steps 3 and 4 in each time step Δt to ensure that the UAV can fly safely in a dynamic environment.

Images