Unmanned aerial vehicle scheduling management and control method and system based on nest

Abstract

The invention relates to an unmanned aerial vehicle scheduling management and control method and system based on a nest, and the method comprises the following steps: S1, segmenting an inspection region, and searching an optimal distribution scheme of unmanned aerial vehicles and the nest by using improved particle swarm optimization (PSO); s2, the scheduling control center receives the task request, decomposes the task into sub-tasks according to the task requirement, and allocates proper aircraft nests and unmanned aerial vehicles to process tasks according to the real-time states of the aircraft nests and the unmanned aerial vehicles; s3, calling a high-precision map and a path planning algorithm to plan a flight path; s4, the unmanned aerial vehicle receives the task instruction and the flight plan, and navigation and obstacle avoidance are autonomously completed through a GNSS and an airborne sensor; s5, the dispatching center monitors the state of the unmanned aerial vehicle in real time; and S6, after the unmanned aerial vehicle completes the task, selecting an optimal parking point according to the distribution and state of the aircraft nest. The unmanned aerial vehicle cluster efficiency, the dynamic response capability and the task execution reliability are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles (UAVs), and in particular to a method and system for dispatching and controlling UAVs based on a nest. Background Art

[0002] Drone technology has rapidly developed in recent years, finding widespread application in logistics, inspection, security, and surveillance. However, as business needs become more complex, challenges with drone mission scheduling, flight path management, and energy endurance have become increasingly prominent. Furthermore, in large-scale mission scenarios, autonomous drones alone are unable to meet the demands of multitasking and efficient operation. Consequently, a drone-based scheduling and control method has emerged. By deploying multiple distributed drone nests as support equipment and combining a global management and scheduling system with the operational capabilities of individual drones, this method achieves centralized, automated management, and efficient dispatching of drones.

[0003] However, traditional drone scheduling methods have limitations: ① The emergency scheduling coordination capabilities between multiple drones are insufficient, resulting in low resource utilization; ② The flight path planning algorithm is slow to respond to dynamic environments, unable to quickly avoid obstacles or adjust paths; ③ The coordination efficiency between drone nests and drones is low, for example, task allocation lacks predictability and charging resources and maintenance operations cannot be dynamically optimized; ④ Complex factors such as terrain and weather are not adequately considered in mission planning, which can easily lead to flight safety issues. Therefore, a system solution that combines evolutionary algorithms and path planning technology to achieve optimal drone nest distribution, intelligent task decomposition, dynamic path planning, and real-time monitoring of the entire process is urgently needed. Summary of the Invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide a drone scheduling and control method and system based on a machine nest, which significantly improves the efficiency, dynamic response capability and task execution reliability of drone clusters.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for dispatching and controlling drones based on a drone nest comprises the following steps:

[0007] S1: Segment the inspection area and use improved particle swarm optimization (PSO) to find the optimal distribution plan for drones and nests;

[0008] S2: The dispatch control center receives the task request, decomposes the task into subtasks according to the task requirements, and assigns appropriate processing tasks to the machine nest and UAV based on the real-time status of the machine nest and UAV;

[0009] S3: Call high-precision maps and path planning algorithms to plan the flight path;

[0010] S4: The machine nest completes the takeoff preparations for the UAV. The UAV receives the mission instructions and flight plan, and autonomously completes navigation and obstacle avoidance through GNSS and onboard sensors.

[0011] S5: The dispatch center monitors the status of the drone in real time. If any abnormality is found, the center will immediately adjust the mission or take over the drone according to the nearest nest, and send the drone to the nearest nest for maintenance or charging.

[0012] S6: After the UAV completes its mission, it selects the optimal parking point based on the distribution and status of the machine nests, and lands at the corresponding machine nest for charging, maintenance, or standby.

[0013] Furthermore, the improved particle swarm optimization (PSO) is used to find the optimal distribution scheme of drones and nests, as follows:

[0014] Set the number of particles n and the number of iterations T max , learning factors c1, c2 and inertia factor ω, initializing particle position and velocity, initializing the global optimal solution g and the historical optimal solution p of particle i i ; Among them, the position of particle i represents a solution, including the inspection area and the allocation combination X of the machine nest and the drone i :

[0015]

[0016] in, Indicates whether the drone j in the i-th particle is responsible for inspecting area a; Indicates whether drone j in particle i is assigned to nest k;

[0017] The speed of each particle in, represents the adjustment weight of the task allocation of UAV j to the inspection area a; represents the allocation adjustment weight of UAV j to machine nest k;

[0018] Calculate the fitness of each particle based on the comprehensive objective function and penalty function;

[0019] Use the improved PSO formula to update the particle velocity in the tth iteration and And the continuous value position of the particle is discretized to 0 or 1 through the sigmoid function:

[0020]

[0021] in, is the speed of particle i corresponding to the task assignment in the tth iteration; is the velocity of particle i in the tth iteration corresponding to the machine nest allocation; is the dynamic inertia factor; is the individual historical optimal position of the i-th particle drone j in the mission area a; is the value of whether the i-th particle drone j is responsible for the inspection task area a at the t-th iteration; Assign the individual historical optimal position of the i-th particle drone j to the machine nest k; is the value of whether the i-th particle drone j is assigned to the machine nest k in the t-th iteration; g a,j is the global optimal solution; r1 and r2 are random values ​​distributed in [0,1] to enhance the randomness of the search; ω max 、ω min is the initial value and minimum value of the inertia factor; t is the current number of iterations; g a,j is the global historical optimal position of UAV j in the mission area a among all particles; g j,k The global historical optimal position of UAV j assigned to the nest k among all particles;

[0022] The position update is calculated using the following formula to update the inspection area task allocation:

[0023]

[0024] Updated drone nest and drone allocation:

[0025]

[0026] The continuous position variable and Discretize to {0,1} to adapt to scheduling requirements:

[0027] For the task assignment part:

[0028]

[0029] Allocation of machine nests and drones:

[0030]

[0031] Among them, σ is the sigmoid function; ζ is the random discretization threshold;

[0032] Record and update the historical optimal position and global optimal position of the particle;

[0033] Repeat until the maximum number of iterations is reached or the fitness converges.

[0034] Furthermore, the fitness of each particle is calculated based on the comprehensive objective function and penalty function, as follows:

[0035] The comprehensive objective function F of particle i ifor:

[0036] F i =αC1+βC2+γC3;

[0037] Among them, C1, C2, and C3 are the balanced cost of inspection area allocation, the cost of drone flight distance and energy consumption, and the constraint cost of nest capacity allocation; α, β, and γ are the weighted coefficients of the objective function;

[0038] The inspection area allocation balance cost C1 ensures that the inspection area task requirements are evenly distributed to the drones, with the goal of minimizing the inspection allocation deviation:

[0039]

[0040] Among them, w a is the task weight of area a; N is the total number of areas; K is the total number of drones;

[0041] The UAV flight distance and energy consumption cost C2, minimize the UAV flight distance and energy consumption:

[0042]

[0043] Among them, d a,j represents the flight distance of UAV j to reach area a during the inspection mission;

[0044] The nest capacity allocation constraint cost C3 ensures that the nest allocation is reasonable and the use of nest k does not exceed the nest capacity C k :

[0045]

[0046] Wherein, M represents the total number of machine nests;

[0047] Check the constraints of the machine nest capacity and task allocation, and the scheme that exceeds the limit will be punished by the penalty function. constraint:

[0048]

[0049] Among them, λ is a penalty factor used to increase the fitness value of particles that deviate from the constraints.

[0050] Furthermore, S2 is specifically:

[0051] Receive Q task requests Tq, cluster and sort the tasks by location, deadline and priority, divide the tasks into regions, and generate a subtask set S;

[0052] Get the real-time status of the drone, including location, power, load, speed; get the drone's H mThe current state, including the position h m and available berth C m ;

[0053] Construct the task allocation optimization objective function F, and use the genetic algorithm GA to optimize the UAV-task allocation vector and the UAV-nest return vector;

[0054] Eliminate infeasible solutions based on mission uniqueness, payload, power, time, and nest capacity constraints;

[0055] Determine the final UAV-task allocation vector and UAV-nest return vector, and send flight mission instructions to the corresponding UAV.

[0056] Furthermore, S3 is specifically:

[0057] Build a 3D grid map based on a high-precision map of the inspection area, marking no-fly zones, obstacles, and other environmental features;

[0058] Use the A* algorithm to plan the initial path from the starting point s to the target point g p ;

[0059] Define a comprehensive cost function c(x,y,z), taking into account length, safety, weather, and altitude factors, and adjust the weight coefficients. Each grid point (x,y,z) is assigned a comprehensive cost c(x,y,z) as the cost evaluation standard in path planning:

[0060] c(x,y,z)=c distance +λ1c weather +λ2c altitude +λ3c safety ;

[0061] Among them, c distance is the unit path length; c weather is the cost of the impact of climate conditions, and the greater the wind speed, the greater the cost; c altitude is the cost of deviating from the optimal flight altitude, and the lower or higher the altitude deviates from the optimal flight altitude, the higher the cost; c safety is the safety cost of the distance to the obstacle, and the closer the distance to the obstacle, the greater the cost; λ1, λ2, and λ3 are the weight coefficients of each factor;

[0062] The final output path points are passed to the UAV flight control system to generate flight mission instructions.

[0063] Furthermore, S4 is specifically as follows: the machine nest completes charging, communication checking and sensor initialization for the drone for takeoff, and the drone receives path points and mission parameters sent by the dispatch center; completes navigation based on GNSS, IMU and path tracking control algorithms; detects obstacles through onboard sensors, and adjusts the flight trajectory based on obstacle avoidance strategies.

[0064] Furthermore, navigation is completed based on GNSS, IMU and path tracking control algorithms, as follows:

[0065] Get the current position of the drone P current and the target position P target The error between p :

[0066] e p =P target -P current ;

[0067] The position error is corrected according to the proportional-integral-differential controller to generate the desired speed command v ref :

[0068]

[0069] Among them, K p , K i , K d are the proportional, integral, and differential gains of the controller; t′ represents the time;

[0070] According to the desired speed command v ref and target heading ψ target , generate the desired attitude angle, including pitch θ ref , yaw ψ ref :

[0071]

[0072] v z 、v x 、v y Represent the expected speed of the drone in the x, y, and z directions respectively;

[0073] The target heading ψ provided by the GNSS signal target Based on the yaw ψ ref To make real-time adjustments:

[0074] Δψ=ψ target -ψ ref ;

[0075] After compensating the heading error Δψ, the target steering of the UAV is adjusted.

[0076] Furthermore, the aircraft detects obstacles through onboard sensors and adjusts the flight trajectory based on the obstacle avoidance strategy, as follows:

[0077] Construct an artificial potential field model and assume that the target point position generates an attractive force F attractive , the obstacle position generates a repulsive force F repulsive, the combined force F of the drone total for:

[0078] F total =F attractive +F repulsive ;

[0079] Attraction:

[0080] F attractive =μ·||P target -P current ||;

[0081] Among them, μ is the attraction weight;

[0082] Repulsive force:

[0083]

[0084] Among them, P obstacle is the obstacle position; η is the repulsive force weight, d safe For safe distance;

[0085] When the current planned path is blocked by an obstacle, the D* algorithm is used to replan the local path. total Calculate the desired velocity of the drone.

[0086] Furthermore, S6 is specifically:

[0087] Retrieve the real-time status of all nests, combine nest selection with the remaining flight capabilities of the UAV, and evaluate the feasibility of all candidate nests;

[0088] Construct a comprehensive cost function and select the nest with the lowest cost. Each nest H m The comprehensive cost C m Defined as:

[0089] C m =λ4c distance,m +λ5c load,m +λ6c charge,m +λ7c weather,m ;

[0090] c distance，m =d m ·η adjusted ;

[0091]

[0092] c weather，m =w v ·(1-cos(w d -ψ m ))

[0093] Among them, λ4, λ5, λ6, and λ7 are the weight coefficients of the cost term; c distance,m 、c load,m 、c charge,m and c weather,m They are respectively the flight distance cost, the nest load cost, the charging capacity cost and the environmental adaptability cost; d m From the current position to the machine nest H m Flight distance; is the actual energy consumption coefficient per unit distance; η fly is the basic energy consumption coefficient per unit distance; w v is the wind speed in the current flight area; w d is the current wind direction angle; ψ is the flight direction angle of the UAV; v cruise is the cruising speed of the UAV; C m For the machine nest H m Total berth capacity; b m For the machine nest H m The current number of free berths; P charge,m For the machine nest H m Single charging port power; p m For the machine nest H m The number of currently available charging ports; ∈ is a very small positive number; ψ m Fly the drone to the target nest H m Flight direction angle at time ;

[0094] Use GNSS and sensors to achieve precise landing at the selected parking point.

[0095] A drone scheduling and control system based on a nest includes a nest base station, drones, and a scheduling control center. Specifically, the nest base station is used for drone parking, charging, battery replacement, data exchange, and task allocation operations; the drone is equipped with positioning, communication, and task execution modules for performing specific scheduling tasks; and the scheduling control center is responsible for task allocation, route planning, resource scheduling, and global monitoring decisions.

[0096] The present invention has the following beneficial effects:

[0097] 1. This invention makes drone mission execution more intelligent through the dispatching control center's task decomposition and matching algorithm, intelligent path planning, and precise autonomous navigation and obstacle avoidance technology. Task allocation fully considers multiple factors such as distance, task type, payload capacity, and the drone's battery status, making task decomposition and allocation more reasonable. At the same time, a high-precision dynamic path planning algorithm combines factors such as weather conditions, terrain, and no-fly zones to generate the optimal route, avoiding flight risks. During autonomous navigation, the obstacle avoidance capabilities of GNSS and sensors are combined to significantly improve operational safety and adapt to mission requirements in complex environments.

[0098] 2. This invention globally optimizes the distribution of inspection areas, machine nests, and drones, comprehensively considering the resource scheduling of drones and machine nests, ensuring reasonable division of inspection areas and balanced resource allocation, minimizing duplication of work and resource waste, and improving overall scheduling efficiency. It also avoids resource retention or shortages caused by unreasonable machine nest locations or inflexible drone resource allocation, achieving optimal matching between drones and machine nests.

[0099] 3. This invention emphasizes real-time monitoring and dynamic scheduling capabilities between the drone nest and the dispatch control center. If an emergency occurs (such as low battery or blocked path), the system can quickly respond, dynamically adjusting the mission or directly taking over the drone's operation and guiding it to the nearest drone nest for maintenance or charging. Once the mission is completed, the drone can find the optimal parking spot based on the drone nest status, ensuring efficient utilization of drone nest resources and significantly enhancing the system's adaptability and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0101] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0102] refer to Figure 1 In this embodiment, a method for dispatching and controlling drones based on a drone nest is provided, comprising the following steps:

[0103] S1: Segment the inspection area and use improved particle swarm optimization (PSO) to find the optimal distribution plan for drones and nests;

[0104] S2: The dispatch control center receives the task request, decomposes the task into subtasks according to the task requirements, and assigns appropriate processing tasks to the machine nest and UAV according to their real-time status;

[0105] S3: Call high-precision maps and path planning algorithms to plan the flight path;

[0106] S4: The machine nest completes the takeoff preparations for the UAV. The UAV receives the mission instructions and flight plan, and autonomously completes navigation and obstacle avoidance through GNSS and onboard sensors.

[0107] S5: The dispatch center monitors the status of the drone (battery level, location, route deviation, etc.) in real time. If any abnormality is found (such as low battery or blocked path), the center will immediately adjust the mission or take over the drone according to the nearest nest, and send the drone to the nearest nest for maintenance or charging.

[0108] S6: After the UAV completes its mission, it selects the optimal parking point based on the distribution and status of the machine nests, and lands at the corresponding machine nest for charging, maintenance, or standby.

[0109] In this embodiment, the improved particle swarm optimization (PSO) is used to find the optimal distribution scheme of drones and nests, as follows:

[0110] Set the number of particles n and the number of iterations T max , learning factors c1, c2 and inertia factor ω, initializing particle position and velocity, initializing the global optimal solution g and the historical optimal solution p of particle i i ; Among them, the position of particle i represents a solution, including the inspection area and the allocation combination X of the machine nest and the drone i :

[0111]

[0112] in, Indicates whether the drone j in the i-th particle is responsible for inspecting area a; Indicates whether drone j in particle i is assigned to nest k;

[0113] The speed of each particle in, represents the adjustment weight of the task allocation of UAV j to the inspection area a; represents the allocation adjustment weight of UAV j to machine nest k;

[0114] Calculate the fitness of each particle based on the comprehensive objective function and penalty function;

[0115] Use the improved PSO formula to update the particle velocity in the tth iteration and And the continuous value position of the particle is discretized to 0 or 1 through the sigmoid function:

[0116]

[0117] in, is the speed of particle i corresponding to the task assignment in the tth iteration; is the velocity of particle i in the tth iteration corresponding to the machine nest allocation; is the dynamic inertia factor; is the individual historical optimal position of the i-th particle drone j in the mission area a; is the value of whether the i-th particle drone j is responsible for the inspection task area a at the t-th iteration; Assign the individual historical optimal position of the i-th particle drone j to the machine nest k; is the value of whether the i-th particle drone j is assigned to the machine nest k in the t-th iteration; g a,j is the global optimal solution; r1 and r2 are random values ​​distributed in [0,1] to enhance the randomness of the search; ω max 、ω min is the initial value and minimum value of the inertia factor; t is the current number of iterations; g a,j is the global historical optimal position of UAV j in the mission area a among all particles; g j,k The global historical optimal position of UAV j assigned to the nest k among all particles;

[0118] The position update is calculated using the following formula to update the inspection area task allocation:

[0119]

[0120] Updated drone nest and drone allocation:

[0121]

[0122] The continuous position variable and Discretize to {0,1} to adapt to scheduling requirements:

[0123] For the task assignment part:

[0124]

[0125] Allocation of machine nests and drones:

[0126]

[0127] Among them, σ is the sigmoid function; ζ is the random discretization threshold;

[0128] Record and update the historical optimal position and global optimal position of the particle;

[0129] Repeat until the maximum number of iterations is reached or the fitness converges.

[0130] In this embodiment, the fitness of each particle is calculated based on the comprehensive objective function and penalty function, as follows:

[0131] The comprehensive objective function F of particle i i for:

[0132] F i =αC1+βC2+γC3;

[0133] Among them, C1, C2, and C3 are the balanced cost of inspection area allocation, the cost of drone flight distance and energy consumption, and the constraint cost of nest capacity allocation; α, β, and γ are the weighted coefficients of the objective function;

[0134] The inspection area allocation balance cost C1 ensures that the inspection area task requirements are evenly distributed to the drones, with the goal of minimizing the inspection allocation deviation:

[0135]

[0136] Among them, w a is the task weight of area a; N is the total number of areas; K is the total number of drones;

[0137] The UAV flight distance and energy consumption cost C2, minimize the UAV flight distance and energy consumption:

[0138]

[0139] Among them, d a,j represents the flight distance of UAV j to reach area a during the inspection mission;

[0140] The nest capacity allocation constraint cost C3 ensures that the nest allocation is reasonable and the use of nest k does not exceed the nest capacity C k :

[0141]

[0142] Among them, M represents the total number of machine nests;

[0143] Check the constraints of the machine nest capacity and task allocation, and the scheme that exceeds the limit will be punished by the penalty function. constraint:

[0144]

[0145] Among them, λ is a penalty factor used to increase the fitness value of particles that deviate from the constraints.

[0146] In this embodiment, S2 is specifically:

[0147] Receive Q task requests Tq, cluster and sort the tasks by location, deadline and priority, divide the tasks into regions, and generate a subtask set S;

[0148] Get the real-time status of the drone, including location, power, load, speed; get the drone's nest H m The current state, including the position h m and available berth C m ;

[0149] Construct the task allocation optimization objective function F, and use the genetic algorithm GA to optimize the UAV-task allocation vector and the UAV-nest return vector;

[0150] Eliminate infeasible solutions based on mission uniqueness, payload, power, time, and nest capacity constraints;

[0151] Determine the final UAV-task allocation vector and UAV-nest return vector, and send flight mission instructions to the corresponding UAV.

[0152] In this embodiment, S3 is specifically:

[0153] Build a 3D grid map based on a high-precision map of the inspection area, marking no-fly zones, obstacles, and other environmental features;

[0154] Use the A* algorithm to plan the initial path from the starting point s to the target point g p ;

[0155] Define a comprehensive cost function c(x,y,z), taking into account length, safety, weather, and altitude factors, and adjust the weight coefficients. Each grid point (x,y,z) is assigned a comprehensive cost c(x,y,z) as the cost evaluation standard in path planning:

[0156] c(x,y,z)=c distance +λ1c weather +λ2c altitude +λ3c safety ;

[0157] Among them, c distance is the unit path length; c weather is the cost of the impact of climate conditions, and the greater the wind speed, the greater the cost; c altitude is the cost of deviating from the optimal flight altitude, and the lower or higher the altitude deviates from the optimal flight altitude, the higher the cost; c safety is the safety cost of the distance to the obstacle, and the closer the distance to the obstacle, the greater the cost; λ1, λ2, and λ3 are the weight coefficients of each factor;

[0158] The final output path points are passed to the UAV flight control system to generate flight mission instructions.

[0159] In this embodiment, S4 is specifically as follows: the machine nest completes charging, communication checking and sensor initialization for the drone for takeoff, and the drone receives path points and mission parameters sent by the dispatch center; completes navigation based on GNSS, IMU and path tracking control algorithms; detects obstacles through onboard sensors, and adjusts the flight trajectory based on obstacle avoidance strategies.

[0160] In this embodiment, navigation is accomplished based on GNSS, IMU, and a path tracking control algorithm, as follows:

[0161] Get the current position of the drone P current and the target position P target The error between p :

[0162] e p =P target -P urrent ;

[0163] The position error is corrected according to the proportional-integral-differential controller to generate the desired speed command v ref :

[0164]

[0165] Among them, K p , K i , K d are the proportional, integral, and differential gains of the controller; t′ represents the time;

[0166] According to the desired speed command v ref and target heading ψ target , generate the desired attitude angle, including pitch θ ref , yaw ψ ref :

[0167]

[0168] v z 、v x 、v y Represent the expected speed of the drone in the x, y, and z directions respectively;

[0169] The target heading ψ provided by the GNSS signal target Based on the yaw ψ ref To make real-time adjustments:

[0170] Δψ=ψ target -ψ ref ;

[0171] After compensating the heading error Δψ, the target steering of the UAV is adjusted.

[0172] In this embodiment, obstacles are detected by onboard sensors, and the flight trajectory is adjusted based on the obstacle avoidance strategy, as follows:

[0173] Construct an artificial potential field model and assume that the target point position generates an attractive force F attractive , the obstacle position generates a repulsive force F repulsive , the combined force F of the drone total for:

[0174] Ftotal =F attractive +F repulsive ;

[0175] Attraction:

[0176] F attractive =μ·||P target -P current ||;

[0177] Among them, μ is the attraction weight;

[0178] Repulsive force:

[0179]

[0180] Among them, P obstacle is the obstacle position; η is the repulsive force weight, d safe For safe distance;

[0181] When the current planned path is blocked by an obstacle, the D* algorithm is used to replan the local path. total Calculate the desired velocity of the drone.

[0182] In this embodiment, S6 is specifically:

[0183] Retrieve the real-time status of all drone nests (candidate parking points), combine the drone nest selection with the remaining flight capabilities (battery power, range), and evaluate the feasibility of all candidate drone nests;

[0184] Construct a comprehensive cost function and select the nest with the lowest cost. Each nest H m The comprehensive cost C m Defined as:

[0185] C m =λ4c distance,m +λ5c load,m +λ6c charge,m +λ7c weather,m ;

[0186] c distance，m =d m ·η adjusted ;

[0187]

[0188] c weather，m =w v ·(1-cos(w d -ψ m ))

[0189] Among them, λ4, λ5, λ6, and λ7 are the weight coefficients of the cost term; cdistance,m 、c load,m 、c charge,m and c weather,m They are respectively the flight distance cost, the nest load cost, the charging capacity cost and the environmental adaptability cost; d m From the current position to the machine nest H m Flight distance; is the actual energy consumption coefficient per unit distance; η fly is the basic energy consumption coefficient per unit distance; w v is the wind speed in the current flight area; w d is the current wind direction angle; ψ is the flight direction angle of the UAV; v cruise is the cruising speed of the UAV; C m For the machine nest H m Total berth capacity; b m For the machine nest H m The current number of free berths; P charge,m For the machine nest H m Single charging port power; p m For the machine nest H m The number of currently available charging ports; ∈ is a very small positive number; ψ m Fly the drone to the target nest H m Flight direction angle at time ;

[0190] Use GNSS and sensors to achieve precise landing at the selected parking point.

[0191] A drone scheduling and control system based on a nest includes a nest base station, drones, and a scheduling control center. Specifically, the nest base station is used for drone parking, charging, battery replacement, data exchange, and task allocation operations; the drone is equipped with positioning, communication, and task execution modules for performing specific scheduling tasks; and the scheduling control center is responsible for task allocation, route planning, resource scheduling, and global monitoring decisions.

[0192] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0193] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0194] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0195] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0196] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other manner. Any person skilled in the art may utilize the above-disclosed technical content to modify or modify the present invention into equivalent embodiments. However, any simple modifications, equivalent variations, and modifications to the above embodiments that do not depart from the technical content of the present invention and are based on the technical essence of the present invention remain within the scope of protection of the present invention.

Claims

1. A method for dispatching and controlling drones based on a drone nest, characterized in that: The following steps are involved: S1: Segment the inspection area and use improved particle swarm optimization (PSO) to find the optimal distribution plan for drones and nests; S2: The dispatch control center receives the task request, decomposes the task into subtasks according to the task requirements, and assigns appropriate processing tasks to the machine nest and UAV according to their real-time status; S3: Call high-precision maps and path planning algorithms to plan the flight path; S4: The machine nest completes the takeoff preparations for the UAV. The UAV receives the mission instructions and flight plan, and autonomously completes navigation and obstacle avoidance through GNSS and onboard sensors. S5: The dispatch center monitors the status of the drone in real time. If any abnormality is found, the center will immediately adjust the mission or take over the drone according to the nearest nest, and send the drone to the nearest nest for maintenance or charging. S6: After the UAV completes its mission, it selects the optimal parking point based on the distribution and status of the machine nests, and lands at the corresponding machine nest for charging, maintenance, or standby.

2. The method for dispatching and controlling unmanned aerial vehicles based on a nest according to claim 1, characterized in that: The improved particle swarm optimization (PSO) is used to find the optimal distribution scheme of drones and nests as follows: Set the number of particles n and the number of iterations T max , learning factors c1, c2 and inertia factor ω, initializing particle position and velocity, initializing the global optimal solution g and the historical optimal solution p of particle i i ; Among them, the position of particle i represents a solution, including the inspection area and the allocation combination X of the machine nest and the drone i : in, Indicates whether the drone j in the i-th particle is responsible for inspecting area a; Indicates whether drone j in particle i is assigned to nest k; The speed of each particle in, represents the adjustment weight of the task allocation of UAV j to the inspection area a; represents the allocation adjustment weight of UAV j to machine nest k; Calculate the fitness of each particle based on the comprehensive objective function and penalty function; Use the improved PSO formula to update the particle velocity in the tth iteration and And the continuous value position of the particle is discretized to 0 or 1 through the sigmoid function: in, is the speed of particle i corresponding to the task assignment in the tth iteration; is the velocity of particle i in the tth iteration corresponding to the machine nest allocation; is the dynamic inertia factor; is the individual historical optimal position of the i-th particle drone j in the mission area a; is the value of whether the i-th particle drone j is responsible for the inspection task area a at the t-th iteration; Assign the individual historical optimal position of the i-th particle drone j to the machine nest k; is the value of whether the i-th particle drone j is assigned to the machine nest k in the t-th iteration; g a,j is the global optimal solution; r1 and r2 are random values ​​distributed in [0,1] to enhance the randomness of the search; ω max 、ω min is the initial value and minimum value of the inertia factor; t is the current number of iterations; g a,j is the global historical optimal position of UAV j in the mission area a among all particles; g j,k The global historical optimal position of UAV j assigned to the nest k among all particles; The position update is calculated using the following formula to update the inspection area task allocation: Updated drone nest and drone allocation: The continuous position variable and Discretize to {0,1} to adapt to scheduling requirements: For the task assignment part: Allocation of machine nests and drones: Among them, σ is the sigmoid function; ζ is the random discretization threshold; Record and update the historical optimal position and global optimal position of the particle; Repeat until the maximum number of iterations is reached or the fitness converges.

3. The method for dispatching and controlling unmanned aerial vehicles based on a nest according to claim 2, characterized in that: The fitness of each particle is calculated based on the comprehensive objective function and penalty function, as follows: The comprehensive objective function F of particle i i for: F i =αC1+βC2+γC3; Among them, C1, C2, and C3 are the balanced cost of inspection area allocation, the cost of drone flight distance and energy consumption, and the constraint cost of nest capacity allocation; α, β, and γ are the weighted coefficients of the objective function; The inspection area allocation balance cost C1 ensures that the inspection area task requirements are evenly distributed to the drones, with the goal of minimizing the inspection allocation deviation: Among them, w a is the task weight of area a; N is the total number of areas; K is the total number of drones; The UAV flight distance and energy consumption cost C2, minimize the UAV flight distance and energy consumption: Among them, d a,j represents the flight distance of UAV j to reach area a during the inspection mission; The nest capacity allocation constraint cost C3 ensures that the nest allocation is reasonable and the use of nest k does not exceed the nest capacity C k : Among them, M represents the total number of machine nests; Check the constraints of the machine nest capacity and task allocation, and the scheme that exceeds the limit will be punished by the penalty function. constraint: Among them, λ is a penalty factor used to increase the fitness value of particles that deviate from the constraints.

4. The method for dispatching and controlling drones based on a nest according to claim 1, characterized in that: The S2 is specifically: Receive Q task requests Tq, cluster and sort the tasks by location, deadline and priority, divide the tasks into regions, and generate a subtask set S; Get the real-time status of the drone, including location, power, payload, and speed; Get the machine nest H m The current state, including the position h m and available berth C m ; Construct the task allocation optimization objective function F, and use the genetic algorithm GA to optimize the UAV-task allocation vector and the UAV-nest return vector; Eliminate infeasible solutions based on mission uniqueness, payload, power, time, and nest capacity constraints; Determine the final UAV-task allocation vector and UAV-nest return vector, and send flight mission instructions to the corresponding UAV.

5. The method for dispatching and controlling unmanned aerial vehicles based on a nest according to claim 1, characterized in that: The S3 is specifically: Build a 3D grid map based on a high-precision map of the inspection area, marking no-fly zones, obstacles, and other environmental features; Use the A* algorithm to plan the initial path from the starting point s to the target point g p ; Define a comprehensive cost function c(x,y,z), taking into account length, safety, weather, and altitude factors, and adjust the weight coefficients. Each grid point (x,y,z) is assigned a comprehensive cost c(x,y,z) as the cost evaluation standard in path planning: c(x,y,z)=c distance +λ1c weather +λ2c altitude +λ3c safety ; Among them, c distance is the unit path length; c weather is the cost of the impact of climate conditions, and the greater the wind speed, the greater the cost; c altitude is the cost of deviating from the optimal flight altitude, and the lower or higher the altitude deviates from the optimal flight altitude, the higher the cost; c safety is the safety cost of the distance to the obstacle, and the closer the distance to the obstacle, the greater the cost; λ1, λ2, and λ3 are the weight coefficients of each factor; The final output path points are passed to the UAV flight control system to generate flight mission instructions.

6. The method for dispatching and controlling unmanned aerial vehicles based on a nest according to claim 1, characterized in that: The S4 is specifically as follows: the machine nest completes charging, communication checking and sensor initialization for the drone for takeoff, and the drone receives path points and mission parameters sent by the dispatch center; completes navigation based on GNSS, IMU and path tracking control algorithms; detects obstacles through onboard sensors and adjusts the flight trajectory based on obstacle avoidance strategies.

7. The method for dispatching and controlling unmanned aerial vehicles based on a nest according to claim 6, characterized in that: The navigation is completed based on GNSS, IMU and path tracking control algorithm, as follows: Get the current position of the drone P current and the target position P target The error between p : e p =P target -P current ; The position error is corrected according to the proportional-integral-differential controller to generate the desired speed command v ref : Among them, K p , K i , K d are the proportional, integral, and differential gains of the controller; t′ represents the time; According to the desired speed command v ref and target heading ψ target , generate the desired attitude angle, including pitch θ ref , yaw ψ ref : v z 、v x 、v y Represent the expected speed of the drone in the x, y, and z directions respectively; The target heading ψ provided by the GNSS signal target Based on the yaw ψ ref To make real-time adjustments: Δψ=ψ target -ψ ref ; After compensating the heading error Δψ, the target steering of the UAV is adjusted.

8. The method for dispatching and controlling unmanned aerial vehicles based on a nest according to claim 6, characterized in that: The onboard sensors detect obstacles and adjust the flight trajectory based on the obstacle avoidance strategy, as follows: Construct an artificial potential field model and assume that the target point position generates an attractive force F attractive , the obstacle position generates a repulsive force F repulsive , the combined force F of the drone total for: F total =F attractive +F repulsive ; Attraction: F attractive =μ·||P target -P current ||; Among them, μ is the attraction weight; Repulsive force: Among them, P obstacle is the obstacle position; η is the repulsive force weight, d safe For safe distance; When the current planned path is blocked by an obstacle, the D* algorithm is used to replan the local path. total Calculate the desired velocity of the drone.

9. The method for dispatching and controlling unmanned aerial vehicles based on a nest according to claim 1, characterized in that: The S6 is specifically: Retrieve the real-time status of all nests, combine nest selection with the remaining flight capabilities of the UAV, and evaluate the feasibility of all candidate nests; Construct a comprehensive cost function and select the nest with the lowest cost. Each nest H m The comprehensive cost C m Defined as: C m =λ4c distance,m +λ5c load,m +λ6c charge,m +λ7c weather,m ; c distance,m =d m ·or adjusted ; c weather,m =in v ·(1-cos(in d -ψ m )) Among them, λ4, λ5, λ6, and λ7 are the weight coefficients of the cost term; c distance,m 、c load,m 、c charge,m and c weather,m They are respectively the flight distance cost, the nest load cost, the charging capacity cost and the environmental adaptability cost; d m From the current position to the machine nest H m Flight distance; is the actual energy consumption coefficient per unit distance; η fly is the basic energy consumption coefficient per unit distance; w v is the wind speed in the current flight area; w d is the current wind direction angle; ψ is the flight direction angle of the UAV; v cruise is the cruising speed of the UAV; C m For the machine nest H m Total berth capacity; b m For the machine nest H m The current number of free berths; P charge,m For the machine nest H m Single charging port power; p m For the machine nest H m The number of currently available charging ports; ∈ is a very small positive number; ψ m Fly the drone to the target nest H m Flight direction angle at time ; Use GNSS and sensors to achieve precise landing at the selected parking point.

10. A drone dispatching and control system based on a machine nest, characterized in that: It includes a machine nest base station, drones and a dispatching control center, as follows: the machine nest base station is used for drone parking, charging, battery replacement, data exchange, and task allocation operations; the drone is equipped with positioning, communication and task execution modules to perform specific dispatching tasks; the dispatching control center is responsible for task allocation, route planning, resource scheduling and global monitoring decisions.

Images