Unmanned aerial vehicle cluster collaborative area search energy consumption optimization method

Abstract

The invention discloses an unmanned aerial vehicle cluster collaborative area search energy consumption optimization method, and relates to the technical field of three-dimensional traffic system and unmanned aerial vehicle communication and control technology intersection, and the method comprises the steps: building a collaborative area search system model of an unmanned aerial vehicle cluster at a current time slot based on an unmanned aerial vehicle cluster collaborative area search scene, then establishing a flight movement energy consumption model of the unmanned aerial vehicle based on a collaborative region search system model, calculating an energy consumption model, and meanwhile, considering a calculation unloading strategy to construct an optimization target of taking the minimum energy consumption of the unmanned aerial vehicle cluster system as an optimization target; according to the method, an unmanned aerial vehicle cluster collaborative area searching energy consumption optimization model with the flight path and the calculation unloading proportion of an unmanned aerial vehicle as optimization variables is established, the energy consumption optimization model is solved, and the flight path and task calculation unloading of the unmanned aerial vehicle are comprehensively considered. Reasonable scheduling and allocation of local computing resources and effective cooperation and communication of the unmanned aerial vehicle cluster are realized, and the system energy consumption of the unmanned aerial vehicle cluster is effectively reduced.

Description

Technical Field

[0001] The present application relates to the interdisciplinary technical field of three-dimensional transportation systems and UAV communication and control technologies, and in particular to a method for optimizing energy consumption in collaborative regional searches of UAV clusters. Background Art

[0002] With the national initiative to build a comprehensive, three-dimensional transportation network, this system, powered by aerial platforms like unmanned aerial vehicles (UAVs) and aircraft, and ground platforms like intelligent connected vehicles and road infrastructure, is entering a new phase of rapid development. It represents a leading edge in the development of integrated transportation. Among these, drone technology has garnered widespread attention in the wireless communications sector due to its continued growth and broad application prospects. Compared to large aerial platforms like airships and hot air balloons, drones offer advantages such as low cost, portability, and ease of deployment. They are frequently used in disaster relief, temporary communications, remote area communications, and aerial inspections.

[0003] Drones' high dynamic mobility and ability to carry communications components have led to their widespread use in area search. Drone area search technology enables comprehensive aerial surveillance and real-time data collection and processing, helping users obtain accurate regional information and take effective early warning or preventive measures. Currently, it's common to use a single drone to search a small, pre-planned target area. As search areas expand, coordinated area searches using drone swarms can significantly reduce mission time, lower search costs, and enable more efficient and flexible search missions.

[0004] Mobile edge computing (MEC) refers to the deployment of computing and storage resources at the edge of mobile networks to provide an IT service environment and cloud computing capabilities for mobile networks, thereby offering users ultra-low latency and high-bandwidth network service solutions. The core technology of MEC is computational offloading, and the key lies in making appropriate offloading decisions. Currently, the computational offloading strategies of some methods aim to optimize task latency, or set the offloading strategy to offload redundant tasks to the base station when the drone cannot process task data locally within the specified time. The computational offloading strategy is designed taking into account the overall energy consumption of the drone system. The drone's local computing system can dynamically and adaptively adjust the offloading ratio based on the current load and working status, while taking into account the transmission distance to the offloading base station, so as to reduce the overall energy consumption to a greater extent.

[0005] Due to the limited capacity of onboard batteries, energy consumption has long been a key factor restricting the performance and application scope of drones performing area search missions. Effectively reducing the energy consumption of drone swarms during coordinated area search and improving energy efficiency is a pressing issue. Optimizing the communication computing resources and flight control during drone operation to reduce both communication computing and flight energy consumption is crucial, thereby improving drone endurance and mission execution efficiency. Currently, when faced with challenges in the processing power and energy consumption of drones completing missions, deploying drones to offload computational tasks to ground base stations is often considered. While this approach reduces the computational burden and energy consumption of drones, it is susceptible to terrain restrictions, limiting missions to designated areas with access to ground base stations and failing to fully utilize the inherent mobility of drones.

[0006] Therefore, how drone clusters schedule and allocate local computing resources, and how they effectively collaborate and communicate to minimize energy consumption, are key issues that need to be addressed when performing regional search missions. Summary of the Invention

[0007] The purpose of this application is to provide a method for optimizing energy consumption of collaborative regional search of drone clusters, which can realize reasonable scheduling and allocation of local computing resources and effective collaboration and communication of drone clusters, and effectively reduce the system energy consumption of drone clusters.

[0008] To achieve the above objectives, this application provides the following solutions.

[0009] The present application provides a method for optimizing energy consumption of collaborative regional search of a drone cluster, comprising the following steps.

[0010] Based on the search scenario of drone clusters, a collaborative area search system model of the current time slot drone cluster is constructed; the drone cluster includes an auxiliary drone and several search drones; the search scenario includes the target area, the mission cycle of the search coverage mission, the flight altitude of the search drone and the flight altitude of the auxiliary drone; the collaborative area search system model includes the flight trajectory model of the search drone and the flight trajectory model of the auxiliary drone; the auxiliary drone is used to move from the specified initial position to the final position within the mission cycle, and to assist in the calculation of the unloading task of the search drone; the search drone is used to perform the search coverage mission of the target area.

[0011] According to the collaborative regional search system model of the current time slot, the mobile energy consumption model, local computing energy consumption model and computing offloading energy consumption model of each search UAV in the current time slot are established, as well as the mobile energy consumption model and local computing energy consumption model of the auxiliary UAV.

[0012] According to the mobile energy consumption model, local computing energy consumption model and computing offloading energy consumption model of each search UAV in the current time slot, as well as the mobile energy consumption model and local computing energy consumption model of the auxiliary UAV, an energy consumption optimization model of the collaborative regional search of the UAV cluster in the current time slot is established; the energy consumption optimization model of the collaborative regional search of the UAV cluster includes an objective function and constraints; the objective function is a function with the goal of minimizing the energy consumption of the system; the constraints include offloading power constraints, task completion time constraints, computing resource constraints, auxiliary UAV energy constraints, task offloading ratio constraints, flight speed constraints and search area coverage constraints; the task offloading ratio is the ratio of tasks offloaded from the search UAV to the auxiliary UAV.

[0013] The energy consumption optimization model of the UAV cluster collaborative regional search in the current time slot is solved to obtain the optimal energy consumption optimization plan corresponding to the current time slot; the optimal energy consumption optimization plan includes the flight trajectory of each search UAV, the flight trajectory of the auxiliary UAV and the task offloading ratio in the current time slot.

[0014] According to the specific embodiments provided in this application, this application discloses the following technical effects: This application provides a method for optimizing the energy consumption of collaborative regional search of drone clusters. First, based on the collaborative regional search scenario of the drone cluster, a collaborative regional search system model of the drone cluster in the current time slot is constructed. Then, based on the collaborative regional search system model, a flight movement energy consumption model and a computing energy consumption model of the drone are established (the computing energy consumption model includes a local computing energy consumption model and a computing unloading energy consumption model). At the same time, considering the computing unloading strategy (i.e., the task unloading ratio), a drone cluster collaborative regional search energy consumption optimization model is constructed with the minimum energy consumption of the drone cluster system as the optimization goal, and the flight trajectory of the drone and the computing unloading ratio as the optimization variables. The drone cluster collaborative regional search energy consumption optimization model is solved, and the drone cluster collaborative regional search energy consumption optimization model is solved. Taking into account the flight trajectory of the drone and the task computing unloading, the reasonable scheduling and allocation of local computing resources and the effective collaboration and communication of the drone cluster are realized, and the system energy consumption of the drone cluster is effectively reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A flow chart of the energy consumption optimization method for collaborative regional search of drone clusters provided in an embodiment of the present application.

[0017] Figure 2Schematic diagram of the specific implementation process of the energy consumption optimization method for collaborative regional search of drone clusters provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0019] With the increasing capabilities of drones in sensing, communication, and computing, they can now carry computing, storage, communication, and sensor hardware. Modified drones can even function as base stations thanks to their computing capabilities. By leveraging their mobility and altitude advantages, drones can fly and perform computing tasks in areas where auxiliary computing is needed, significantly improving data processing capabilities while reducing data transmission latency.

[0020] The purpose of this application is to provide a method for optimizing the energy consumption of collaborative regional search of drone clusters, which aims to comprehensively consider the flight trajectory and task calculation offloading of drones, realize the reasonable scheduling and allocation of local computing resources and the effective collaboration and communication of drone clusters, and effectively reduce the system energy consumption of drone clusters. A drone cluster is composed of a group of search drones and auxiliary drones. The search drones perform search and coverage tasks in the target area, and the auxiliary drones serve as mobile base stations for auxiliary calculations. They dynamically adjust the deployment position according to the situation of the task object to provide computing services, and maneuverably reduce the time delay and energy consumption costs generated during the transmission process. At the same time, a dynamic calculation offloading strategy is designed by jointly considering the flight trajectory planning of drones, and an optimization model is constructed with the goal of minimizing the energy consumption of the drone cluster system. The model is solved based on the particle swarm algorithm, which can effectively reduce the energy consumption of the drone cluster system.

[0021] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0022] In one example of this application, Figure 1 and Figure 2 As shown, this embodiment is used to provide a method for optimizing energy consumption of collaborative regional search of a drone cluster, including the following steps.

[0023] S1: Based on the search scenario of the drone cluster, a collaborative area search system model of the drone cluster in the current time slot is constructed; the drone cluster includes an auxiliary drone and several search drones; the search scenario includes the target area, the mission cycle of the search coverage mission, the flight altitude of the search drone and the flight altitude of the auxiliary drone; the collaborative area search system model includes the flight trajectory model of the search drone and the flight trajectory model of the auxiliary drone; the auxiliary drone is used to move from the specified initial position to the final position within the mission cycle, and assist in the calculation of the unloading task of the search drone; the search drone is used to perform the search coverage mission of the target area.

[0024] S2: Based on the collaborative regional search system model of the current time slot, establish the mobile energy consumption model, local computing energy consumption model and computing offloading energy consumption model of each search drone in the current time slot, as well as the mobile energy consumption model and local computing energy consumption model of the auxiliary drone.

[0025] S3: Establish a UAV cluster collaborative area search energy consumption optimization model for the current time slot based on the mobile energy consumption model, local computing energy consumption model and computing offloading energy consumption model of each search UAV in the current time slot, as well as the mobile energy consumption model and local computing energy consumption model of the auxiliary UAV; the UAV cluster collaborative area search energy consumption optimization model includes an objective function and constraints; the objective function is a function aimed at minimizing system energy consumption; the constraints include offloading power constraints, task completion time constraints, computing resource constraints, auxiliary UAV energy constraints, task offloading ratio constraints, flight speed constraints and search area coverage constraints; the task offloading ratio is the ratio of tasks offloaded from the search UAV to the auxiliary UAV.

[0026] S4: Solve the energy consumption optimization model of the UAV cluster collaborative regional search in the current time slot to obtain the optimal energy consumption optimization plan corresponding to the current time slot; the optimal energy consumption optimization plan includes the flight trajectory of each search UAV in the current time slot, the flight trajectory of the auxiliary UAV and the task offloading ratio.

[0027] In step S1, based on the search scene of the drone cluster, a collaborative area search system model of the drone cluster in the current time slot is constructed, specifically including: rasterizing the search scene to obtain a grid map; using a time discretization method to divide the task cycle of the search coverage task into several equal time slots; for the current time slot, constructing the collaborative area search system model of the drone cluster in the current time slot according to the grid map.

[0028] Specifically, this embodiment constructs a UAV cluster collaborative regional search system model consisting of 1 search UAV and 1 auxiliary UAV. The search scene is rasterized and modeled as a rectangular map with side lengths of M and W grids, respectively, where each cell length is l. The search coverage task is completed by the collaboration of 1 search UAV and 1 auxiliary UAV for assisting the calculation task. The search scene is represented by a Cartesian coordinate system. The UAV takes off at a preset take-off position. The search UAV flies at an altitude of H0 and the auxiliary UAV flies at an altitude of H1. A The maximum flight speed of the drone is V, which remains constant during the flight. max (Unit is m / s).

[0029] The time discretization method is used to divide the search coverage task period T into N equal time slots. Assuming that δ represents the time slot length, then δ=T / N. In the nth time slot, the expression of the search UAV flight trajectory model is U i (n)=(x i (n),y i (n)); the expression of the auxiliary UAV flight trajectory model is Q A (n)=(x A (n),y A (n)).

[0030] Among them, U i (n) is the horizontal position of the i-th search drone in the current time slot, x i (n) is the lateral coordinate of the i-th search drone in the current time slot, y i (n) is the longitudinal coordinate of the i-th search UAV in the current time slot; Q A (n) is the horizontal position of the auxiliary UAV in the current time slot, x A (n) is the lateral coordinate of the auxiliary UAV in the current time slot, y A (n) is the longitudinal coordinate of the auxiliary UAV in the current time slot.

[0031] I search drones collaborate to perform an area search mission and possess information processing capabilities. During the mission cycle, local computation is permitted, with only parameter data transferred to the auxiliary drone. Alternatively, computational tasks can be offloaded to the auxiliary drone. The auxiliary drone moves from a designated initial position to a final position within the mission cycle T, assisting in the computation of the offloaded tasks from the search drone.

[0032] The specific process of step S2 is as follows: Based on the UAV cluster collaborative area search system model obtained in step S1, the UAV's mobile energy consumption model and computing energy consumption model are constructed.

[0033] The energy consumption of the UAV is derived from the resistance in the air and is determined by the speed of the UAV during flight. The power consumption of the auxiliary UAV is expressed as formula (1).

[0034]

[0035] The total energy consumption of the auxiliary UAV during flight is expressed as follows: the energy consumption model of the auxiliary UAV is expressed as formula (2).

[0036]

[0037] in, is the total energy consumption of the auxiliary UAV’s flight movement in the current time slot, δ represents the time slot length, and P A0 and P A are respectively the blade profile power and induced power of the auxiliary UAV in the hovering state, U Atip To assist the UAV’s blade tip speed, V A0 is the induced speed of the auxiliary UAV in the hovering state, ρ is the air density, A A is the rotor wheel area of ​​the auxiliary UAV, d A0 is the fuselage drag ratio of the auxiliary UAV, s A To assist the UAV’s rotor strength, P[V An ] is the mobile power of the auxiliary UAV in the nth time slot, V An is the moving speed of the auxiliary UAV in the nth time slot.

[0038] The computing energy consumption of the search drone consists of local computing energy consumption and computing offloading energy consumption. Assume that each time the search drone i flies to a cell, it has a set of tasks to perform. The task size b corresponding to the cell is i is a set of random numbers that obey Gaussian distribution (b i ∈(0,10]). The search drone i can choose to offload to the auxiliary drone UAV A The task offloading ratio a i The ratio of tasks to local computing is 1-a i .

[0039] The amount of local computing tasks for the search drone is (1-a i )b i , the drone follows a constant CPU cycle f i The calculation process of executing the task, the local calculation delay of the drone can be expressed as formula (3).

[0040]

[0041] The local computing energy consumption of the UAV can be calculated by formula (4), that is, the local computing energy consumption model of the search UAV is formula (4).

[0042]

[0043] in, is the local computing energy consumption of the i-th search drone, k i is the effective capacitance coefficient of the processor chip carried by the i-th search drone, f i is the calculation speed of the i-th search drone, t i is the local computation delay of the i-th search drone, c i is the number of CPU cycles for a 1-bit task, a i is the task offloading ratio of the i-th search drone, b i For the amount of tasks.

[0044] In the search scenario constructed in this embodiment, the UAVs all operate in an open work area when performing their missions, and the wireless communication channel between the search UAV and the auxiliary UAV is mainly line-of-sight (LoS) transmission. Based on this, the channel gain between the UAVs can be expressed as formula (5).

[0045]

[0046] Among them, h i,A [n] is the channel gain between the search UAV and the auxiliary UAV, β0 is the channel gain when the reference distance d0 = 1m, H0 is the flight height of the search UAV, U i [n] is the coordinate of the trajectory point of UAV i in the nth time slot, Q A is the coordinate of the trajectory point of the auxiliary UAV in the nth time slot. It is assumed that the channel state remains unchanged in each time slot.

[0047] To avoid channel interference between different search drones, this embodiment uses the TDMA protocol, setting up I search drones to independently and sequentially offload tasks to the auxiliary drone in the time dimension. According to Shannon's formula, the data offload rate of search drone i in time slot n can be expressed as formula (4).

[0048]

[0049] Among them, R i,A is the data unloading rate of the search UAV i in time slot n, B is the total bandwidth of the UAV communication channel, γ0 is the noise power, H A To assist the UAV to fly at a certain height, U i[n] Search for the trajectory point coordinates of UAV i for the nth time slot.

[0050] The transmission delay and task volume of the drone offloading process can also be used to express the data offloading rate. By integrating equations (5) and (6), we can obtain the data unloading power of the search drone i in the current time slot n, and thus obtain the computational unloading energy consumption of the search drone i based on the data unloading power of the search drone i in the current time slot n. That is, the computational unloading energy consumption model of the search drone is equation (8).

[0051]

[0052] Under the premise of meeting the search scene coverage, optimization is carried out from two aspects: the flight trajectory of the UAV and the computation offloading. The specific optimization variables include the flight trajectory matrix U of the search UAV i (The flight trajectory matrix of the search drone includes the flight position coordinates of all search drones in the current time slot), the flight trajectory matrix of the auxiliary drone (the flight trajectory matrix of the auxiliary drone includes the flight position coordinates of the auxiliary drone in the current time slot) Q A , and the task offloading ratio of the search drone a i The system energy consumption comes from the energy consumption of one search UAV and one auxiliary UAV. The search UAV's energy consumption is composed of flight energy consumption, local computing energy consumption, and computation offloading energy consumption; the auxiliary UAV's energy consumption is composed of flight energy consumption and local computing energy consumption. Therefore, by optimizing the above variables, the goal of minimizing system energy consumption is achieved. The optimization problem is expressed as Equations (9)-(16).

[0053]

[0054] st unloading power constraint:

[0055] Task completion time constraints:

[0056] Computing resource constraints:

[0057] Auxiliary UAV energy constraints:

[0058] Task offloading ratio constraint:

[0059] Flight speed constraint: V[n]≤V max (15).

[0060] Search area coverage constraint: G = 0 (16).

[0061] Wherein, formula (9) is the objective function in step S3 with the goal of minimizing system energy consumption, is the local computing energy consumption of the i-th search drone, Offload the computational energy consumption of the i-th search drone, is the flight movement energy consumption of the i-th search drone, To assist the local computing energy consumption of the UAV, The flight movement energy consumption of the auxiliary UAV is constrained by formula (10) on the non-negativity and maximum value of the UAV data unloading power, P i The data offloading power for the i-th search drone, P max is the maximum transmission power of the UAV node, U is the search UAV set, including all search UAVs; Formula (11) constrains the time for the UAV to complete the task in each time slot, t i is the local computation delay of the i-th search UAV, δ represents the time slot length; Formula (12) constrains the computational resources of the UAV, f i is the calculation speed of the i-th search drone, f max is the maximum computing speed of the UAV; Formula (13) constrains the energy of the auxiliary UAV, indicating that the flight and local computing energy consumption of the auxiliary UAV cannot exceed the battery power of the auxiliary UAV Formula (14) constrains the offloading ratio of tasks, a i is the task offloading ratio of the i-th search UAV; Formula (15) constrains the flight speed of the UAV, V[n] is the flight speed of the UAV, V max is the maximum flight speed of the UAV; Formula (16) constrains the search area coverage, G is the search area coverage matrix, the search area is initialized as a matrix with all elements 1, and becomes a 0 matrix after the task is completed; U represents the UAV set, that is, the search UAV set.

[0062] It should be noted that the flight energy consumption of the search drone is The calculation process of is the same as that of the flight movement energy consumption of the auxiliary UAV, that is, it is calculated through formulas (1) and (2). The difference from the calculation process of the flight movement energy consumption of the auxiliary UAV is that the flight movement energy consumption of the search UAV is substituted into formulas (1) and (2) with the relevant parameters of the search UAV, such as the blade profile power, induced power, induced speed, fuselage drag ratio, rotor strength, movement power and movement speed of the search UAV in the hovering state.

[0063] That is, search for the flight energy consumption of the drone The calculation formula is formula (17).

[0064]

[0065] in,

[0066] In the formula, P[V in ] is the mobile power of the i-th search drone in the n-th time slot, V in is the moving speed of the i-th search drone in the n-th time slot, P i0 and P i are respectively the blade profile power and induced power of the search UAV in the hovering state, U itip To search for the tip speed of the UAV, V i0 To search for the induced speed of the drone in hovering state, A i is the rotor wheel area of ​​the search UAV, d i0 is the drag ratio of the search UAV, s i To search for the rotor strength of the drone.

[0067] The calculation process of the local computing energy consumption of the auxiliary UAV is the same as that of the search UAV, that is, it is calculated by formula (4). The difference is that during the calculation, the input into formula (4) is the effective capacitance coefficient, computing speed, local computing delay, and task offloading ratio of the processor chip carried by the auxiliary UAV.

[0068] Local computing energy consumption of auxiliary drones The calculation formula is as follows.

[0069]

[0070] in,

[0071] Where k A is the effective capacitance coefficient of the processor chip carried by the auxiliary drone, f A To assist the calculation speed of the UAV, t A To assist the UAV’s local computation delay, c A is the number of CPU cycles for a 1-bit task, b A To assist the UAV's mission volume.

[0072] In step S4, the energy consumption optimization model of the UAV cluster collaborative regional search in the current time slot is solved, specifically including: using the particle swarm algorithm to solve the energy consumption optimization model of the UAV cluster collaborative regional search in the current time slot.

[0073] The solution process of the UAV cluster collaborative regional search energy consumption optimization model based on particle swarm algorithm includes the following steps.

[0074] Assume that there are Q particles in the D-dimensional target search space, that is, the population size of the particle swarm is Q; the model solution implementation process based on the particle swarm algorithm includes the following steps.

[0075] S41: Get the search area data and rasterize the search area. Initialize the PSO parameters of the drone and the speed and position of all particles. The optimal solution found by each individual is defined as P best , P best is an energy consumption optimization scheme, which includes a set of solutions for the flight trajectory of each search UAV, the flight trajectory of the auxiliary UAV, and the task offloading ratio in the current time slot. The optimal solution found by the entire community is defined as G best , G best is the optimal energy consumption optimization solution, where the qth particle can be represented by a D-dimensional vector, specifically as formula (19).

[0076] X q =(x q1 ,x q2 ,…,x qD ),q=1,2,…,Q (19).

[0077] Among them, X q represents the qth particle, x q1 The first dimension vector representing the position of the qth particle, x q2 The second-dimensional vector representing the position of the qth particle, x qD The D-th dimension vector representing the position of the q-th particle, Q is the total number of particles in the population.

[0078] S42: Construct the fitness function of the drone cluster based on the optimization problem. In each generation of evolution, calculate the fitness function value of each particle.

[0079] S43: If the particle's current fitness function value y is better than its historical optimal value P best , then P best Will be replaced by the current location.

[0080] S44: If the historical optimal value P of the particle's current position best Better than the global optimal value G best , then G best The historical optimal value P of the particle's current position will be best The velocity and position of the qth particle are updated separately. q It is expressed by formula (20).

[0081] V q =(v q1 ,v q2 ,…,v qD ),q=1,2,…,Q (20).

[0082] Among them, vq1 The first dimension vector representing the velocity of the qth particle, v q2 The second-dimensional vector representing the velocity of the qth particle, v qD The D-th dimension vector representing the velocity of the q-th particle.

[0083] The update rule for the velocity and position of the qth particle is expressed by formula (21).

[0084]

[0085] x qD =x qD-1 +v qD (twenty two).

[0086] in, is the inertia weight, which indicates the degree to which the current iteration round is affected by the speed of the previous iteration round; c1 and c2 are learning factors, rand() is a random number between (0, 1), and P bestq is the historical optimal value of the current position of the qth particle velocity.

[0087] S45: If the end condition is not met, repeat S41-S44, otherwise output G best The loop ends and the optimal position coordinates and unloading strategy for the current time slot are output, that is, the optimal energy consumption optimization plan. The optimal energy consumption optimization plan includes the flight trajectory of each search drone, the flight trajectory of the auxiliary drone and the task unloading ratio in the current time slot. The coordinates of the drone cluster are updated and the solution is entered into the next time slot.

[0088] This embodiment has the following beneficial effects.

[0089] (1) This embodiment constructs a drone swarm composed of search drones and auxiliary drones to collaboratively perform tasks. The drone swarm can flexibly adjust its scale according to task requirements, achieve efficient adaptation to different scenarios, greatly improve search efficiency, and save task time. At the same time, considering the drone's flight trajectory and task computation offloading, the optimization problem is jointly designed and solved from two aspects. In terms of flight trajectory, the drone's flight energy consumption is reduced by planning the optimal route; in terms of task computation offloading, the drone can maximize the use of its computing power while performing the task by rationally allocating computing tasks.

[0090] (2) In this embodiment, the auxiliary UAV serves as a mobile base station for offloading task calculations, which greatly eliminates the restrictions of the terrain and fixed base station conditions at the search task execution location. On the one hand, it reduces the computing burden and energy consumption of the search UAV, which not only helps to extend the UAV's flight time, but also reduces the operation and maintenance costs and improves the overall search efficiency. On the other hand, the UAV cluster can perform search tasks in any designated area, giving full play to the unique mobility of the UAV itself. With the support of the mobile base station, the UAV cluster can quickly respond and adjust the search strategy to quickly cover the target area.

[0091] (3) This embodiment adaptively adjusts the task offloading ratio between the search UAV and the auxiliary UAV based on the real-time relative position and communication status between the UAVs, and takes into account the flight capability and load of the UAVs while meeting the mission requirements, thereby further optimizing the structure of the UAV cluster, improving the cluster coordination performance, and avoiding the waste of computing resources and energy consumption caused by setting a fixed task offloading ratio.

[0092] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0093] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for optimizing energy consumption of UAV cluster collaborative area search, characterized in that: include: Based on the search scenario of the drone cluster, a collaborative regional search system model of the drone cluster in the current time slot is constructed; the drone cluster includes an auxiliary drone and several search drones; the search scenario includes the target area, the mission cycle of the search coverage mission, the flight altitude of the search drone and the flight altitude of the auxiliary drone; the collaborative regional search system model includes the flight trajectory model of the search drone and the flight trajectory model of the auxiliary drone; the auxiliary drone is used to move from the specified initial position to the final position within the mission cycle and assist in calculating the offloading task of the search drone; the search drone is used to perform the search coverage mission of the target area; According to the collaborative regional search system model of the current time slot, the mobile energy consumption model, local computing energy consumption model and computing offloading energy consumption model of each search UAV in the current time slot are established, as well as the mobile energy consumption model and local computing energy consumption model of the auxiliary UAV; An energy consumption optimization model for cooperative regional search of a UAV cluster in the current time slot is established based on the mobile energy consumption model, local computing energy consumption model and computing offloading energy consumption model of each search UAV in the current time slot, as well as the mobile energy consumption model and local computing energy consumption model of the auxiliary UAV. The energy consumption optimization model for cooperative regional search of a UAV cluster includes an objective function and constraints. The objective function is a function aimed at minimizing system energy consumption. The constraints include offloading power constraints, task completion time constraints, computing resource constraints, auxiliary UAV energy constraints, task offloading ratio constraints, flight speed constraints and search area coverage constraints. The task offloading ratio is the ratio of tasks offloaded from the search UAV to the auxiliary UAV. The energy consumption optimization model of the UAV cluster collaborative regional search in the current time slot is solved to obtain the optimal energy consumption optimization plan corresponding to the current time slot; the optimal energy consumption optimization plan includes the flight trajectory of each search UAV, the flight trajectory of the auxiliary UAV and the task offloading ratio in the current time slot.

2. The energy consumption optimization method for cooperative regional search of UAV clusters according to claim 1 is characterized in that: Based on the search scenario of drone clusters, a collaborative regional search system model of drone clusters in the current time slot is constructed, specifically including: Rasterize the search scene to obtain a grid map; The time discretization method is used to divide the task cycle of the search coverage task into several equal time slots; For the current time slot, a collaborative area search system model of the drone cluster in the current time slot is constructed according to the grid map.

3. The energy consumption optimization method for cooperative regional search of UAV clusters according to claim 2 is characterized in that: The expression for searching the UAV flight trajectory model is as follows: U i (n)=(x i (n),y i (n)); The expression of the auxiliary UAV flight trajectory model is as follows: Q A (n)=(x A ( n ), y A (n))? Among them, U i (n) is the horizontal position of the i-th search drone in the current time slot, x i (n) is the lateral coordinate of the i-th search drone in the current time slot, y i (n) is the longitudinal coordinate of the i-th search UAV in the current time slot; Q A (n) is the horizontal position of the auxiliary UAV in the current time slot, x A (n) is the lateral coordinate of the auxiliary UAV in the current time slot, y A (n) is the longitudinal coordinate of the auxiliary UAV in the current time slot.

4. The energy consumption optimization method for cooperative regional search of UAV clusters according to claim 1 is characterized in that: The expression of the mobile energy consumption model of the auxiliary UAV is as follows: in, is the total energy consumption of the auxiliary UAV’s flight movement in the current time slot, δ represents the time slot length, and P A0 and P A are respectively the blade profile power and induced power of the auxiliary UAV in the hovering state, U A tip To assist the UAV’s blade tip speed, V A0 is the induced speed of the auxiliary UAV in the hovering state, ρ is the air density, A A is the rotor wheel area of ​​the auxiliary UAV, d A0 is the fuselage drag ratio of the auxiliary UAV, s A To assist the UAV’s rotor strength, P[V An ] is the mobile power of the auxiliary UAV in the nth time slot, V An is the moving speed of the auxiliary UAV in the nth time slot.

5. The energy consumption optimization method for cooperative regional search of UAV clusters according to claim 1 is characterized in that: The expression of the local computing energy consumption model of the search drone is as follows: in, is the local computing energy consumption of the i-th search drone, k i is the effective capacitance coefficient of the processor chip carried by the i-th search drone, f i is the calculation speed of the i-th search drone, t i The delay is calculated locally for the i-th search drone, c i is the number of CPU cycles for a 1-bit task, a i is the task offloading ratio of the i-th search drone, b i For the amount of tasks.

6. The energy consumption optimization method for cooperative regional search of UAV clusters according to claim 1 is characterized in that: The objective function is as follows: Unload power constraints: Task completion time constraints: Computing resource constraints: Auxiliary UAV energy constraints: Task offloading ratio constraint: Flight speed constraint: V[n]≤V max ; Search area coverage constraint: G = 0; in, is the local computing energy consumption of the i-th search drone, Offload the computational energy consumption of the i-th search drone, is the flight movement energy consumption of the i-th search drone, To assist the local computing energy consumption of the UAV, To assist the UAV’s flight and movement energy consumption, P i The data offloading power for the i-th search drone, P max is the maximum transmission power of the UAV node, U is the search UAV set, including all search UAVs; t i The delay is calculated locally for the i-th search drone, where δ represents the time slot length and f i is the calculation speed of the i-th search drone, f max is the maximum computing speed of the UAV, To assist the drone’s battery power, a i is the task offloading ratio of the i-th search drone, V max is the maximum flight speed of the UAV, and G is the search area coverage matrix.

7. The energy consumption optimization method for cooperative regional search of UAV clusters according to claim 1 is characterized in that: Solve the energy consumption optimization model of the UAV cluster collaborative area search in the current time slot, including: The particle swarm algorithm is used to solve the energy consumption optimization model of UAV cluster collaborative area search in the current time slot.

Images