Unmanned aerial vehicle relay communication method for wind field environment and ground circular orbit

Abstract

The invention discloses an unmanned aerial vehicle relay communication method for a wind field environment and a ground circular orbit. For a full-duplex amplification forwarding fixed-wing unmanned relay communication system, according to the wind speed in a wind field environment and a horizontal wind speed angle and a vertical wind speed angle corresponding to the wind speed, an appropriate flight airspeed, an appropriate elevation angle, an appropriate flight time and an appropriate ground circular orbit radius are selected for an unmanned aerial vehicle, and then the course angle, the ground speed and the yaw angle of the unmanned aerial vehicle at each flight moment are determined; the unmanned aerial vehicle can fly along a preset ground circular orbit through adjustment of a course change angle and a roll angle of the unmanned aerial vehicle, and on the basis, an unmanned aerial vehicle relay communication method suitable for a wind field environment and the ground circular orbit is provided, so that the unmanned aerial vehicle relay can realize the unmanned aerial vehicle relay communication with the minimum energy consumption. And required data forwarding services are provided for the two ground users. Simulation experiments show that the method can meet design target requirements, and is higher in correctness and lower in energy consumption.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communications, and in particular relates to a UAV relay communication method for use in a wind farm environment and a ground circular track. Background Art

[0002] In recent years, UAV cooperative communication technology has become a hot topic in the field of wireless communication. Compared with traditional ground communication, UAV cooperative communication is easy to achieve on-demand allocation, thus having higher communication efficiency; UAVs are highly mobile, so they are more flexible and quick to deploy; the channel links between UAVs and ground terminals are mostly line-of-sight links, which can provide a better channel environment. Therefore, UAVs will play an extremely important role in the field of wireless communication in the future. Their applications mainly include: (1) as temporary base stations; (2) as mobile relays; and (3) for the Internet of Things.

[0003] At present, a large number of literatures have studied the capacity and spectrum efficiency optimization of relay communication systems when drones are used as mobile relays. In fact, drones have limited onboard energy, and energy efficiency is considered to be one of the key issues in drone communication. At present, most studies on drone relay communication consider the form of decode-and-forward relay communication, and there are relatively few studies on full-duplex amplification-and-forward relay communication assisted by fixed-wing drones. On the other hand, the wind field in the natural environment will have different effects on the airspeed, ground speed and energy consumption of drones, but most of the current research on drone-assisted communication is still based on windless conditions. In the actual wind field environment, when performing flight missions, drones not only have to counteract the additional resistance brought by airflow, but also need to adjust their flight attitude to resist the influence of wind speed and wind direction on flight. In view of this, exploring the adjustment of drone flight attitude in wind field environments and designing reasonable ground tracks are of great practical significance for improving the performance of drone relay communication. Summary of the invention

[0004] The purpose of the present invention is to provide a UAV relay communication method for a wind farm environment and a ground circular orbit, aiming to solve the optimization problem that a fixed-wing UAV relay flies along a ground circular orbit and, during the flight, acts as a full-duplex amplification and forwarding relay to forward data of a data volume Q from a ground source node to a ground destination node with minimal energy consumption.

[0005] The present invention is implemented as follows: a UAV relay communication method for a wind farm environment and a ground circular track, the method comprising the following steps:

[0006] S1. Get wind speed V w , horizontal wind speed angle χ wx and vertical wind speed angle χ wz; Initialize the number of iterations n = 1, set the roll angle φ of the drone relay for the n-1th iteration n-1 To ensure the maximum roll angle of the drone for safe flight m , that is, φ n-1 =φ m , set the error parameter ε>0 (ε is usually a decimal close to 0), set the energy consumption E of the drone in the n-1th iteration n-1 =∞;

[0007] S2. Use Lemma 4 to obtain the optimal solution V for the problem of optimizing the space velocity subproblem n , V n is the optimal solution of the UAV airspeed U obtained in the nth iteration;

[0008] S3, when the drone relays at airspeed V n and roll angle φ n-1 During flight, the product of the thrust provided by the drone engine and the airspeed is P EG (V n ,φ n-1 )<0, if If it holds, then the optimal solution to the rolling angle optimization problem is φ n =φ n-1 , otherwise, the optimal solution to the rolling angle optimization problem is in, Indicates that the drone relays at airspeed V n The upper limit of the ground speed when flying along a circular orbit on the ground, g is the acceleration due to gravity, l 12 is the distance between the ground source node and the ground destination node, jump to step S6;

[0009] S4, when P EG (V n ,φ n-1 )=0, Lemma 2 is used to obtain the optimal solution r for the problem of finding the optimal radius. n , then, r n Substitute the following formula:

[0010]

[0011] The optimal solution φ of the optimal rolling angle subproblem is calculated n =ψ n (r n )(in, and Respectively represent the airspeed V of the drone n The lower and upper limits of the ground speed corresponding to flying along the circular orbit on the ground, g is the acceleration due to gravity), jump to step S6;

[0012] S5. Use Lemma 3 to obtain the optimal solution φ to the problem of optimizing the rolling angle n ;

[0013] S6. Based on the optimal solution V of the optimized airspeed subproblem n and the optimal solution φ of the rolling angle subproblem n , use Lemma 5 to obtain the radius r of the UAV's circular orbit on the ground and the approximate energy consumption E of the UAV n ;

[0014] S7. If the absolute value of the difference in energy consumption between the nth and n-1th iterations of the UAV is greater than the error parameter, that is, |E n-1 -E n |>ε holds true, then increase the number of iterations by one, that is, set n=n+1, and jump to step S2. Otherwise, set the optimal airspeed U * is equal to the optimal solution of the optimization airspeed subproblem obtained at the nth iteration, that is, U * =V n , the optimal radius r of the UAV ground circular track * is equal to r, that is, r * = r, the optimal flight time of the drone is δ * =[Q / Q LB (r)]·|P EG (V n ,φ n )|, where Q is the amount of data that the ground source node needs to send to the ground destination node, Q LB (r) is the lower bound of the amount of data that the ground destination node can receive per unit time when the UAV relay flies along a ground circular orbit with a radius of r and provides data forwarding services. EG (V n ,φ n )|Relay for drones at airspeed V n and roll angle φ n Power consumption during flight;

[0015] S8, the optimal time δ * Discretize into K time intervals, each time interval has a length of δ * / K, calculate the elevation angle of the drone θ = arcsin (-v w cos wz / U * ), let the UAV’s flight path angle χ=0, k=1;

[0016] S9, calculate t = (k-1)δ according to the following formula * / UAV ground speed v at time K g (t), the drone airspeed U in the following formula is replaced by the optimal airspeed U * replace:

[0017]

[0018] In the formula, v g (t) is the ground speed of the UAV at time t, v w is the wind speed, χ wx and χ wz are the horizontal and vertical wind speed angles, respectively, U is the airspeed of the UAV, and χ is the flight path angle of the UAV, which ranges from 0° to 360°;

[0019] S10, calculate the yaw angle of the drone at this time At this time, the UAV's heading change angle And the roll angle of the drone at this time

[0020] S11, control the drone to a constant airspeed U * , Course change angle and roll angle φ, the flight time is δ * / K, provides data forwarding services to two ground nodes during flight;

[0021] S12. If k < K, update the UAV's flight path angle. Let k=k+1, then jump to step S9, otherwise, jump to step S13;

[0022] S13, end.

[0023] Preferably, in step S2, the sub-problem of optimizing the airspeed is specifically expressed as:

[0024]

[0025] Us min ≤U≤U max (10b)

[0026] U≥v w (10c)

[0027] Among them, V n represents the optimal solution of the UAV airspeed U obtained in the nth iteration, |P EG (U,φ n-1 )| is the UAV with airspeed U and roll angle φ n-1 Power consumption during flight, U min and U max are the minimum and maximum airspeed limits of the drone, v w is the wind speed;

[0028] Lemma 4 gives the optimal solution to the optimization problem (10) consisting of equations (10a), (10b) and (10c).

[0029] Preferably, in step S4, the rolling angle optimization subproblem can be transformed into a problem of finding an optimal radius, which is specifically expressed as:

[0030]

[0031] Among them, r n is the optimal solution of the radius r of the UAV ground circular orbit obtained in the nth iteration, ψ n (r) is the UAV moving at a constant airspeed V n The intermediate value between the minimum roll angle and the maximum roll angle corresponding to flying along a circular orbit with a radius of r. Specifically, r in formula (11) can be replaced by r n Calculated, φ n-1 is the optimal solution to the optimized rolling angle subproblem obtained at the n-1th iteration, l 12 is the ground source node T 1 With the ground destination node T 2 The distance between The UAV is moving at a constant airspeed V n Ground speed v when flying along a circular orbit on the ground g (t), g is the upper bound of gravitational acceleration, φ m The maximum roll angle for safe flight of the drone;

[0032] Lemma 2 gives the optimal solution to problem (12) consisting of equations (12a), (12b) and (12c).

[0033] Preferably, in step S5, the rolling angle optimization subproblem can be expressed as:

[0034]

[0035] sttanφ≤tanφ n-1 (13b)

[0036]

[0037] Among them, φ n represents the optimal solution of the UAV roll angle φ at the nth iteration, Q is the ground source node T 1 Need to be sent to the ground destination node T 2 The amount of data, Q LB (r) is the ground destination node T when the UAV relay flies along the ground circular orbit with a radius of r 2 The lower limit of the amount of data that can be received per unit time, |P EG (V n,φ)| is the airspeed V of the drone n and the power consumption during flight with a roll angle φ, φ n-1 is the optimal solution to the optimized rolling angle subproblem obtained at the n-1th iteration, The UAV is moving at a constant airspeed V n Ground speed v when flying along a circular orbit on the ground g (t), g is the upper bound of gravitational acceleration, l 12 is the ground source node T 1 With the ground destination node T 2 The distance between

[0038] Lemma 3 gives the optimal solution to problem (13) consisting of equations (13a), (13b) and (13c).

[0039] Preferably, in step S6, Lemma 5 gives the radius of the UAV's ground circular track and the approximate energy consumption of the UAV, including:

[0040] Condition 1: The drone relays at an airspeed of V n and roll angle φ n During flight, the product of the thrust provided by the drone engine and the airspeed is P EG (V n ,φ n ) is less than 0, that is, P EG (V n ,φ n )<0, the radius of the UAV's circular orbit is (in, The airspeed V of the drone n The upper limit of the ground speed when flying along a circular orbit on the ground, g is the acceleration due to gravity, φ n is the optimal solution of the UAV roll angle φ at the nth iteration), and the approximate energy consumption of the UAV is (Where Q is the amount of data that the ground source node needs to send to the ground destination node, Q LB (r) is the lower limit of the amount of data that can be received by the ground destination node in unit time when the UAV relay flies along the ground circular orbit with a radius of r. The airspeed V of the drone n The lower limit of the ground speed when flying along the circular orbit on the ground, g is the acceleration of gravity, r is the radius of the circular orbit of the drone on the ground, The airspeed V of the drone n and roll angle Power consumption during flight);

[0041] Condition 2: P EG (V n ,φ n)=0, the radius r of the UAV’s circular orbit can be calculated based on V n and φ n Calculate the inverse function of the function given by the following formula to obtain:

[0042]

[0043] Where ψn(rn) is the speed of the drone at a constant airspeed V n Along the radius r n The intermediate value of the minimum roll angle and the maximum roll angle corresponding to the ground circular orbit flight (here ψ n (r n )=φ n , r=r n ), and The UAV has an airspeed V n The lower and upper limits of the ground speed corresponding to flying along a circular orbit on the ground, g is the acceleration due to gravity;

[0044] Then calculate The approximate energy consumption of the drone is E n =(E n1 +E n2 ) / 2;

[0045] Condition 3: P EG (V n ,φ n )>0, the radius of the UAV's circular orbit is The approximate energy consumption of the drone is Among them, Q is the amount of data that the ground source node needs to send to the ground destination node, Q LB (r) is the lower limit of the amount of data that can be received by the ground destination node in unit time when the UAV relay flies along the ground circular orbit with a radius of r. The airspeed V of the drone n The upper limit of the ground speed when flying along the circular orbit on the ground, g is the acceleration of gravity, r is the radius of the circular orbit of the drone on the ground, The airspeed V of the drone n and roll angle Power consumption during flight.

[0046] Compared with the shortcomings and deficiencies of the prior art, the present invention has the following beneficial effects: the present invention is directed to a full-duplex amplification and forwarding fixed-wing unmanned relay communication system, and selects appropriate flight airspeed, elevation angle, flight time and ground circular orbit radius for the UAV according to the wind speed in the wind field environment and its corresponding horizontal wind speed angle and vertical wind speed angle, and then determines the course angle, ground speed and yaw angle of the UAV at each flight moment, and enables the UAV to fly along a predetermined ground circular orbit by adjusting the heading change angle and roll angle of the UAV. On this basis, a UAV relay communication method suitable for the wind field environment and the ground circular orbit is given, so that the UAV relay can provide the required data forwarding service for two ground users with minimal energy consumption. Simulation experiments show that the method of the present invention can meet the design target requirements, and has higher accuracy and lower energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic diagram of the full-duplex amplification and forwarding UAV relay communication system in the present invention;

[0048] Figure 2 The energy consumption of the UAV relay flying along the circular track on the ground and providing different data forwarding services in the wind field environment;

[0049] Figure 3 The actual amount of data received by the ground destination node when the UAV relay flies along the ground circular track and provides different data forwarding services in a wind field environment;

[0050] Figure 4 The completion time of data forwarding service when the UAV relay flies along a circular track on the ground and provides different data forwarding services in a wind field environment;

[0051] Figure 5 The ground speed change of the UAV relay completing a circle of ground orbit in a wind field environment;

[0052] Figure 6 The roll angle change of the UAV relay completing one circle of ground circular orbit in a wind field environment. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0054] The embodiment of the present invention relates to a full-duplex amplification and forwarding fixed-wing UAV relay communication system, such as Figure 1 As shown, the source node T on the ground 1 The data with a data volume of Q needs to be sent to the destination node T2 , assuming T 1 Located at the origin, its coordinates are (0,0,0), T 2 The coordinates of the 12 ,0,0),l 12 T 1 and T 2 The distance between 1 and T 2 The distance is far, and there is no direct link. It is necessary to use a fixed-wing drone flying at a height of h to relay T 3 Forwarding data; Fixed-wing UAV relay T 3 Always flying at height h, its initial position is q 0 =(l 12 / 2,r,h), the drone flies from the initial position along a circular track with a radius of r, and the drone relays T 3 The airspeed vector of the flight is U and the ground speed vector is V g Assume that the wind speed vector is V w . Drone T 3 Receive T during flight 1 While sending the signal, it sends it to the destination node T 2 Forwarding data. The drone receives signals and forwards data simultaneously on the same frequency band, that is, it works in full-duplex mode, and the relay forwarding adopts the Amplify-and-Forward (AF) relay protocol.

[0055] For ease of analysis, let the magnitude of the drone ground speed vector be v g (t) [hereinafter referred to as v g (t) is the “ground speed”], that is, v g (t)=||v g ||, where t represents time. Since the drone T 3 It flies along the circular orbit on the ground and acts as a full-duplex amplification and forwarding relay to provide data forwarding services for ground nodes (connecting the ground source node T 1 The data is forwarded to the ground destination node T 2 ), so that the local source node T 1 With drone relay T 3 And drone relay T 3 With the ground destination node T 2 When there is a line-of-sight link between the ground destination node T 2 The actual amount of received data at time t has a lower bound, specifically Q LB (r)t, where:

[0056]

[0057] In formula (1), e is a natural constant, β 0 Indicates the reference value of channel gain at a distance of 1m, P T is the ground source node T 1 and drone relay T 3 The transmission power, σ 2 is the noise power, r is the radius of the circular orbit, l 12 T 1 and T 2 The distance between them, h is the distance between the drone relays T 3 It should be noted that: Q in formula (1) LB (r) can be regarded as the ground destination node T when the UAV relay flies along the ground circular orbit with a radius of r and provides data forwarding services. 2 The lower bound of the amount of data that can be received per unit time. That is, at time t, the ground destination node T 2 The actual amount of data received will be greater than or equal to Q LB (r)t.

[0058] In fact, the ground speed vector, air speed vector and wind speed vector of the drone satisfy the wind triangle relationship, that is, V g =U+V w For the convenience of analysis, let the magnitude of the airspeed vector be U (hereinafter referred to as "airspeed"), that is, U = ||U||; let the magnitude of the wind speed vector be v w (hereinafter referred to as v w is the wind speed), that is, v w =||v w ||. Let χ wz is the vertical wind speed angle, that is, the wind speed vector V w The angle between the vertical plane and the z-axis; let x wx is the horizontal wind speed angle, that is, the wind speed vector V w The angle between the horizontal plane and the x-axis. Assume that 0°≤χ wz ≤360°、0°≤χ wx ≤360°. In order to allow the drone to relay T 3 Able to fly along a predetermined circular trajectory on the ground, the drone relays T 3 Need to adjust the elevation angle θ and yaw angle χ c (In fact, the adjustment of the yaw angle is achieved by adjusting the heading change angle, which is explained in detail below.) Here, the elevation angle θ is the angle between the airspeed vector of the drone and the x-axis in the vertical plane (the angle of attack of the drone is ignored here); the yaw angle χ c is the UAV’s route angle χ and heading change angle The difference between The UAV’s course angle χ is the angle between the UAV’s ground speed vector in the horizontal plane and the x-axis. is the angle between the airspeed vector of the drone in the horizontal plane and the x-axis. Assuming that the elevation angle θ of the drone satisfies -90°<θ<90°, then θ=arcsin(-v w cos wz / U) to calculate the elevation angle of the drone. Since the drone relay flies along the circular track on the ground at a constant airspeed U, when the airspeed, wind speed and vertical wind speed angle are determined, the elevation angle θ of the drone relay is also determined, and the range of its route angle χ is [0°, 360°]. According to the wind triangle relationship and the cosine theorem, we can get Therefore, when the airspeed U and ground speed v are known g (t) and wind speed v w , can be based on Calculate the yaw angle χ of the drone c Finally, according to Calculate the UAV's heading change angle The UAV trajectory angle χ here can be obtained from the UAV ground speed, the radius of the ground circular orbit and time. g (t) Flying along a circular orbit with a radius of r, at time t, the flight angle of the UAV is Among them, mod 360° {·} indicates the calculation modulo 360°. If time t is discretized into K intervals, the length of each time interval is t / K, and assuming that the initial course angle of the drone is 0°, that is, the course angle at time 0 is 0°, then at time t, the course angle of the drone can be calculated as

[0059] Because the drone relays T 3 When flying along a circular track on the ground at a fixed altitude, the resistance, gravity component and thrust force on the drone in the direction parallel to the fuselage should be balanced; the lift component and gravity component obtained by the drone in the vertical plane should be balanced; the centrifugal force on the horizontal plane should be balanced with the horizontal component of the lift, that is, equation (2) holds.

[0060]

[0061] In formula (2), W is the overall weight of the drone (i.e., the overall mass m of the drone multiplied by the gravitational acceleration g), θ is the elevation angle of the drone, D is the drag force on the drone, F is the thrust generated by the drone engine, L is the lift force on the drone, φ is the roll angle of the drone (the angle between the wingspan axis of the drone and the horizontal plane), m is the overall mass of the drone, and v is g(t) is the ground speed of the UAV at time t (i.e., the magnitude of the ground speed vector), and r is the radius of the circular orbit on the ground. The drag D encountered by a fixed-wing UAV flying along a circular orbit on the ground at a constant airspeed U can be calculated by formula (3):

[0062]

[0063] In formula (3), is the zero lift drag coefficient, ρ is the air density, A is the wing area, U is the airspeed of the drone (i.e., the magnitude of the airspeed vector), W is the overall weight of the drone, θ is the elevation angle of the drone, and e o is the wingspan efficiency, b is the wingspan of the UAV, and φ is the roll angle of the UAV.

[0064] For fixed-wing UAVs, the engine power consumption can be regarded as a function of the UAV airspeed U and the roll angle φ, denoted as |P EG (U,φ)|, and |P EG (U,φ)| is equal to the thrust F generated by it multiplied by the airspeed U, that is, |P EG (U,φ)|=|F|U. Here |F| represents the absolute value of the driving force F. According to formula (2), we can get F=Wsinθ+D. Then, using formula (3) and θ=arcsin(-v w cos wz / U), at this time, the power consumption of the UAV engine can be written as

[0065]

[0066] In formula (4), W is the overall weight of the drone (i.e., the overall mass m of the drone multiplied by the gravitational acceleration g), v w is the wind speed (i.e., the magnitude of the wind speed vector), χ wz is the vertical wind speed angle, is the zero lift drag coefficient, ρ is the air density, A is the wing area, U is the airspeed of the drone (i.e., the magnitude of the airspeed vector), and e o is the wingspan efficiency, b is the wingspan of the UAV, and φ is the roll angle of the UAV.

[0067] In the embodiment of the present invention, the UAV is required to fly along a predetermined circular orbit on the ground and provide data forwarding services to the ground nodes with minimum energy consumption. Therefore, according to the system requirements, the wind triangle relationship and the cosine theorem, equation (5) needs to hold:

[0068]

[0069] In formula (5), v g (t) is the ground speed of the UAV at time t (i.e., the magnitude of the ground speed vector), v wis the wind speed (i.e., the magnitude of the wind speed vector), χ wz is the vertical wind speed angle, χ is the flight path angle of the UAV, and χ wx is the horizontal wind speed angle. Since the drone flies along a circular orbit on the ground, the range of the drone's route angle χ is 0° to 360°. In a wind field environment, in order for the drone to fly along a circular orbit on the ground, positive values ​​of U and v are required. g (t), so that equation (5) holds true, that is, Need to be satisfied. Since the range of the course angle χ is 0° to 360°, cos 2 (χ-χ wx ) will take any value from 0 to 1. Therefore, in order to make the inequality Established, must ensure Lemma 1 gives the equation (5) with respect to v g (t) has positive solutions.

[0070] Introduction 1: For a fixed-wing UAV relay flying in a wind environment at a constant airspeed U, the condition for it to fly along a circular orbit on the ground is that its airspeed is greater than or equal to the wind speed, that is, U ≥ v w Established, at this time, the ground speed of the drone v g (t) can be written as:

[0071]

[0072] In formula (6), v g (t) is the ground speed of the UAV at time t, v w is the wind speed, χ wx and χ wz are the horizontal and vertical wind speed angles respectively, U is the airspeed of the UAV, and χ is the flight angle of the UAV, which ranges from 0° to 360°.

[0073] Since the value range of χ is 0° to 360°, cos 2 (χ-χ wx ) will take any value from 0 to 1. Therefore, the ground speed v given by equation (6) g (t) will satisfy in and are the lower and upper bounds of the ground speed of the UAV when it flies along a circular orbit on the ground at a constant airspeed U, which are specifically given by formula (7).

[0074]

[0075] In formula (7), U is the airspeed of the UAV, v w is the wind speed, χ wz is the vertical wind speed angle.

[0076] Next, according to the system requirements, that is, the drone relay T 3 In a wind field environment, the satellite flies along a circular orbit with a radius of r at a constant airspeed U. During the flight, it acts as a full-duplex amplification and forwarding relay, transmitting data of size Q from the ground source node T with minimal energy consumption. 1 Forwarded to the ground destination node T 2 , establish the optimization problem, which can be specifically given by equations (8a), (8b), (8c), (8d), (8e), (8f) and (8g):

[0077]

[0078] Us min ≤U≤U max (8b)

[0079] U≥v w (8c)

[0080] QQ LB (r)δ≤0 (8d)

[0081]

[0082] Problem (8), that is, the optimization problem consisting of equations (8a), (8b), (8c), (8d), (8e), (8f) and (8g), U * , δ * and r * are the optimal solutions for the UAV airspeed U, flight time δ and ground circular orbit radius r, respectively. EG (U, φ)| is the power consumption of the UAV when it flies at an airspeed U and a rolling angle φ, U min and U max are the minimum and maximum airspeed limits of the drone, v w is the wind speed, Q is the ground source node T 1 Need to be sent to the ground destination node T 2 The amount of data, Q LB (r) is the ground destination node T when the UAV relay flies along a ground circular orbit with a radius of r 2 The lower limit of the amount of data that can be received per unit time, v g (t) is the ground speed of the UAV at time t (0≤t≤δ), g is the acceleration due to gravity, r is the radius of the UAV's circular orbit, φ m To ensure the maximum roll angle of the drone for safe flight, and are the lower and upper bounds of the ground speed of the UAV when it flies along a circular orbit on the ground at a constant airspeed U, which are specifically given by formula (7). 12 is the ground source node T 1 With the ground destination node T 2 The distance between.

[0083] It should be noted that: for the convenience of mathematical analysis, when using formula (8a) to calculate the energy consumption of the UAV, it is assumed that the roll angle φ of the UAV remains constant within the flight time δ; after solving problem (8) and obtaining the optimal UAV airspeed, flight time and ground track radius, it is necessary to calculate the UAV's course angle, ground speed, yaw angle, heading change angle and roll angle at each moment, so as to adjust the UAV's flight attitude and let the UAV fly along the predetermined ground circular track.

[0084] Observing equation (8a), it is found that the energy consumption of the drone is a monotonically increasing function of the flight time δ. Therefore, the optimal solution to problem (8) will be obtained when the inequality constraint equation (8d) takes equality. In addition, the constraint equations (8e) and (8f) can be combined into

[0085] Therefore, problem (8) can be transformed into:

[0086]

[0087] Us min ≤U≤U max (9b)

[0088] U≥v w (9c)

[0089]

[0090] Among them, U * and r * are the optimal solutions for the UAV airspeed U and the ground circular orbit radius r, respectively, and Q is the ground source node T 1 Need to be sent to the ground destination node T 2 The amount of data, Q LB (r) is the ground destination node T when the UAV relay flies along the ground circular orbit with a radius of r 2 The lower limit of the amount of data that can be received per unit time, |P EG (U, φ)| is the power consumption of the UAV when it flies at an airspeed U and a rolling angle φ, U min and U max are the minimum and maximum airspeed limits of the drone, v w is the wind speed, is the upper limit of the ground speed of the UAV when it flies along the circular orbit on the ground at a constant airspeed U, which is specifically given by formula (7), g is the gravitational acceleration, r is the radius of the circular orbit of the UAV on the ground, φ m To ensure the maximum roll angle of the drone for safe flight, l 12 is the ground source node T 1 With the ground destination node T 2 The distance between.

[0091] It should be noted that: when the optimal solution U of problem (9) is obtained, that is, the problem composed of equations (9a), (9b), (9c), (9d) and (9e), * and r * After that, the optimal flight time δ * It can be directly obtained by δ * =Q / Q LB (r * ) calculated.

[0092] Next, problem (9) is decomposed into two sub-problems, namely, the airspeed optimization sub-problem and the roll angle optimization sub-problem. In the airspeed optimization sub-problem and the roll angle optimization sub-problem, only the airspeed and roll angle of the drone are optimized respectively. Let V n is the optimal solution of the optimization airspeed subproblem obtained in the nth iteration, φ n-1 is the optimal solution of the optimized roll angle subproblem obtained in the n-1th iteration, then the optimized airspeed subproblem can be written as:

[0093]

[0094] Us min ≤U≤U max (10b)

[0095] U≥v w (10c)

[0096] Among them, V n represents the optimal solution of the UAV airspeed U obtained in the nth iteration, |P EG (U,φ n-1 )| is the UAV with airspeed U and roll angle φ n-1 Power consumption during flight, U min and U max are the minimum and maximum airspeed limits of the drone, v w is the wind speed.

[0097] It should be noted that: in the first iteration, φ n-1 Set to the maximum roll angle φ to ensure safe flight of the drone m .

[0098] When problem (10) is obtained, that is, the optimal solution V of the optimization problem composed of equations (10a), (10b) and (10c) n After that, V n Instead of the UAV airspeed U in the roll angle optimization problem, that is, in the roll angle optimization problem, the UAV airspeed U is constant and equal to V n , and then optimize the roll angle of the drone. Let φ n is the optimal solution of the optimized rolling angle subproblem obtained in the nth iteration. Next, three cases are considered for the optimized rolling angle subproblem.

[0099] Case 1: P EG (V n ,φ n-1 )<0 holds true. At this time, it is impossible to reduce the power consumption and flight time of the drone by reducing the roll angle of the drone. Therefore, when When it holds, the optimal solution to the rolling angle problem is φ n =φ n-1 , otherwise, the optimal solution to the rolling angle optimization problem is

[0100] Case 2: P EG (V n ,φ n-1 )=0. At this time, the calculated power consumption of the drone is 0. As mentioned above, the ground speed v of the drone g (t) There is a lower bound and upper bound That is to say, when the UAV flies along a circular orbit with a radius of r, there is also a lower limit of the roll angle. and upper limit At this point, the problem of optimizing the rolling angle can be transformed into finding an optimal radius r n , so that ψ in equation (11) n (r n ) is equal to φ n-1 Or as close as possible to φ n-1 .

[0101]

[0102] In this way, the problem of optimizing the rolling angle can be transformed into the problem of finding the optimal radius, which can be written as:

[0103]

[0104] Problem (12), that is, the problem consisting of equations (12a), (12b) and (12c), r n is the optimal solution of the radius r of the UAV ground circular orbit obtained in the nth iteration, ψ n(r) can be replaced by r in formula (11) n Calculated, φ n-1 is the optimal solution to the optimal rolling angle subproblem obtained at the n-1th iteration (at the first iteration, φ n-1 Set to the maximum roll angle φ to ensure safe flight of the drone m ), l 12 is the ground source node T 1 With the ground destination node T 2 The distance between The UAV is moving at a constant airspeed V n Ground speed v when flying along a circular orbit on the ground g The upper bound of (t) can be specifically V n Substituting U in equation (7) to calculate, g is the acceleration due to gravity, φ m The maximum roll angle for safe flight of the drone.

[0105] The following Lemma 2 gives the optimal solution to problem (12).

[0106] Lemma 2: The optimal solution of problem (12), i.e., the problem composed of equations (12a), (12b) and (12c), can be divided into the following four cases.

[0107] Situation 2.1: When The optimal solution to problem (12) is r n = l 12 / 2.

[0108] Case 2.2: When and The optimal solution to problem (12) is n , can be used for ψ n (r)-φ n-1 =0, in The range is obtained by binary search.

[0109] Case 2.3: When and φ n-1 <ψ n (l 12 / 2) holds. The optimal solution to problem (12) is r n = l 12 / 2.

[0110] Case 2.4: When and The optimal solution to problem (12) is

[0111] Get the optimal solution r of problem (12) n After that, the optimal solution of the rolling angle subproblem is φn By changing r n Substituting into formula (11) we can get: n =ψ n (r n ).

[0112] Case 3: P EG (V n ,φ n-1 )>0 holds true. At this time, the problem of optimizing the rolling corner can be written as:

[0113]

[0114] sttanφ≤tanφ n-1 (13b)

[0115]

[0116] Among them, φ n represents the optimal solution of the UAV roll angle φ, Q is the ground source node T 1 Need to be sent to the ground destination node T 2 The amount of data, Q LB (r) is the ground destination node T when the UAV relay flies along the ground circular orbit with a radius of r 2 The lower limit of the amount of data that can be received per unit time, |P EG (V n ,φ)| is the airspeed V of the drone n and the power consumption during flight with a roll angle φ, n-1 is the optimal solution to the optimal rolling angle subproblem obtained at the n-1th iteration (at the first iteration, φ n-1 Set to the maximum roll angle φ to ensure safe flight of the drone m ), is the ground speed v when the UAV flies along the circular track on the ground at a constant airspeed Vn g The upper bound of (t) can be obtained by replacing U in equation (7) with V n Instead of calculating, g is the acceleration due to gravity, l 12 is the ground source node T 1 With the ground destination node T 2 The distance between.

[0117] The following Lemma 3 gives the optimal solution to problem (13), i.e., the problem consisting of equations (13a), (13b) and (13c).

[0118] Lemma 3: The optimal solution of problem (13), i.e., the problem composed of equations (13a), (13b) and (13c), can be divided into the following three cases.

[0119] Situation 3.1: When The optimal solution to problem (13) is

[0120] Case 3.2: When and At this time, there will be a roll angle φ that makes the value of equation (13a) equal to 0. The roll angle of the drone at this time is Can be targeted at P EG (V n ,φ)=0 In this way, the problem of optimizing the rolling angle can be transformed into finding an optimal radius r n , so that ψ in equation (11) n (r n ) is equal to Or as close as possible The optimal radius r is found n The problem can be written as:

[0121]

[0122] Among them, r n is the optimal solution of the radius r of the UAV ground circular orbit obtained in the nth iteration, ψ n (r) can be replaced by r in formula (11) n Calculated, is P obtained by dichotomy above EG (V n ,φ)=0 The solution within the range, l 12 is the ground source node T 1 With the ground destination node T 2 The distance between The UAV is moving at a constant airspeed V n Ground speed v when flying along a circular orbit on the ground g The upper bound of (t) can be specifically V n Substituting U in equation (7) to calculate, g is the acceleration due to gravity, φ n-1 is the optimal solution to the optimal rolling angle subproblem obtained at the n-1th iteration (at the first iteration, φ n-1 Set to the maximum roll angle φ to ensure safe flight of the drone m ). Problem (14), that is, the optimal solution r of the optimization problem composed of equations (14a), (14b) and (14c) n It can be obtained by Lemma 2 (at this time, φ in Lemma 2 m You need to use φ here n-1 Instead, φ in Lemma 2n-1 Need to use ). We obtain the optimal solution r of problem (14) n After that, the optimal solution of the rolling angle subproblem is φ n By changing r n Substituting into formula (11) we can get: n =ψ n (r n ).

[0123] Case 3.3: When and The optimal rolling corner problem can be written as:

[0124]

[0125] Among them, φ n represents the optimal solution of the UAV roll angle φ, Q is the ground source node T 1 Need to be sent to the ground destination node T 2 The amount of data, For drone relay along the radius The ground destination node T during ground circular orbit flight 2 The lower limit of the amount of data that can be received per unit time, |P EG (V n ,φ)| is the airspeed V of the drone n and the power consumption during flight with a roll angle φ, φ n-1 is the optimal solution to the optimal rolling angle subproblem obtained at the n-1th iteration (at the first iteration, φ n-1 Set to the maximum roll angle φ to ensure safe flight of the drone m ), The airspeed V of the drone n Ground speed v when flying along a circular orbit on the ground g The upper bound of (t) can be obtained by replacing U in equation (7) with V n Instead of calculating, g is the acceleration due to gravity, l 12 is the ground source node T 1 With the ground destination node T 2 The distance between them. Problem (15), that is, the optimal solution φ of the optimization problem composed of equations (15a) and (15b) n It can be obtained by one-dimensional search method.

[0126] The following Lemma 4 gives the optimal solution to the optimization spacespeed subproblem, i.e., the optimization problem (10) consisting of equations (10a), (10b) and (10c).

[0127] Introduction 4: When U max <v wIf U is established, equation (10b) and equation (10c) cannot be satisfied at the same time. In this case, problem (10) is not feasible. max ≥v w holds, equation (10b) and equation (10c) can be combined into max(v w ,U min )≤U≤U max , then the optimal solution V of problem (10) n It can be divided into the following eight situations.

[0128] Scenario 1: The optimal solution to problem (10) is V n =max(v w ,U min ).

[0129] Scenario 2: The optimal solution to problem (10) can be further divided into the following three sub-cases:

[0130] 1) If The optimal solution to problem (10) is V n =max(v w ,U min );

[0131] 2) If and The optimal solution to problem (10) is

[0132] 3) If The optimal solution to problem (10) is V n =U max .

[0133] Scenario 3: and V 1 ≤max(v w ,U min ) holds. The optimal solution to problem (10) can be further divided into the following three sub-cases:

[0134] 1) If P EG [max(v w ,U min ),φ n-1 ]>0 holds, the optimal solution to problem (10) is V n =max(v w ,U min );

[0135] 2) If P EG [max(v w ,U min ),φn-1 ]≤0 and P EG (U max ,φ n-1 )≥0 holds, the optimal solution V of problem (10) n YesP EG (U,φ n-1 )=0at max(v w ,U min )≤U≤U max The solution within the range can be obtained by binary search;

[0136] 3) If P EG (U max ,φ)<0 holds, the optimal solution of problem (10) is V n =U max .

[0137] Scenario 4: and V 2 ≤max(v w ,U min )<V 1 ≤U max The optimal solution to problem (10) can be further divided into the following five sub-cases:

[0138] 1) If P EG (V 1 ,φ n-1 )>0 holds, the optimal solution to problem (10) is V n =V 1 ;

[0139] 2) If P EG (V 1 ,φ n-1 )≤0 and P EG [max(v w ,U min ),φ n-1 ]≥0 holds, the optimal solution V of problem (10) n P EG (U,φ n-1 )=0at max(v w ,U max )≤U≤V 1 The solution within the range can be obtained by binary search method;

[0140] 3) If P EG [max(v w ,U min ),φ n-1 ]<0 and P EG (U max ,φ n-1)≥0 holds, the optimal solution V of problem (10) n P EG (U,φ n-1 )=0 at V 1 ≤U≤U max The solution within the range can be obtained by binary search method;

[0141] 4) If P EG [max(v w ,U min ),φ n-1 ]<0,P EG [max(v w ,U min ),φ n-1 ]<P EG (U max ,φ n-1 ) and P EG (U max ,φ n-1 )<0 holds, the optimal solution to problem (10) is V n =U max ;

[0142] 5) If P EG [max(v w ,U min ),φ n-1 ]<0,P EG [max(v w ,U min ),φ n-1 ]≥P EG (U max ,φ n-1 ) and P EG (U max ,φ n-1 )<0 holds, the optimal solution to problem (10) is V n =max(v w ,U min ).

[0143] Scenario 5: V 1 ≤U max and max(v w ,U min )<V 2 The optimal solution to problem (10) can be further divided into the following six sub-cases:

[0144] 1) If 0<P EG [max(v w ,U min ),φ n-1 ]≤P EG (V1 ,φ n-1 ) holds true, the optimal solution to problem (10) is V n =max(v w ,U min );

[0145] 2) If 0<P EG (V 1 ,φ n-1 )<P EG [max(v w ,U min ),φ n-1 ] holds, the optimal solution to problem (10) is V n =V 1 ;

[0146] 3) If P EG [max(v w ,U min ),φ n-1 ]≤0 and P EG (V 2 ,φ n-1 )≥0 holds, the optimal solution V of problem (10) n YesP EG (U,φ n-1 )=0at max(v w ,U min )≤U≤V 2 The solution within the range can be obtained by binary search method;

[0147] 4) If P EG (V 2 ,φ n-1 )<0 and P EG (U max ,φ n-1 )≥0 holds, the optimal solution V of problem (10) n YesP EG (U,φ n-1 )=0 at V 1 ≤U≤U max The solution within the range can be obtained by binary search method;

[0148] 5) If P EG (V 2 ,φ n-1 )<0、P EG (U max ,φ n-1 )<0 and P EG (V 2 ,φ n-1 )≥P EG (U max ,φn-1 ) holds true, the optimal solution to problem (10) is V n =V 2 ;

[0149] 6) If P EG (V 2 ,φ n-1 )<0、P EG (U max ,φ n-1 )<0 and P EG (V 2 ,φ n-1 )<P EG (U max ,φ n-1 ) holds true, the optimal solution to problem (10) is V n =U max .

[0150] Scenario 6: V 2 ≤max(v w ,U min ) and U max <V 1 The optimal solution to problem (10) can be further divided into the following three sub-cases:

[0151] 1) If P EG (U max ,φ n-1 )>0 holds, the optimal solution to problem (10) is V n =U max ;

[0152] 2) If P EG (U max ,φ n-1 )≤0 and P EG [max(v w ,U min ),φ n-1 ]≥0 holds, the optimal solution V of problem (10) n YesP EG (U,φ n-1 )=0at max(v w ,U min )≤U≤U max The solution within the range can be obtained by binary search method;

[0153] 3) If P EG [max(v w ,U min ),φ n-1 ]<0 holds, the optimal solution to problem (10) is V n =max(vw ,U min ).

[0154] Scenario 7: max(v w ,U min )<V 2 and V 2 <U max <V 1 The optimal solution to problem (10) can be further divided into the following five sub-cases:

[0155] 1) If P EG [max(v w ,U min ),φ n-1 ]>0、P EG [max(v w ,U min ),φ n-1 ]>P EG (U max ,φ n-1 ) and P EG (U max ,φ n-1 )≥0 holds, the optimal solution to problem (10) is V n =U max ;

[0156] 2) If P EG [max(v w ,U min ),φ n-1 ]>0、P EG [max(v w ,U min ),φ n-1 ]≤P EG (U max ,φ n-1 ) and P EG (U max ,φ n-1 )≥0 holds, the optimal solution to problem (10) is V n =max(v w ,U min );

[0157] 3) If P EG [max(v w ,U min ),φ n-1 ]≤0 and P EG (V 2 ,φ n-1 )≥0 holds, the optimal solution V of problem (10) n P EG(U,φ n-1 )=0at max(v w ,U max )≤U≤V 2 The solution can be obtained by binary search method;

[0158] 4) If P EG [max(v w ,U min ),φ n-1 ]>0 and P EG (U max ,φ n-1 )<0 holds, the optimal solution V n P EG (U,φ n-1 )=0 at V 2 ≤U≤U max The solution within the range can be obtained by binary search method;

[0159] 5) If P EG (V 2 ,φ n-1 )<0 holds, the optimal solution to problem (10) is V n =V 2 .

[0160] Scenario 8: and U max ≤V 2 The optimal solution to problem (10) can be further divided into the following three sub-cases:

[0161] 1) If P EG [max(v w ,U max ),φ n-1 ]>0, the optimal solution of problem (10) is V n =max(v w ,U max );

[0162] 2) If P EG [max(v w ,U max ),φ n-1 ]≤0 and P EG (U max ,φ n-1 )≥0 holds, the optimal solution V of problem (10) n YesP EG (U,φ n-1 )=0at max(v w ,U min )≤U≤U max The solution within the range can be obtained by binary search method;

[0163] 3) If P EG (U max ,φ n-1 )<0 holds, the optimal solution to problem (10) is V n =U max .

[0164] It should be noted that the above V 1 and V 2 Given by formula (16):

[0165]

[0166] In formula (16), η 1 and η 2 They are defined as: and Where W is the overall weight of the drone, e o is the span efficiency, is the zero lift drag coefficient, A is the wing area, ρ is the air density, b is the wingspan of the drone, φ n-1 is the optimal solution to the optimal rolling angle subproblem obtained at the n-1th iteration (at the first iteration, φ n-1 Set to the maximum roll angle φ to ensure safe flight of the drone m ), v w is the wind speed, χ wz is the vertical wind speed angle.

[0167] In addition, in Lemma 4, if we use binary search to find P EG (U,φ n-1 )=0, if multiple solutions are found, the smallest one is returned as the final solution. Because a small airspeed will produce a small ground speed, which is conducive to reducing the radius of the circular orbit on the ground, thus facilitating the UAV to forward data as a relay.

[0168] In obtaining the optimal solution V of the space velocity subproblem n and the optimal solution φ of the rolling angle subproblem n After that, the radius of the UAV’s circular orbit on the ground and the approximate energy consumption of the UAV are calculated, which is given by Lemma 5.

[0169] Lemma 5: The optimal solution V of the space velocity subproblem is known n and the optimal solution φ of the rolling angle subproblem n ,The radius of the UAV’s ground circular track and the approximate energy consumption of the UAV can be divided into the following three conditions.

[0170] Condition 1: P EG (V n ,φ n)<0 holds. The radius of the UAV’s circular orbit is The approximate energy consumption of the drone is

[0171] Condition 2: P EG (V n ,φ n )=0. The radius r of the UAV's circular orbit can be calculated based on V n and φ n Calculate the inverse function of the function given by equation (11) to obtain [ψ in equation (11) n (r n )=φ n , r n = r]; then, calculate and The approximate energy consumption of the drone is E n =(E n1 +E n2 ) / 2.

[0172] Condition 3: P EG (V n ,φ n )>0. The radius of the UAV's circular orbit is The approximate energy consumption of the drone is

[0173] According to the above discussion, the present invention discloses a UAV relay communication method for a wind farm environment and a ground circular track, which is used to solve the optimization problem (8) composed of equations (8a), (8b), (8c), (8d), (8e), (8f) and (8g), that is, the fixed-wing UAV relay flies along the ground circular track, and during the flight, as a full-duplex amplification and forwarding relay, forwards data of a data volume Q from a ground source node to a ground destination node with minimum energy consumption. The method specifically includes the following steps:

[0174] S1. Get wind speed v w , horizontal wind speed angle χ wx and vertical wind speed angle χ wz ; Initialize n = 1, φ n-1 =φ m ,ε>0,E n-1 =∞, where ε is a small positive constant, such as 0.001;

[0175] S2. Use Lemma 4 to obtain the optimal solution V for the problem of optimizing the space velocity subproblem n ;

[0176] S3, when P EG (V n ,φn-1 )<0, if If it holds, then the optimal solution to the rolling angle optimization problem is φ n =φ n-1 , otherwise, the optimal solution to the rolling angle optimization problem is Then, jump to step S6;

[0177] S4, when P EG (V n ,φ n-1 )=0, Lemma 2 is used to obtain the optimal solution r for the problem of finding the optimal radius. n , then, r n Substituting into equation (11), we can obtain the optimal solution φ of the optimal rolling angle subproblem: n =ψ n (r n ); then, jump to step S6;

[0178] S5. Use Lemma 3 to obtain the optimal solution φ to the problem of optimizing the rolling angle n ;

[0179] S6. Based on the optimal solution V of the optimized airspeed subproblem n and the optimal solution φ of the rolling angle subproblem n , use Lemma 5 to obtain the radius r of the UAV's circular orbit on the ground and the approximate energy consumption E of the UAV n ;

[0180] S7: If | E n-1 -E n |>ε holds, then let n=n+1 and jump to step S2, otherwise, let U * =V n , r * = r and δ * =[Q / Q LB (r)]·|P EG (V n ,φ n )|;

[0181] S8, the optimal time δ * Discretize into K time intervals (K is a large positive integer), the length of each time interval is δ * / K; calculate the elevation angle of the drone θ = arcsin (-v w cos wz / U * ); let the UAV’s flight path angle χ=0, k=1;

[0182] S9: Calculate t = (k-1)δ according to equation (6) * / UAV ground speed v at time Kg (t), at this time, the UAV airspeed U in formula (6) is replaced by the optimal airspeed U * replace;

[0183] S10, calculate the yaw angle of the drone at this time Calculate the heading change angle of the drone at this time Calculate the roll angle of the drone at this time

[0184] S11, the UAV is at a constant airspeed U * , Course change angle and roll angle φ, the flight time is δ * / K, provides data forwarding services to two ground nodes during flight;

[0185] S12, if k < K, then update the UAV's course angle Let k=k+1, then jump to step S9, otherwise, jump to step S13;

[0186] Step 13, end.

[0187] A simulation experiment was conducted for the above method in the embodiment of the present invention, and the experimental environment was a Matlab environment. In the experiment, the "sqp" algorithm in the Matlab optimization toolbox was used to verify the correctness of the method of the present invention. In the experimental result figure, "sqp" is used to represent the data obtained by the "sqp" algorithm in the Matlab optimization toolbox. The parameter values ​​in the experiment are shown in Table 1 below.

[0188] In addition, in order to compare with the method of the present invention, the system performance curves of three comparison schemes are also given in the simulation experiment diagram, namely, “l 12 / 2” scheme, “(U max +U min ) / 2” scheme and “φ m "Plan." 12 / 2" solution means that the drone relay always has a radius of l 12 / 2 ground circular orbit flight; "(U max +U min ) / 2” solution means that the drone relay always flies at a constant airspeed (U max +U min ) / 2 flies along a circular orbit on the ground; "φ m The solution indicates that the UAV relay will always reach the maximum roll angle φ when flying along the circular track on the ground m .

[0189] Table 1

[0190]

[0191] Attached Figure 2 , Figure 3 , Figure 4 The energy consumption, the amount of data actually received by the ground destination node, and the completion time of the data forwarding service are given when the UAV relay flies along a circular orbit on the ground and provides different data forwarding services in a wind field environment. It should be noted that in order to obtain the actual amount of data received by the ground destination node, the residual self-interference channel power is set to -60dB in the simulation experiment.

[0192] Table 2 gives the airspeed, ground circular orbit radius and elevation angle of the UAV relay for the four schemes. Figure 5 , Figure 6 The ground speed change and roll angle change of the UAV relay completing a circle of ground circular orbit in a wind field environment are given, where the value range of the route angle χ of the UAV relay is [0°, 360°].

[0193] Table 2

[0194]

[0195] Figure 2 It shows that the energy consumption of the drone increases with the increase of the Q value (the amount of data required to be forwarded). Figure 2 The approximate energy consumption of the UAV relay calculated by Lemma 5 is given in the figure (indicated by “the proposed algorithm (approximate)”). The results show that the approximate energy consumption of the UAV is very close to the actual value, which confirms the correctness of the method of the present invention. max +U min ) / 2” solution has the largest energy consumption, “l 12 / 2” and “φ m The energy consumption of the schemes is almost the same, the energy consumption of the method of the present invention is the smallest, and it can save at least 11.65% of energy consumption compared with the three comparison schemes. In addition, Figure 2 It is shown that the energy consumption of the UAV is very close to the simulation results obtained by the “sqp” algorithm in the Matlab optimization toolbox, which proves the correctness of the method of the present invention.

[0196] Figure 3 It shows that the actual amount of data received by the destination node under all four schemes is greater than the Q value (the amount of data required to be forwarded), which proves the correctness of the method of the present invention. max +U min ) / 2” scheme and “φ m "The actual amount of data received by the destination node under the scheme is very close to the completion time of the data forwarding service (e.g. Figure 3 and Figure 4 However, “l 12 / 2” scheme, the destination node actually receives the least amount of data, and the data forwarding service takes the longest time to complete (e.g. Figure 3 and Figure 4 As shown), because “l 12 The ground circular orbit radius of the / 2” scheme is the largest, which is not conducive to the UAV relay to forward data.

[0197] Figure 5 and Figure 6 It is shown that the ground speed and roll angle of the drone relay have a strong correlation, which means that the drone needs to adjust its roll angle according to the ground speed so that it can fly according to the predetermined circular orbit on the ground. Figure 6 Display, "φ m ” and “(U max +U min The roll angle variation ranges of the UAV relay of the "l ) / 2" scheme are 9.1767° to 45° and 7.0925° to 15.4598° respectively, while the method of the present invention and the "l 12 The roll angle of the UAV relay of the " / 2" scheme varies from 0.4938° to 4.5852° and 0.0772° to 0.7213° respectively. max +U min ) / 2” solution, the maximum ground speed of the drone relay (e.g. Figure 5 However, since its orbit radius is also large (as shown in Table 2), the roll angle and its variation range of the UAV are medium (as shown in Table 2). Figure 6 “φ m The UAV relay ground speed under the "scheme is medium (such as Figure 5 As shown in Table 2), the rolling angle and range of the UAV relay are the largest because its orbital radius is very small (as shown in Table 2). Figure 6 On the one hand, a large UAV roll angle will result in high energy consumption (such as Figure 2 On the other hand, in the actual wind field environment, the roll angle of the UAV relay varies greatly, requiring the UAV relay to have a fast and good roll angle adjustment capability. Therefore, from a practical point of view, the method of the present invention and "l 12 / 2” solutions are more suitable because their rolling angle and range of variation are smaller. However, “l 12 The / 2" scheme requires more data transmission time, and the amount of data transmitted is small, but the energy consumption is higher. Therefore, the method of the present invention is the best in general. In addition, Table 2 and Figure 5 It shows that a high airspeed of the drone means a high ground speed, and the elevation angle of the drone will be relatively small, and vice versa. This phenomenon matches the wind triangle, proving the correctness of the method of the present invention.

[0198] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A UAV relay communication method for wind farm environment and ground circular track, characterized in that: The method comprises the following steps: S1. Get wind speed v w , horizontal wind speed angle χ wx and vertical wind speed angle χ wz ; Initialize the number of iterations n = 1, set the roll angle φ of the drone relay in the n-1th iteration n-1 To ensure the maximum roll angle of the drone for safe flight m , set the energy consumption E of the drone in the n-1th iteration n-1 =∞; S2. Use Lemma 4 to obtain the optimal solution V for the problem of optimizing the space velocity subproblem n , V n is the optimal solution of the UAV airspeed U obtained in the nth iteration; S3, when the drone relays at airspeed V n and roll angle φ n-1 During flight, the product of the thrust provided by the drone engine and the airspeed is P EG (V n ,φ n-1 )<0, if If it holds, then the optimal solution to the rolling angle optimization problem is φ n =φ n-1 , otherwise, the optimal solution to the rolling angle optimization problem is in, Indicates that the drone relays at airspeed V n The upper limit of the ground speed when flying along a circular orbit on the ground, g is the acceleration due to gravity, l 12 is the distance between the ground source node and the ground destination node, jump to step S6; S4, when P EG (V n ,φ n-1 )=0, Lemma 2 is used to obtain the optimal solution r for the problem of finding the optimal radius. n , r n Substitute the following formula: The optimal solution φ of the optimal rolling angle subproblem is calculated n =ψ n (r n ),in, and Respectively represent the airspeed V of the drone n The lower and upper limits of the ground speed corresponding to flying along the circular orbit on the ground, g is the acceleration due to gravity, and jump to step S6; S5. Use Lemma 3 to obtain the optimal solution φ to the problem of optimizing the rolling angle n ; S6. According to the optimal solution V of the optimized airspeed subproblem n and the optimal solution φ of the rolling angle subproblem n , use Lemma 5 to obtain the radius r of the UAV's circular orbit on the ground and the approximate energy consumption E of the UAV n ; S7. If the absolute value of the difference in energy consumption between the nth and n-1th iterations of the UAV is greater than the error parameter ε, that is, |E n-1 -E n |>ε holds true, then increase the number of iterations by one, that is, set n=n+1, and jump to step S2. Otherwise, set the optimal airspeed U * is equal to the optimal solution of the optimization airspeed subproblem obtained at the nth iteration, that is, U * =V n , the optimal radius r of the UAV ground circular track * is equal to r, that is, r * = r, the optimal flight time of the drone is δ * =[Q / Q LB (r)]·|P EG (V n ,φ n )|, where Q is the amount of data that the ground source node needs to send to the ground destination node, Q LB (r) is the lower bound of the amount of data that the ground destination node can receive per unit time when the UAV relay flies along a ground circular orbit with a radius of r and provides data forwarding services. EG (V n ,φ n )|Relay for drones at airspeed V n and roll angle φ n Power consumption during flight; S8, the optimal time δ * Discretize into K time intervals, each time interval has a length of δ * / K, calculate the elevation angle of the drone θ = arcsin (-v w cos wz / U * ), let the UAV’s flight path angle χ=0, k=1; S9, calculate t = (k-1)δ according to the following formula * / UAV ground speed v at time K g (t), the drone airspeed U in the following formula is replaced by the optimal airspeed U * replace: In the formula, v g (t) is the ground speed of the UAV at time t, v w is the wind speed, χ wx and χ wz are the horizontal and vertical wind speed angles, respectively, U is the airspeed of the UAV, and χ is the flight path angle of the UAV, which ranges from 0° to 360°; S10, calculate the yaw angle of the drone at this time At this time, the UAV's heading change angle And the roll angle of the drone at this time S11, control the drone to a constant airspeed U * , Course change angle and roll angle φ, the flight time is δ * / K, provides data forwarding services to two ground nodes during flight; S12. If k < K, update the UAV's flight path angle. Let k=k+1, then jump to step S9, otherwise, jump to step S13; S13, end.

2. The method according to claim 1, characterized in that In step S2, the sub-problem of optimizing airspeed is specifically expressed as: s.t.U min ≤U≤U max (10b) U≥v w (10c) Among them, V n represents the optimal solution of the UAV airspeed U obtained in the nth iteration, |P EG (U,φ n-1 )| is the UAV with airspeed U and roll angle φ n-1 Power consumption during flight, U min and U max are the minimum and maximum airspeed limits of the drone, v w is the wind speed; Lemma 4 gives the optimal solution to the optimization problem (10) consisting of equations (10a), (10b) and (10c).

3. The method according to claim 1, characterized in that In step S4, the rolling angle optimization subproblem can be transformed into the problem of finding the optimal radius, which is specifically expressed as: Among them, r n is the optimal solution of the radius r of the UAV ground circular orbit obtained in the nth iteration, ψ n (r) is the UAV moving at a constant airspeed V n The intermediate value between the minimum roll angle and the maximum roll angle corresponding to flying along a circular orbit with a radius of r. Specifically, r in formula (11) can be replaced by r n Calculated, φ n-1 is the optimal solution to the optimized rolling angle subproblem obtained at the n-1th iteration, l 12 is the distance between the ground source node T1 and the ground destination node T2, The UAV is moving at a constant airspeed V n Ground speed v when flying along a circular orbit on the ground g (t), g is the acceleration due to gravity, φ m The maximum roll angle for safe flight of the drone; Lemma 2 gives the optimal solution to problem (12) consisting of equations (12a), (12b) and (12c).

4. The method according to claim 1, characterized in that In step S5, the optimization rolling angle subproblem can be expressed as: s.t.tanφ≤tanφ n-1 (13b) Among them, φ n represents the optimal solution of the UAV roll angle φ at the nth iteration, Q is the amount of data that the ground source node T1 needs to send to the ground destination node T2, and Q LB (r) is the lower limit of the amount of data that the ground destination node T2 can receive in unit time when the UAV relay flies along the ground circular orbit with a radius of r, |P EG (V n ,φ) is the airspeed V of the drone n and the power consumption during flight with a roll angle φ, n-1 is the optimal solution to the optimized rolling angle subproblem obtained at the n-1th iteration, The UAV is moving at a constant airspeed V n Ground speed v when flying along a circular orbit on the ground g (t), g is the upper bound of gravitational acceleration, l 12 is the distance between the ground source node T1 and the ground destination node T2; Lemma 3 gives the optimal solution to problem (13) consisting of equations (13a), (13b) and (13c).

5. The method according to claim 1, characterized in that In step S6, Lemma 5 gives the radius of the UAV's circular orbit on the ground and the approximate energy consumption of the UAV, including: Condition 1: The drone relays at an airspeed of V n and roll angle φ n During flight, the product of the thrust provided by the drone engine and the airspeed is P EG (V n ,φ n ) is less than 0, that is, P EG (V n ,φ n )<0, the radius of the UAV's circular orbit is (in, The airspeed V of the drone n The upper limit of the ground speed when flying along a circular orbit on the ground, g is the acceleration due to gravity, φ n is the optimal solution of the UAV roll angle φ at the nth iteration), and the approximate energy consumption of the UAV is (Where Q is the amount of data that the ground source node needs to send to the ground destination node, Q LB (r) is the lower limit of the amount of data that can be received by the ground destination node in unit time when the UAV relay flies along the ground circular orbit with a radius of r. The airspeed V of the drone n The lower limit of the ground speed when flying along the circular orbit on the ground, g is the acceleration of gravity, r is the radius of the circular orbit of the drone on the ground, The airspeed V of the drone n and roll angle Power consumption during flight); Condition 2: P EG (V n ,φ n )=0, the radius r of the UAV's circular orbit can be calculated based on V n and φ n Calculate the inverse function of the function given by the following formula to obtain: Where ψ n (r n ) is the UAV at a constant airspeed V n Along the radius r n The intermediate value of the minimum roll angle and the maximum roll angle corresponding to the ground circular orbit flight (here ψ n (r n )=φ n , r=r n ), and The UAV has an airspeed V n The lower and upper limits of the ground speed corresponding to flying along a circular orbit on the ground, g is the acceleration due to gravity; Then calculate and The approximate energy consumption of the drone is E n =(E n1 +E n2 ) / 2; Condition 3: P EG (V n ,φ n )>0, the radius of the UAV's circular orbit is The approximate energy consumption of the drone is Among them, Q is the amount of data that the ground source node needs to send to the ground destination node, Q LB (r) is the lower limit of the amount of data that can be received by the ground destination node in unit time when the UAV relay flies along the ground circular orbit with a radius of r. The airspeed V of the drone n The upper limit of the ground speed when flying along the circular orbit on the ground, g is the acceleration of gravity, r is the radius of the circular orbit of the drone, The airspeed V of the drone n and roll angle Power consumption during flight.

Images