Dynamic homeward voyage method of unmanned aerial vehicle

Abstract

The invention provides a dynamic homeward voyage method for an unmanned aerial vehicle. The method comprises the following steps: acquiring position information of a landing point and position information of a homeward voyage point; the position information of the return point is used as the coordinate origin (0, 0, 0) of a space Cartesian coordinate system, the coordinate of the landing point is (xend, yend, zend) under the space Cartesian coordinate system, and xend is greater than 0, yend is greater than 0, and zend is greater than 0; the real-time wind speed vws is obtained, and the components vwsX, vwsY and vwsZ of the real-time wind speed vws on the X axis, the Y axis and the Z axis are determined; the whole process of returning the unmanned aerial vehicle from the return point to the landing point is divided into N time periods with the time interval being T, and in each time interval, according to the channel gain between the current position of the unmanned aerial vehicle and the return point, the optimal airspeed components of the unmanned aerial vehicle on the X axis, the Y axis and the Z axis in the current time interval are determined by using a game system; and the unmanned aerial vehicle returns from the return point to the landing point according to the corresponding optimal airspeed component in each time interval.

Description

Technical Field

[0001] The present invention belongs to the technical field of UAV return technology in wireless communication, and specifically relates to a dynamic UAV return method. Background Art

[0002] In the context of wind speed influence, the UAV return technology is particularly important. Wind speed is one of the key factors affecting the flight stability and safety of UAVs. During the process of UAVs performing flight tasks, if unexpected situations occur or communication with the ground base station is lost, and at the same time affected by wind speed, the flight state of the UAV may become more complex and dangerous. The UAV autonomous return technology can automatically make a decision to return when the UAV encounters unexpected situations, and plan a safe return path according to the current meteorological information such as wind speed and wind direction. This technology not only relies on positioning technologies such as GPS and Beidou, but also needs to combine advanced technologies such as wind speed measurement, flight control system and autonomous navigation algorithm to ensure that the UAV can return safely under complex meteorological conditions.

[0003] Therefore, under complex meteorological conditions such as wind speed, the UAV autonomous return technology is of great significance for ensuring the safety of UAVs and improving the reliability and efficiency of flight tasks. Summary of the Invention

[0004] Object of the Invention: To solve the problem of the UAV returning from the return point to the landing point where the landing point is located and performing effective communication, the present invention proposes a dynamic UAV return method.

[0005] Technical Solution: A dynamic UAV return method includes the following steps:

[0006] Step 1: Obtain the position information of the landing point and the position information of the return point;

[0007] Step 2: Use the position information of the return point as the coordinate origin (0, 0, 0) of the spatial Cartesian coordinate system. In this spatial Cartesian coordinate system, the coordinates of the landing point are (x end , y end , z end ), and x end > 0, y end > 0, z end > 0;

[0008] Step 3: Obtain the real-time wind speed v ws , and determine the components v ws of the real-time wind speed v wsx , v wsy and v wsz on the X-axis, Y-axis and Z-axis;

[0009] Step 4: Divide the whole process of the UAV returning from the return point to the landing point into N time intervals with a time interval of T. In each time interval, according to the channel gain between the current position of the UAV and the return point, use the game system to determine the optimal airspeed components of the UAV on the X-axis, Y-axis, and Z-axis in the current time interval.

[0010] Step 5: The UAV returns from the return point to the landing point according to its corresponding optimal airspeed components in each time interval.

[0011] Furthermore, for determining the components v ws of the real-time wind speed v wsx on the X-axis, Y-axis, and Z-axis, the specific operations include: wsy and v wsz

[0012] Assume that the clockwise angle between the positive Z-axis and the real-time wind speed v ws is α, then the component of the wind speed on the Z-axis is expressed as:

[0013]

[0014] where, when 0° ≤ α < 90° or 270° ≤ α < 360°, v wsz is along the positive Z-axis direction; when 90° ≤ α < 180° or 180° ≤ α < 270°, v wsz is along the negative Z-axis direction.

[0015] The projection of the real-time wind speed v ws on the X-Y plane is expressed as:

[0016]

[0017] Define v wsh as the horizontal wind speed.

[0018] Assume that the clockwise angle between the positive Y-axis and the horizontal wind speed v wsh is β, then the component of the real-time wind speed v ws on the X-axis is expressed as:

[0019]

[0020] where, when 0° ≤ β < 180°, v wsx is along the positive X-axis direction; when 180° ≤ β < 360°, v wsx is along the negative X-axis direction.

[0021] The component of the real-time wind speed v ws on the Y-axis is expressed as:

[0022]

[0023] Among them, when 0° ≤ β < 90° or 270° ≤ β < 360°, v wsy is in the positive Y-axis direction; when 90° ≤ β < 180°, v wsy is in the negative Y-axis direction.

[0024] Furthermore, in step 4, the channel gain between the current position of the UAV and the return point is calculated according to the following steps:

[0025] Divide the time interval T into K small time intervals Δt, expressed as T = KΔt, define k = {1,..., k,..., K} as the set of small time intervals, and within each small time interval Δt, the channel gain of the UAV to the remote sensing at the landing point remains unchanged;

[0026] Assume that within the nth time interval T, the airspeed of the UAV is v AS,n =(v ASx,n , v ASy,n , v ASz,n ), and the actual flight speed of the UAV is expressed as:

[0027] v UAVx,n = v ASx,n ± v wsx

[0028] v UAVy,n = v ASy,n ± v wsy

[0029] v UAVz,n = v ASz,n ± v wsz

[0030] Among them, v ASx,n , v ASy,n and v ASz,n are the airspeed components of the UAV's airspeed on the X-axis, Y-axis, and Z-axis within the nth time interval T respectively, and v UAVx,n , v UAVy,n and v UAVz,n are the components of the UAV's actual flight speed on the X-axis, Y-axis, and Z-axis within the nth time interval T respectively;

[0031] When the kth small time interval Δt within the nth time interval T ends, the UAV position information is:

[0032] x n,k = x n,k-1 + v UAVx,n × Δt

[0033] y n,k= y n,k-1 + v UAVy,n ×Δt

[0034] z n,k = z n,k-1 + v UAVz,n ×Δt

[0035] where x n,0 = x n-1,K , y n,0 = y n-1,K and z n,0 = z n-1,K , and x 1,0 = 0, y 1,0 = 0 and z 1,0 = 0;

[0036] Therefore, when the k-th small time interval Δt within the n-th time interval T ends, the channel gain between the UAV and the landing point is expressed as:

[0037]

[0038] where L n,k = (x n,k , y n,k , z n,k ) represents the position coordinates of the UAV when the k-th small time interval Δt within the n-th time interval T ends; ρ0 is the unit channel gain value; and are respectively the probabilities of the line-of-sight LoS and non-line-of-sight NLoS of the additional path loss between the UAV and the landing point, and there is Ω LoS and Ω NLoS respectively represent the additional path losses of the LoS and NLoS links, and satisfy Ω LoS << Ω NLoS .

[0039] Furthermore, in step 4, in each time interval, according to the channel gain between the current position of the UAV and the return point, using the game system, the optimal airspeed components of the UAV on the X-axis, Y-axis, and Z-axis in the current time interval are determined. The specific operations include:

[0040] The game mechanism models the resource game behavior of the components of the UAV's airspeed on the X-axis, Y-axis, and Z-axis, including:

[0041] In any time interval, the X-axis, Y-axis, and Z-axis in the spatial Cartesian coordinate system obtain benefits by purchasing limited airspeed resources as followers in the game, while the leader is the one who sells the airspeed components on the X-axis, Y-axis, and Z-axis to obtain benefits;

[0042] The optimal unit - rate pricing γ is obtained by maximizing the utility functions that represent the respective interests of the leader and the competing followers. * ;

[0043] The followers determine their optimal airspeed components purchased on the X - axis, Y - axis, and Z - axis according to the optimal unit - rate pricing γ. * , deciding their optimal airspeed components purchased on the X - axis, Y - axis, and Z - axis.

[0044] Furthermore, the utility function of the leader is defined as the revenue value obtained from selling the airspeed of the UAV on the X - axis, Y - axis, and Z - axis, expressed as:

[0045] γv ASx,n +γv ASy,n +γv ASz,n

[0046] In the formula, γ represents the unit - rate pricing, and v ASx,n , v ASy,n and v ASz,n are the airspeed components of the UAV airspeed on the X - axis, Y - axis, and Z - axis respectively within the nth time interval T;

[0047] The corresponding constraint conditions are expressed as:

[0048]

[0049]

[0050] In the formula, v AS is the maximum value of the UAV airspeed, p is the transmission power of the UAV, σ 2 is the receiver noise at the landing point, B is the communication bandwidth, H n,k is the channel gain between the UAV and the landing point at the end of the kth small time interval Δt within the nth time interval T of the UAV, is the minimum demand value.

[0051] Furthermore, for the X - axis, its utility function is expressed as:

[0052]

[0053] Among them, γ represents the unit - rate pricing, p is the transmission power of the UAV, σ 2 is the receiver noise at the landing point, B is the communication bandwidth, H n,k,x is the channel gain between the UAV at the point (x n,k-1 +v UAVx,n ×Δt,y n,k-1 ,z n,k-1 ) and the landing point;

[0054] The corresponding constraint conditions are expressed as:

[0055] 0 ≤ v UAVx,n T ≤ x end -x n,0

[0056] In the formula, x end and x n,0 respectively represent the coordinate of the landing point on the X-axis and the position coordinate of the UAV on the X-axis at the initial moment of the nth time interval T.

[0057] Furthermore, for the Y-axis, its utility function is expressed as:

[0058]

[0059] where γ represents the unit rate pricing, p is the transmission power of the UAV, σ 2 is the receiver noise at the landing point, B is the communication bandwidth, and H n,k,y is the channel gain between the UAV at the point (x n,k-1 , y n,k-1 + v UAVy,n × Δt, z n,k-1 ) and the landing point;

[0060] The corresponding constraint conditions are expressed as:

[0061] 0 ≤ v UAVy,n T ≤ y end -y n,0 ;

[0062] In the formula, y end and y n,0 respectively represent the coordinate of the landing point on the Y-axis and the position coordinate of the UAV on the Y-axis at the initial moment of the nth time interval T.

[0063] Furthermore, for the Z-axis, its utility function is expressed as:

[0064]

[0065] where γ represents the unit rate pricing, p is the transmission power of the UAV, σ 2 is the receiver noise at the landing point, B is the communication bandwidth, and H n,k,z is the channel gain between the UAV at the point (x n,k-1 , y n,k-1 , z n,k-1 + v UAVz,n × Δt) and the landing point;

[0066] The corresponding constraint conditions are expressed as:

[0067] 0 ≤ v UAVz,n T ≤ z end -z n,0

[0068] In the formula, z end and z n,0 respectively represent the coordinate of the landing point on the Z-axis and the position coordinate of the UAV on the Z-axis at the initial moment of the nth time interval T.

[0069] If, during a certain game, the airspeed component does not need to be purchased for the X-axis, Y-axis, or Z-axis, then that axis is deleted from the game;

[0070] If there is only one axis in the game for the game of airspeed resources, then all of the maximum airspeed of the UAV is assigned to that axis;

[0071] If there is no airspeed component to be purchased for the X-axis, Y-axis, and Z-axis, it indicates that the UAV has reached the landing point.

[0072] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0073] (1) The method of the present invention solves the problem of the UAV returning from the return point to the landing point in a static wind speed and wind direction environment and performing effective communication;

[0074] (2) The method of the present invention can give a dynamic return method of the UAV in a dynamic wind speed and wind direction scenario through online calculation, physically conforms to the realistic application scenario, and will be effectively applied to engineering practice. Description of the Drawings

[0075] Figure 1 It is a schematic diagram of the spatial Dillka coordinates of a dynamic return of a UAV proposed by the present invention;

[0076] Figure 2 It is a schematic diagram of a dynamic return method of a UAV proposed by the present invention. Detailed Embodiments

[0077] The technical solution of the present invention will be further elaborated below in conjunction with the drawings and embodiments.

[0078] Embodiment 1:

[0079] As Figure 2 shown, this embodiment proposes a dynamic return method of a UAV in a static wind speed and wind direction scenario, including the following steps:

[0080] Step 1: Before the UAV takes off, record the position information of the landing point B;

[0081] Step 2: When the UAV located at point A receives the user's return instruction or triggers the system automatic return instruction due to less remaining energy, etc., record the position information at the return point A;

[0082] Step 3: According to the position information of the return point A and the landing point B, establish a spatial Cartesian coordinate system as shown in Figure 1 . Take the return point A as the origin of the spatial coordinates (0, 0, 0), and the coordinates of the landing point B are (x end , y end , z end ), where x end > 0, y end > 0, z end > 0, that is, the landing point B is located in the first octant of the spatial Cartesian coordinate system.

[0083] Step 4: Obtain the real-time wind speed v ws , and determine the components of this real-time wind speed v ws on the X-axis, Y-axis, and Z-axis as v wsx , v wsy , and v wsz .

[0084] Considering the scenario of static wind speed and direction, assume that the clockwise angle between the positive Z-axis and the real-time wind speed v ws is α. Then the component of the wind speed on the Z-axis can be expressed as:

[0085]

[0086] where, when 0° ≤ α < 90° or 270° ≤ α < 360°, v wsz is along the positive Z-axis direction; when 90° ≤ α < 180° or 180° ≤ α < 270°, v wsz is along the negative Z-axis direction.

[0087] The projection of the real-time wind speed v ws on the X-Y plane can be expressed as:

[0088]

[0089] Define v wsh as the horizontal wind speed.

[0090] Assume that the clockwise angle between the positive Y-axis and the horizontal wind speed v wsh is β. Then the component of the real-time wind speed v ws on the X-axis can be expressed as:

[0091]

[0092] where, when 0° ≤ β < 180°, v wsx is along the positive X-axis direction; when 180° ≤ β < 360°, v wsx is along the negative X-axis direction.

[0093] The real-time wind speed vws The component on the Y-axis can be expressed as:

[0094]

[0095] where, when 0° ≤ β < 90° or 270° ≤ β < 360°, v wsy is along the positive Y-axis direction; when 90° ≤ β < 180°, v wsy is along the negative Y-axis direction.

[0096] In the scenario of static wind speed and direction, the wind speed v ws and both α and β are known values. Therefore, the components of the wind speed along the X-axis, Y-axis, and Z-axis, v wsx , v wsy and v wsz can be obtained through the above formulas.

[0097] Step 5: During the process of the UAV returning from the return point A to the landing point B, the entire process is divided into N time periods with a time interval of T. In each time period, the UAV needs to make a decision on the appropriate UAV airspeed (v ASx,n , v ASy,n , v ASz,n ) in the three-dimensional coordinate system, so as to achieve a safe return.

[0098] In the present invention, it is assumed that the static wind speed is less than or equal to the maximum wind speed that the UAV can tolerate, that is, there will be no situation where the component of the UAV on any axis moves away from the landing point B. If the static wind speed is greater than the maximum wind speed that the UAV can tolerate, the UAV cannot return, which does not fall within the technical scope discussed in the present invention.

[0099] Since the UAV moves during the time interval T, the channel gain of the UAV to the landing point B for controlling remote sensing changes. For the convenience of explaining the data transmission problem in a time interval T, T is further divided into K small time intervals Δt, that is, T = KΔt, and the is defined as the set of small time intervals. In each small time interval Δt, since the UAV runs a short distance, it can be assumed that the channel gain of the UAV to the landing point B for controlling remote sensing remains unchanged during this period.

[0100] The path loss of the wireless transmission of the UAV to the return point for controlling remote sensing depends on the distance-based fading and the line-of-sight (LoS) probability-based fading. The LoS occurrence probability of the additional path loss between nodes p and q is first given as follows, that is:

[0101]

[0102] Among them, c0 and d0 depend on environmental variables, which are constants; is the elevation angle of the air-to-ground wireless transmission link, h is the vertical height of the UAV relative to the user, is the horizontal distance between nodes p and q.

[0103] Therefore, the channel gain between nodes p and q is defined as:

[0104]

[0105] Among them, ρ0 is the unit channel gain value (known); and are the probabilities of occurrence of line-of-sight LoS and non-line-of-sight NLoS (Non Line of Sight) of the additional path loss between nodes p and q (known), and there is Ω LoS and Ω NLoS represent the additional path losses of the LoS and NLoS links respectively (known), and satisfy Ω LoS << Ω NLoS ; d p,q is the distance between nodes p and q.

[0106] Assume that within the nth time interval T, the airspeed of the UAV is v AS,n =(v ASx,n , v ASy,n , v ASz,n ). Therefore, the speed of the UAV relative to the ground, that is, the actual flight speed of the UAV can be expressed as:

[0107] v UAVx,n = v ASx,n ± v wsx

[0108] v UAVy,n = v ASy,n ± v wsy

[0109] v UAVz,n = v ASz,n ± v wsz

[0110] Among them, v ASx,n 、v ASy,n and v ASz,n are the components of the airspeed of the UAV on the X-axis, Y-axis and Z-axis within the nth time interval T respectively, v UAVx,n 、v UAVy,n and v UAVz,nThey are the components of the actual flight speed of the drone on the X-axis, Y-axis, and Z-axis within the nth time interval T. It should be noted that the addition or subtraction of the airspeed and the wind speed depends on the positive or negative values of the components of the wind speed on the X-axis, Y-axis, and Z-axis. For example, if v wsx is in the positive X-axis direction, then v UAVx,n = v ASx,n + v wsx , otherwise v UAVx,n = v ASx,n - v wsx , and so on for v UAVy,n and v UAVz,n .

[0111] Therefore, the position information of the drone at the end of the kth small time interval Δt within the nth time interval T is:

[0112] x n,k = x n,k-1 + v UAVx,n ×Δt

[0113] y n,k = y n,k-1 + v UAVy,n ×Δt

[0114] z n,k = z n,k-1 + v UAVz,n ×Δt

[0115] where x n,0 = x n-1,K , y n,0 = y n-1,K and z n,0 = z n-1,K , and x 1,0 = 0, y 1,0 = 0 and z 1,0 = 0.

[0116] Therefore, at the end of the kth small time interval Δt within the nth time interval T, the channel gain between the drone and point B is expressed as:

[0117]

[0118] where L n,k = (x n,k , y n,k , z n,k ) represents the position coordinates of the drone at the end of the kth small time interval Δt within the nth time interval T.

[0119] Step 6: In order to enable the drone to adjust its flight direction in a timely manner according to the channel gain, it is necessary to make a decision on the airspeed of the drone in each time interval T. Since the magnitude of the drone's airspeed is constant, there is a competitive relationship among the velocity components on the X-axis, Y-axis, and Z-axis.

[0120] Therefore, in each time interval T, a game mechanism is used to model the resource game behavior of the airspeed components of the drone on the X-axis, Y-axis, and Z-axis, and the optimal airspeed components of the drone on each axis are obtained. Let n = n + 1 and repeat Step 6.

[0121] Among them, the modeling of the resource game behavior of the airspeed components of the drone on the X-axis, Y-axis, and Z-axis through the game mechanism mentioned above means that the problem of airspeed allocation is modeled as an economic game model within a time interval T.

[0122] In any time interval T of the game model, the X-axis, Y-axis, and Z-axis in the spatial Cartesian coordinate system obtain benefits by purchasing limited airspeed resources as followers in the game.

[0123] This game model includes a leader and followers.

[0124] The utility function of the leader is defined as the revenue value obtained from selling the airspeed of the drone on the X-axis, Y-axis, and Z-axis.

[0125] The utility function of the follower on the X-axis is defined as the difference between the average data transmission rate of the drone when the airspeed component obtained on the X-axis is v in the nth time interval T and the cost function paid by the drone for obtaining the airspeed component v on the X-axis, which is expressed as: ASx,n When, and the cost function paid by the drone for obtaining the airspeed component v on the X-axis ASx,n The difference is expressed as:

[0126]

[0127] Among them, p is the transmission power of the drone (a fixed value and known), σ 2 is the receiver noise at the landing point, B is the communication bandwidth, H n,k,x is the channel gain between the drone at the point (x n,k-1 + v UAVx,n ×Δt, y n,k-1 , z n,k-1 ) (only the position on the X-axis changes) and the landing point, which is:

[0128]

[0129] Then it is expressed as when the rate component obtained on the X-axis is v in the nth time interval T ASx,nThe average data transmission rate of the UAV at a certain time is its revenue; γ is the price per unit rate, γv ASx,n represents the value of the rate v obtained on the X-axis for the UAV to pay ASx,n and the cost function incurred.

[0130] The constraint condition is the effective range of the displacement of the UAV in the X-axis direction within the nth time interval T, expressed as:

[0131] 0 ≤ v UAVx,n T ≤ x end -x n,0

[0132] The utility function of the follower on the Y-axis is defined as the difference between the average data transmission rate of the UAV when the airspeed component obtained on the Y-axis is v within the nth time interval T ASy,n and the cost function incurred by the UAV to pay for the airspeed component v obtained on the Y-axis ASy,n and is expressed as:

[0133]

[0134] where H n,k,y is the channel gain between the point where the UAV is located at (x n,k-1 , y n,k-1 + v UAVy,n ×Δt, z n,k-1 )(only the position on the Y-axis changes) and the landing point, and is:

[0135]

[0136] It is then expressed as the average data transmission rate of the UAV when the rate component obtained on the Y-axis is v within the nth time interval T ASy,n which is its revenue; γv ASy,n represents the value of the rate v obtained on the Y-axis for the UAV to pay ASy,n and the cost function incurred.

[0137] The constraint condition is the effective range of the displacement of the UAV in the Y-axis direction within the nth time interval T, expressed as:

[0138] 0 ≤ v UAVy,n T ≤ y end -y n,0

[0139] The utility function of the follower on the Z-axis is defined as the difference between the average data transmission rate of the UAV when the airspeed component obtained on the Z-axis is v within the nth time interval T ASz,n and the cost function incurred by the UAV to pay for the airspeed component v obtained on the Z-axis ASz,n and is expressed as:

[0140]

[0141] where H n,k,z is the channel gain between the UAV at the point (x n,k-1 , y n,k-1 , z n,k-1 + v UAVz,n ×Δt) (only the position changes in the Z - axis) and the landing point, and is

[0142]

[0143] Then it represents the average data transmission rate of the UAV when the rate component obtained in the Z - axis is v ASz,n during the nth time interval T, which is its revenue; γv ASz,n Then it represents the cost function that the UAV pays for obtaining the rate value v ASz,n in the Z - axis.

[0144] The constraint condition is the effective range of the displacement of the UAV in the Z - axis direction during the nth time interval T, which is expressed as:

[0145] 0 ≤ v UAVz,n T ≤ z end - z n,0

[0146] Based on the above analysis, the three - dimensional rate allocation problem of the UAV during the nth time interval T is:

[0147] The optimization problem on the X - axis component is expressed as:

[0148] OP1:

[0149] s.t. 0 ≤ v UAVx,n T ≤ x end - x n,0

[0150] The optimization problem on the Y - axis component is expressed as:

[0151] OP2:

[0152] s.t. 0 ≤ v UAVy,n T ≤ y end - y n,0

[0153] The optimization problem on the Z - axis component is expressed as:

[0154] OP3:

[0155] s.t. 0 ≤ vUAVz,n T ≤ z end -z n,0

[0156] Due to the limited airspeed value of the UAV, the optimization problem at the leader level in the game within the nth time interval T is expressed as:

[0157] OP4:

[0158]

[0159] where v AS is the maximum value of the UAV's airspeed. Therefore, the first constraint condition is that the total airspeed value obtained by the UAV within the nth time interval T should be less than or equal to its maximum value; the second constraint condition is that the average data transmission rate back to the landing point by the UAV within the nth time interval T should be greater than or equal to its minimum required value

[0160] OP1, OP2, and OP3 constitute the optimization problem of the followers in the game. OP4 is the optimization problem of the leader in the game. OP1 - OP4 together constitute the Stackelberg game. Through the game actions of the leader and the followers in the game according to certain rules, the final Stackelberg equilibrium can be obtained, that is, the optimal unit resource pricing γ is obtained by maximizing the utility functions representing the respective interests of the leader and the competing followers * . Then, the followers' optimal airspeed components of their purchase rates on the X-axis, Y-axis, and Z-axis are determined according to γ * and

[0161] If, during a certain game, the X-axis, Y-axis, or Z-axis does not need to purchase airspeed resources, then this axis is removed from the game

[0162] If there is only one axis in the game for the airspeed resource game, then all of v AS is given to this axis

[0163] When there is no need for the X-axis, Y-axis, and Z-axis to purchase airspeed resources, it indicates that the UAV's X-axis, Y-axis, and Z-axis have reached the landing point B, that is, x N,K = x end , y N,K = y end and z N,K = z end , then the algorithm ends

[0164] It should be noted that the above game is carried out within each time interval T

[0165] ​If the constraint condition in OP1 becomes 0 ≤ v UVAx,n T ≤ 0 during a certain game, it indicates that the UAV has reached the landing point B in the X-axis component, and the game behavior in the X-axis direction ends in the subsequent game.

[0166] If the constraint condition in OP2 becomes 0 ≤ v UVAy,n T ≤ 0 during a certain game, it indicates that the UAV has reached the landing point B in the Y-axis component, and the game behavior in the Y-axis direction ends in the subsequent game.

[0167] If the constraint condition in OP3 becomes 0 ≤ v UVAz,n T ≤ 0 during a certain game, it indicates that the UAV has reached the landing point B in the Z-axis component, and the game behavior in the Z-axis direction ends in the subsequent game.

[0168] The method of this embodiment can give all the optimal airspeed components of the UAV during the entire return flight after the return flight command of the UAV is issued and without the need for online real-time calculation, and only offline calculation is required. It can also give the corresponding optimal airspeed components before the start of a time interval T and and the optimal airspeed components for the subsequent time intervals need to be calculated online in real time.

[0169] The method proposed in this embodiment can not only be applied to the scenarios of static wind speed and wind direction, that is, the wind speed magnitude is constant and the direction is constant (the considered wind speed has components in all three spatial dimensions), but also be converted to the scenarios of dynamic wind speed and wind direction. Assuming that the wind speed and wind direction remain unchanged within each time interval T (the scenario of rapidly changing wind can be approximated by reducing T), it is only necessary to solve the values of the current wind speed on the X-axis, Y-axis, and Z-axis before each game in the time interval T, and then conduct the game of the UAV airspeed according to the real-time wind speed value.

Claims

1. A method for dynamic return of an unmanned aerial vehicle, characterized in that: Including the following steps: Step 1: Obtain the position information of the landing point and the position information of the return point; Step 2: Take the position information of the return point as the coordinate origin (0, 0, 0) of the spatial Cartesian coordinate system. In this spatial Cartesian coordinate system, the coordinates of the landing point are (x end , y end , z end ), and x end > 0, y end > 0, z end > 0; Step 3: Obtain the real-time wind speed v ws , and determine the components of the real-time wind speed v ws on the X-axis, Y-axis, and Z-axis, namely v wsx , v wsy , and v wsz ; Step 4: Divide the whole process of the UAV returning from the return point to the landing point into N time intervals with a time interval of T. In each time interval, according to the channel gain between the current position of the UAV and the return point, use the game system to determine the optimal airspeed components of the UAV on the X-axis, Y-axis, and Z-axis in the current time interval. Step 5: The drone returns from the return point to the landing point at the corresponding optimal airspeed component in each time interval.

2. The method for dynamic return of an unmanned aerial vehicle according to claim 1, wherein: The determination of the real-time wind speed v ws Components v wsx 、v wsy and v wsz on the X-axis, Y-axis, and Z-axis, and the specific operations include: Assume that the clockwise angle between the positive Z-axis and the real-time wind speed v ws is α, then the component of the wind speed on the Z-axis is expressed as: Wherein, when 0° ≤ α < 90° or 270° ≤ α < 360°, v wsz is in the positive Z-axis direction; when 90° ≤ α < 180° or 180° ≤ α < 270°, v wsz is in the negative Z-axis direction; Real-time wind speed v ws The projection on the X-Y plane is represented as: Define v wsh as the horizontal wind speed; Assume that the clockwise angle between the positive Y-axis and the horizontal wind speed v wsh is β, then the component of the real-time wind speed v ws on the X-axis is expressed as: where, when 0° ≤ β < 180°, v wsx is along the positive X-axis direction; when 180° ≤ β < 360°, v wsx is along the negative X-axis direction; Real-time wind speed v ws The component on the Y-axis is expressed as: Wherein, when 0° ≤ β < 90° or 270° ≤ β < 360°, v wsy is in the positive Y-axis direction; when 90° ≤ β < 180°, v wsy is in the negative Y-axis direction.

3. A dynamic return method for an unmanned aerial vehicle according to claim 2, characterized in that: In Step 4, the channel gain between the current position of the drone and the return point is calculated according to the following steps: The time interval T is divided into K small time intervals Δt, expressed as T = KΔt, and it is defined that is the set of small time intervals. Within each small time interval Δt, the channel gain of the UAV to the remote sensing at the landing point remains unchanged; Assume that within the nth time interval T, the airspeed of the drone is v AS,n =(v ASx,n , v ASy,n , v ASz,n ), and the actual flight speed of the drone is expressed as: v UAVx,n = v ASx,n ± v wsx v UAVy,n = v ASy,n ± v wsy v UAVz,n = v ASz,n ± v wsz where v ASx,n , v ASy,n and v ASz,n are the airspeed components of the unmanned aerial vehicle on the X-axis, Y-axis and Z-axis within the nth time interval T respectively, and v UAVx,n , v UAVy,n and v UAVz,n are the components of the actual flight speed of the unmanned aerial vehicle on the X-axis, Y-axis and Z-axis within the nth time interval T respectively; When the k-th small time interval Δt within the n-th time interval T ends, the drone position information is: x n,k = x n,k-1 + v UAVx,n × Δt y n,k = y n,k-1 + v UAVy,n × Δt z n,k = z n,k-1 + v UAVz,n × Δt where x n,0 = x n-1,K , y n,0 = y n-1,K and z n,0 = z n-1,K , and x 1,0 = 0, y 1,0 = 0 and z 1,0 = 0; Therefore, when the k-th small time interval Δt within the n-th time interval T ends, the channel gain between the drone and the landing point is expressed as: where L n,k =(x n,k , y n,k , z n,k ) represents the position coordinates of the UAV at the end of the k-th small time interval Δt within the n-th time interval T; ρ0 is the unit channel gain value; and are the line-of-sight LoS and non-line-of-sight NLoS occurrence probabilities of the additional path loss between the UAV and the landing point, respectively, and there is Ω LoS and Ω NLoS represent the additional path losses of the LoS and NLoS links, respectively, and satisfy Ω LoS << Ω NLoS .

4. A method for dynamic return of an unmanned aerial vehicle according to claim 3, characterized in that: In Step 4, in each time interval, according to the channel gain between the current position of the drone and the return point, using the game system, determine the optimal airspeed components of the drone on the X-axis, Y-axis, and Z-axis in the current time interval. The specific operations include: The game mechanism models the resource game behavior of the components of the drone's airspeed on the X-axis, Y-axis, and Z-axis, including: In any time interval, the X-axis, Y-axis, and Z-axis in the space Cartesian coordinate system act as followers in the game and obtain benefits by purchasing limited airspeed resources, while the leader is the one who sells the airspeed components on the X-axis, Y-axis, and Z-axis and obtains benefits; The optimal unit rate pricing γ is obtained by maximizing the utility functions that represent the respective interests of the leader and the competing followers. * ; The follower determines its optimal airspeed components purchased on the X, Y, and Z axes according to the optimal unit rate pricing γ * , 5. A dynamic return method for an unmanned aerial vehicle according to claim 4, characterized in that: The utility function of the leader is defined as the revenue value obtained by selling the drone's airspeed on the X-axis, Y-axis, and Z-axis, expressed as: γv ASx,n +γv ASy,n +γv ASz,n where γ represents the unit rate pricing, and v ASx,n , v ASy,n and v ASz,n are respectively the airspeed components of the UAV's airspeed on the X-axis, Y-axis, and Z-axis within the nth time interval T; The corresponding constraint conditions are expressed as: where v AS is the maximum value of the airspeed of the UAV, p is the transmission power of the UAV, σ 2 is the receiver noise at the landing point, B is the communication bandwidth, H n,k is the channel gain between the UAV and the landing point at the end of the k-th small time interval Δt within the n-th time interval T of the UAV, is the minimum required value.

6. The dynamic return method of an unmanned aerial vehicle according to claim 4, wherein: For the X-axis, its utility function is expressed as: where γ represents the unit rate pricing, p is the transmission power of the UAV, and σ 2 is the receiver noise at the landing point, B is the communication bandwidth, and H n,k,x is the channel gain between the UAV at the point (x n,k-1 +v UAVx,n ×Δt, y n,k-1 , z n,k-1 ) and the landing point; The corresponding constraint conditions are expressed as: 0≤v UAVx,n T≤x end -x n,0 where x end and x n,0 respectively represent the coordinate of the landing point on the X-axis and the position coordinate of the UAV on the X-axis at the initial moment of the nth time interval T.

7. A dynamic return method for an unmanned aerial vehicle according to claim 4, characterized in that: For the Y-axis, its utility function is expressed as: Among them, γ represents the unit rate pricing, p is the transmission power of the UAV, and σ 2 is the receiver noise at the landing point, B is the communication bandwidth, and H n,k,y is the channel gain between the UAV at the point (x n,k-1 , y n,k-1 + v UAVy,n ×Δt, z n,k-1 ) and the landing point; The corresponding constraint conditions are expressed as: 0≤v UAVy,n T≤y end -y n,0 ; where y end and y n,0 respectively represent the coordinate of the landing point on the Y-axis and the position coordinate of the UAV on the Y-axis at the initial moment of the nth time interval T.

8. A dynamic return method for an unmanned aerial vehicle according to claim 4, characterized in that: For the Z-axis, its utility function is expressed as: Among them, γ represents the unit rate pricing, p is the transmission power of the UAV, and σ 2 is the receiver noise at the landing point, B is the communication bandwidth, and H n,k,z is the channel gain between the UAV at the point (x n,k-1 , y n,k-1 , z n,k-1 + v UAVz,n ×Δt) and the landing point; The corresponding constraint conditions are expressed as: 0≤v UAVz,n T≤z end -z n,0 where z end and z n,0 represent the coordinate of the landing point on the Z-axis and the position coordinate of the UAV on the Z-axis at the initial moment of the nth time interval T, respectively.

9. A dynamic return method for a drone according to claim 4, characterized in that: If in a certain game, the X-axis or Y-axis or Z-axis does not need to purchase airspeed components, then delete this axis from the game; If there is only one axis in the game for the game of airspeed resources, then assign all the maximum values of the drone's airspeed to this axis; When the X-axis, Y-axis, and Z-axis do not purchase airspeed components, it indicates that the drone has reached the landing point.

Images