UAV positioning method that can cover agricultural and forestry robot signals

Abstract

The present invention relates to a method for positioning an unmanned aerial vehicle (UAV) that can cover robot signals based on optimization technology. The method proposes the concept of agricultural and forestry robots and their demand market. Aiming at the shortcomings of agricultural and forestry robots in terms of signal source and positioning functions, the present invention uses optimization technology to establish a multi-constrained complex semidefinite programming model. Through the structure of the multi-constrained complex semidefinite programming model, the model is decomposed into a multi-constrained simple semidefinite programming problem model and a sequence maximum element problem model. The loop strategy is used to solve the sequence maximum element problem, and an efficient random algorithm for solving multi-constrained semidefinite regularization is designed. The present invention is used to solve the problems of signal loss covering robot signals faced by UAV positioning in agricultural and forestry production.

Description

Technical Field

[0001] The invention belongs to the technical field of information processing, and relates to a drone positioning method capable of covering agricultural and forestry robot signals based on optimization technology. Background Art

[0002] In agricultural and forestry production activities, robots that can be operated automatically without human intervention are needed, which are called agricultural and forestry robots. In terms of communication, agricultural and forestry operations are large in area and wide in scope, and it is not easy to establish a signal coverage network. In addition, compared with the height of farmland crops, the height of agricultural and forestry crops cannot be ignored, which is one of the important factors affecting signal coverage and signal communication.

[0003] Robots in the agricultural and forestry area have narrower fields than farmland, so agricultural and forestry robots are smaller in size than farmland robots. The number of agricultural and forestry robots required is also related to the growth cycle of trees and the planting area. In addition, due to the high height of agricultural and forestry crops, they block signals, which brings non-negligible errors to the position monitoring of agricultural and forestry robots. Therefore, supplementary signals are needed for accurate monitoring.

[0004] The present invention will use the base station onboard the drone to supplement the signal. Its goal is to accurately supplement the signal coverage of the agricultural robot, use the drone to locate the agricultural and forestry robot, and solve the signal coverage problem based on optimization technology. Summary of the invention

[0005] In order to overcome the shortcomings of the prior art, the present invention designs a UAV positioning method based on optimization technology that can cover the signals of agricultural and forestry robots. First, the concept of agricultural and forestry robots is proposed. Aiming at the shortcomings of agricultural and forestry robots in terms of signal source and positioning functions, the UAV airborne base station is used to supplement the signal, and the optimization technology is used to establish a model that can be decomposed into semidefinite programming and sequence maximum element problem. The classical loop strategy is used to solve the sequence maximum element problem, and an efficient random algorithm for solving multi-constrained semidefinite regularization is designed. This technology is used to solve the problems of signal loss and robot anti-collision faced by UAV positioning in agricultural and forestry production.

[0006] The agricultural and forestry robot involved in the present invention refers to a machine that uses artificial intelligence technologies such as detection and calculation in agricultural and forestry production, is controlled by different program software, adapts to various operations, can sense and adapt to changes in crop types and environment in agriculture and forestry, and can be operated automatically without human intervention. This type of robot has broad market prospects in many agricultural and forestry operations such as fruit tree picking and pesticide spraying.

[0007] Robots in agriculture and forestry areas have a narrower field of view in forest gardens than in farmland, so agricultural and forestry robots are smaller in size than farmland robots. Agricultural and forestry crops are tall and block signals, so supplementary signals are needed for accurate monitoring.

[0008] The technical solution of the present invention is: a method for positioning a UAV that can cover the signal of an agricultural and forestry robot based on optimization technology. The hardware relied on by the method includes a UAV and multiple agricultural and forestry robots M. The UAV is provided with an airborne base station, and the airborne base station is used to supplement the signal of the agricultural and forestry robot M. The distance measurement unit used in the method is meter, and the time unit is minute. The optimization model for positioning a UAV that can cover the signal of an agricultural and forestry robot is established as follows:

[0009] The activity range of the i-th agricultural and forestry robot Mi is defined as an elliptical area, denoted by E i , i is a positive integer, and its major axis is The short axis is E i The center of the coordinate system is the origin, the north-south direction is the vertical axis, represented by t2, and the east-west direction is the horizontal axis, represented by t1. i The angle between the long axis and the positive semi-axis of the horizontal axis is θi, E i The rotation transformation matrix is E i The equation is:

[0010]

[0011] E i The equation data matrix is

[0012] Represents the matrix A i is positive definite, which is A i The inverse matrix of , i = 1,…,m, m is a positive integer;

[0013] Translate the origin of the coordinate system to ηi=(ηi1,ηi2) T ,ηi is E i The coordinate point of the center of the ellipse in the plane position coordinates of the GPS position, b i =A i ηi,c i =ηi T A i ηi-1, construct a maximum eigenvalue is λ 1i and the second eigenvalue is λ 2i The matrix Q i , the matrix yes The reciprocal of Q i Corresponding to λ 1i and λ 2i The eigenvectors of 1i and u 2i Then E i The minor axis is parallel to u 1i , the direction of the major axis is parallel to u 2i , E i Re-expressed as a matrix:

[0014]

[0015] i=1,…,m,x represents The point in is a two-dimensional real number space, E i The center is The signal coverage range of the airborne base station of the drone is circle B. Where β = (β1, β2) represents the coordinate point in the plane position coordinate of the GPS position of the drone, γ is a real number, and the coverage radius of the signal of the drone before takeoff is The tree height data set is H = {h1,…,h n}, n represents the number of trees, when the height of the drone from the ground is ρ, ρ is required to satisfy ρ≥1+h i , i=1,…n, then according to the Pythagorean theorem, the radiation radius of the signal of the drone in the air is because So there exists a positive integer τ i , so that the following holds:

[0016]

[0017] The UAV positioning optimization model is:

[0018]

[0019] Since the optimization of variable ρ is independent of variables β, γ, and τ, the UAV positioning optimization model is decomposed into two unrelated simple optimization sub-problems:

[0020] Multi-constrained simple semidefinite programming problem:

[0021]

[0022]

[0023] And the sequence maximum element problem:

[0024] ρ=1+max j＝1,…,n hj (3)

[0025] The method for solving the optimization model of UAV positioning is:

[0026] The first step is to solve the multi-constrained simple semidefinite programming problem:

[0027] Step 1.1: Initialization

[0028] Set the drone departure coordinates β 0 , and the coverage parameter γ 0 , τ 0 , set the step size {a k} and the penalty parameter sequence {σ k}, a k , σ k is a real number;

[0029] Step 1.2: Iteration

[0030] Iteration index k = 0, ..., N-1, N is a positive integer, and a positive integer i is randomly selected from 1, ...m k ,

[0031] Calculating the Matrix The maximum eigenvalue λ max , and the maximum eigenvalue λ max The corresponding eigenvector

[0032] If max >0.

[0033] β k+1 =β k -2a k (β k -σ k u β )

[0034] γ k+1 =γ k -2a k (-1-σ k )

[0035]

[0036] otherwise

[0037] β k+1 =β k -2a k (β k )

[0038] γ k+1 =γ k -2a k (-1)

[0039] Step1.3: Output ((β N )) T , γ N , (τ N )) T ), T , to obtain the coordinate β N ;

[0040] The second step

[0041] Use the loop strategy to solve for the maximum element ρ of the sequence = 1 + max j＝1,…,n h j ;

[0042] Step2.1: ρ = 0, j = 1;

[0043] Step2.2: If ρ ≤ h j , ρ = h j , j = j + 1;

[0044] Step2.3: If j < N, go to Step2.2;

[0045] Step2.4: Output ρ, and finally obtain the coordinate β of the agricultural and forestry drone N , with a radiation radius of

[0046] Step2.5: End.

[0047] Description of the accompanying drawings:

[0048] Figure 1 Represents the top view of agricultural and forestry operations, showing the signal transmission path of the robot.

[0049] Figure 2 Represents the altitude map of the drone, showing the altitude limit of the drone.

[0050] Figure 3 Represents the signal coverage effect diagram of the fixed - direction agricultural and forestry robot.

[0051] Figure 4 Represents the signal coverage effect diagram of the variable - direction agricultural and forestry robot.

[0052] Fixed - direction agricultural and forestry robot: The included angle between the major axis of E i and the positive semi - axis of the horizontal axis of the coordinate system is θi and is constant.

[0053] Variable - direction agricultural and forestry robot: The included angle between the major axis of E i and the positive semi - axis of the horizontal axis of the coordinate system is θi and is variable.

[0054] Advantages of the present invention

[0055] 1. The present invention combines the characteristics of agricultural production and forestry production, and proposes the concept of agricultural and forestry robots and their demand market. Compared with farmland robots, agricultural and forestry robots are smaller in size, have higher agricultural and forestry crops, and have signal shielding properties. A strategy of using drone-mounted base stations to supplement signals is proposed, which is conducive to accurate information monitoring and ensures that drones and robots will not collide with agricultural and forestry crops.

[0056] 2. Combining the elliptical coverage of the horizontal range with the information of the longitudinal height of the UAV, a multi-constrained complex semi-definite programming model is established, and the model can be decomposed into simple sub-problems using the structure, providing a guarantee for the design of efficient and fast algorithms, reducing the performance indicators required for the UAV and saving the cost of manufacturing the UAV.

[0057] 3. Design a random algorithm for solving semidefinite programming and a loop strategy for solving the maximum sequential element, which solves the problem of a large number of agricultural and forestry robots working in non-stationary conditions and ensures the real-time positioning. DETAILED DESCRIPTION

[0058] See also Figures 1 to 4 As shown, a method for positioning a UAV based on optimization technology that can cover the signal of an agricultural and forestry robot is shown. The hardware relied on by the method includes a UAV and multiple agricultural and forestry robots M. The UAV is provided with an airborne base station, and the airborne base station is used to supplement the signal of the agricultural and forestry robot M. The distance measurement unit used in the method is meter, and the time unit is minute. The optimization model for positioning a UAV that can cover the signal of an agricultural and forestry robot is established below:

[0059] The activity range of the i-th agricultural and forestry robot Mi is defined as an elliptical area, denoted by E i , i is a positive integer, and its major axis is The short axis is E i The center of the coordinate system is the origin, the north-south direction is the vertical axis, represented by t2, and the east-west direction is the horizontal axis, represented by t1. i The angle between the long axis and the positive semi-axis of the horizontal axis is θi, E i The rotation transformation matrix is E i The equation is:

[0060]

[0061] E i The equation data matrix is

[0062] Represents the matrix A i is positive definite, which is A iThe inverse matrix of , i = 1,…,m, m is a positive integer;

[0063] Translate the origin of the coordinate system to ηi=(ηi1,ηi2) T ,ηi is E i The coordinate point of the center of the ellipse in the plane position coordinates of the GPS position, b i =A i ηi,c i =ηi T A i ηi-1, construct a maximum eigenvalue is λ 1i and the second eigenvalue is λ 2i The matrix Q i , the matrix yes The reciprocal of Q i Corresponding to λ 1i and λ 2i The eigenvectors of 1i and u 2i Then E i The minor axis is parallel to u 1i , the direction of the major axis is parallel to u 2i , E i Re-expressed as a matrix:

[0064]

[0065] i=1,…,m,x represents The point in is a two-dimensional real number space, E i The center is The signal coverage range of the airborne base station of the drone is circle B. Where β = (β1, β2) represents the coordinate point in the plane position coordinate of the GPS position of the drone, γ is a real number, and the coverage radius of the signal of the drone before takeoff is The tree height data set is H = {h1,…,h n +, n represents the number of trees, when the height of the drone from the ground is ρ, ρ is required to satisfy ρ≥1+h i , i=1,…n, then according to the Pythagorean theorem, the radiation radius of the signal of the drone in the air is because So there exists a positive integer τ i , so that the following holds:

[0066]

[0067] The UAV positioning optimization model is:

[0068]

[0069] Since the optimization of variable ρ is independent of variables β, γ, and τ, the UAV positioning optimization model is decomposed into two unrelated simple optimization sub-problems:

[0070] Multi-constrained simple semidefinite programming problem:

[0071]

[0072]

[0073] And the sequence maximum element problem:

[0074] ρ=1+max j＝1,…,n h j (3)

[0075] The method for solving the optimization model of UAV positioning is:

[0076] The first step is to solve the multi-constrained simple semidefinite programming problem:

[0077] Step 1.1: Initialization

[0078] Set the drone departure coordinates β 0 , and the coverage parameter γ 0 , τ 0 , set the step size {a k} and the penalty parameter sequence {σ k}, a k , σ k is a real number;

[0079] Step 1.2: Iteration

[0080] Iteration index k = 0, ..., N-1, N is a positive integer, and a positive integer i is randomly selected from 1, ...m k ,

[0081] Calculating the Matrix The maximum eigenvalue λ max , and the maximum eigenvalue λ max The corresponding eigenvector

[0082] If max >0.

[0083] β k+1 =β k -2a k (β k -σ k u β )

[0084] γ k+1 = γ k -2a k (-1 - σ k )

[0085]

[0086] Otherwise

[0087] β k+1 = β k -2a k (β k )

[0088] γ k+1 = γ k -2a k (-1)

[0089] Step1.3: Output ((β N ) T , γ N , (τ N ) T ) T , to obtain the coordinate β N ;

[0090] The second step

[0091] Use the loop strategy to solve for the maximum element ρ of the sequence = 1 + max j＝1,…,n h j ;

[0092] Step2.1: ρ = 0, j = 1;

[0093] Step2.2: If ρ ≤ h j , ρ = h j , j = j + 1;

[0094] Step2.3: If j < N, go to Step2.2;

[0095] Step2.4: Output ρ, and finally obtain the coordinate β of the agricultural and forestry UAV N , with a radiation radius of

[0096] Step2.5: End.

[0097] Numerical results

[0098] In the numerical experiments of the present invention, fixed - direction robots and steerable robots are involved. The signal coverage effect diagrams are shown in Figure 3 and Figure 4 . In Figure 3 and Figure 4In the figure, each ellipse represents an agricultural robot, and its long axis is the real-time walking direction of the agricultural robot. The entire ellipse represents the location where the robot is expected to appear after a period of time. The dotted line represents the area where the UAV signal coverage intersects with the ground, which has covered all agricultural robots. The circle indicates that the robot is temporarily stationary. Within 20 seconds, if the agricultural robot moves at a speed of v meters / second in the working state, the long axis of the ellipse can be set to 20v meters, which can ensure that the mobile agricultural robot is always within the coverage of the UAV signal.

Claims

1. A UAV positioning method based on optimization technology that can cover agricultural and forestry robot signals is characterized by: The hardware relied on by the method includes a drone and multiple agricultural and forestry robots M. The drone is equipped with an airborne base station, which is used to supplement the signal of the agricultural and forestry robot M. The distance measurement unit used in the method is meter, and the time unit is minute. The following is an optimization model for drone positioning that can cover the agricultural and forestry robot signal: The activity range of the i-th agricultural and forestry robot Mi is defined as an elliptical area, denoted by E i , i is a positive integer, and its major axis is The short axis is E i The center of the coordinate system is the origin, the north-south direction is the vertical axis, represented by t2, and the east-west direction is the horizontal axis, represented by t1. i The angle between the long axis and the positive semi-axis of the horizontal axis is θi, E i The rotation transformation matrix is E i The equation is: E i The equation data matrix is Represents the matrix A i is positive definite, which is A i The inverse matrix of , i = 1, ..., m, m is a positive integer; Translate the origin of the coordinate system to ηi=(ηi1,ηi2) T , ηi is E i The coordinate point of the center of the ellipse in the plane position coordinates of the GPS position, b i =A i ηi,c i =ηi T A i ηi-1, construct a maximum eigenvalue is λ 1i and the second eigenvalue is λ 2i The matrix Q i , the matrix yes The reciprocal of Q i Corresponding to λ 1i and λ 2i The eigenvectors of 1i and u 2i Then E i The minor axis is parallel to u 1i , the direction of the major axis is parallel to u 2i , E i Re-expressed as a matrix: i=1,…,m,x represents The point in is a two-dimensional real number space, E i The center is The signal coverage range of the airborne base station of the drone is circle B. Where β = (β1, β2) represents the coordinate point in the plane position coordinate of the GPS position of the drone, γ is a real number, and the coverage radius of the signal of the drone before takeoff is The tree height data set is H = {h1, ..., h n }, n represents the number of trees, when the height of the drone from the ground is ρ, ρ is required to satisfy ρ≥1+h i , i = 1, ... n, then according to the Pythagorean theorem, the radiation radius of the signal of the drone in the air is because So there exists a positive integer τ i , so that the following holds: The UAV positioning optimization model is: Since the optimization of variable ρ is independent of variables β, γ, and τ, the UAV positioning optimization model is decomposed into two unrelated simple optimization sub-problems: Multi-constrained simple semidefinite programming problem: And the maximum element problem of the sequence: ρ = 1 + max j＝1，…，n h j (3) The method for solving the optimization model of UAV positioning is: The first step is to solve the multi-constrained simple semidefinite programming problem: Step 1.1: Initialization Set the drone departure coordinatesβ 0 , and the coverage parameter γ 0 , T 0 , set the step size {a k } and the penalty parameter sequence {σ k }, a k , σ k is a real number; Step 1.2: Iteration Iteration index k = 0, ..., N-1, N is a positive integer, and a positive integer i is randomly selected from 1, ...m k , Calculating the Matrix The maximum eigenvalue λ max , and the maximum eigenvalue λ max The corresponding eigenvector If max >0, b k+1 =b k -2a k (b k -s k you β ) c k+1 =c k -2a k (-1-s k ) otherwise β k+1 ＝β k -2a k (β k ) c k+1 =c k -2a k (-1) Step 1.3: Output ((β N ) T , γ N , (τ N ) T ) T , and get the coordinates β N ; Step 2 Use the loop strategy to solve the maximum element of the sequence ρ = 1 + max j＝1，…，n h j ; Step 2.1: ρ = 0, j = 1; Step 2.2: If ρ≤h j ,ρ=h j , j = j + 1; Step 2.3: If j < N, go to Step 2.2; Step 2.4: Output, ρ, and finally get the coordinates of the agricultural and forestry drone β N , the radiation radius is Step 2.5: End.

Images