In-field path planning method for reel sprinkler irrigation to reduce soil mechanical compaction

Abstract

A method for planning a path in a field for a reel-type sprinkler aimed at reducing mechanical soil compaction relates to the technical field of path planning in a field. The present invention aims to address the problems of low irrigation efficiency according to the planned path and soil structure destruction caused by soil compaction in existing field path planning methods. The present invention includes using a convex edge scanning algorithm to divide the field to obtain regular and irregular fields; obtaining the starting ridges of the irrigation paths for the regular and irregular fields; planning the irrigation paths in the field to obtain the planned field paths, and then starting irrigation from the starting ridges. The present invention is used for field path planning for a reel-type sprinkler.

Description

Technical Field

[0001] The present invention relates to the technical field of field path planning, and in particular to a field path planning method for a reel-type sprinkler aimed at reducing soil mechanical compaction. Background Art

[0002] With the rapid development of modern technology and the advancement of smart agriculture, the intelligent and precise operation of agricultural machinery has gradually become an important way to improve agricultural production efficiency and achieve sustainable development. As a key piece of equipment for farmland irrigation, reel sprinklers are widely used in modern agricultural production due to their strong adaptability, ease of mobility, and uniform irrigation.

[0003] Traditional sprinkler irrigation uses fixed, regular paths for irrigation. However, in irregular fields, this traditional irrigation path planning can lead to overlapping coverage of some areas, resulting in repeated irrigation of some areas while leaving others under-irrigated. Consequently, traditional field path planning methods lead to low irrigation efficiency and waste of water resources. Furthermore, repeated irrigation along paths planned by traditional field path planning methods can lead to repeated soil compaction, reducing soil permeability and water retention, and damaging soil structure. Summary of the Invention

[0004] The purpose of the present invention is to solve the problems of low irrigation efficiency according to the planned path and soil compaction leading to destruction of soil structure in the existing field path planning methods, and to propose a field path planning method for a reel-type sprinkler for reducing soil mechanical compaction.

[0005] A reel irrigation machine path planning method for reducing soil mechanical compaction is described as follows:

[0006] Step 1: Use the convex edge scanning algorithm to divide the fields into regular fields and irregular fields. Specifically:

[0007] Step 11: Obtain the boundary segment set of each field, and use the boundary segments of the field to construct the edge table ET;

[0008] Step 1 and 2: Use the working width L of the reel sprinkler as the moving step length of the scanning line to scan each field, and use the scanning results and the edge table ET to construct the active edge table AET;

[0009] Step 13: Obtain the intersection of each boundary line segment in the active edge table AET and the scan line, thereby obtaining a set of intersection pairs Pair;

[0010] Step 14: Use the intersection point pair set Pair to divide the field into regular fields and irregular fields;

[0011] Step 2: Get the starting ridges of the irrigation paths for regular and irregular fields, specifically:

[0012] Step 21: Use the bat algorithm to obtain the starting ridge of sprinkler irrigation in the irregular field;

[0013] Step 22: Get the starting ridge of the regular field:

[0014] The boundary lines A and B are used as tangents of the working circle to obtain the working circle. The ridge where the center of the working circle is located is the starting ridge of the regular field.

[0015] The starting points of boundary lines A and B are the same boundary point;

[0016] Step 3: Plan the irrigation path within the field, obtain the planned field path, and then start irrigation from the starting ridge.

[0017] Furthermore, in the step 11, the boundary line segment set of each field is obtained, and the edge table ET is constructed using the boundary line segments of the field, specifically:

[0018] First, obtain the field boundary point set P = {(x i ,y i )|i=1,2,...,n}, and obtain the field boundary line segment set E={e j =(x i1 ,y i1 )→(x i2 ,y i2 )|j=1,2,...,m};

[0019] Among them, (x i ,y i ) is the coordinate of the i-th field boundary point, n is the total number of field boundary points, (x i1 ,y i1 ) is the coordinate of the i1th field boundary point, (x i2 ,y i2 ) is the coordinate of the i2th field boundary point, j is the field boundary segment number, m is the total number of field boundary segments, e j is the jth boundary segment of the field, → is the connection relationship;

[0020] Then, the field boundary segment is used to obtain the field boundary segment increment Δx j :

[0021]

[0022] Among them, || is the absolute value symbol;

[0023] Finally, using the field boundary segment increment Δx jGet the edge table ET:

[0024] ET={(y min,j ,y max,j ,x min,j ,Δx j )|j=1,2,...,m}

[0025] y min,j =min(y i1 ,y i2 )

[0026] y max,j =max(y i1 ,y i2 )

[0027] Among them, y min,j It is e j The minimum value of the two endpoints of the ordinate, y max,j It is e j The maximum value of the two endpoints of the horizontal coordinate, x min,j It is y min,j The horizontal coordinate of the corresponding endpoint.

[0028] Furthermore, in steps 1 and 2, the working width L of the reel sprinkler is used as the moving step length of the scanning line to scan each field, and the active edge table AET is constructed using the scanning results and the edge table ET, specifically:

[0029] First, obtain the field circumscribed rectangle R = [(x min ,y min ),(x max ,y max )];

[0030] x min =min(x i ),x max =max(x i ),y min =min(y i ),y max =max(y i )

[0031] Among them, x min is the minimum value of the horizontal coordinate of the field boundary point, x max is the maximum value of the horizontal coordinate of the field boundary point, y min is the minimum value of the vertical coordinate of the field boundary point, y max is the maximum value of the vertical coordinate of the field boundary point, (x min ,y min ) is the coordinate of the lower left corner of the circumscribed rectangle of the field boundary, (x max ,y max) is the coordinate of the upper right corner of the circumscribed rectangle of the field boundary;

[0032] Then, y min As the starting ordinate of the scanning line y', with L as the moving step of the scanning line, each field is scanned along the y-axis direction until y'>y max , get y min to y max All boundary segments within the range, and add the minimum abscissa value, maximum ordinate value and corresponding boundary segment increment of each boundary segment to the active edge table AET:

[0033] AET={(x min,j ,Δx j ,y max,j )|j=1,2,...,m}.

[0034] Furthermore, in step 13, the intersection of each boundary line segment in the active edge table AET and the scan line is obtained to obtain a set of intersection pairs Pair, specifically:

[0035] First, obtain the horizontal coordinates of the intersection of each boundary line segment in the active edge table AET and the scan line:

[0036] x current,j =x min,j +(y'-y min,j )·Δx j

[0037] Among them, x current,j is the horizontal coordinate of the intersection of the jth boundary segment and the scan line;

[0038] Then, arrange the horizontal coordinates of the intersection points of each boundary line segment and the scan line in ascending order, and form an intersection point pair with each two horizontal coordinates in order to obtain the intersection point pair set Pair, specifically:

[0039] Pair={(x 2k-1 ,x 2k ),k=1,...,K}

[0040] Among them, x 2k-1 is the horizontal coordinate of the 2k-1th intersection point, x 2k is the horizontal coordinate of the 2kth intersection point, and 2K is the total number of intersection points between all boundary segments and scan lines.

[0041] Furthermore, in step 14, the field is divided using the intersection point pair set Pair to obtain regular field plots and irregular field plots, specifically:

[0042] First, we use the intersection point pair set Pair to obtain the regular field set R = {R k|k=1,2,...,K}:

[0043] R k =[(x 2k-1 ,y),(x 2k ,y+L)]

[0044] Among them, R k is the kth regular region, (x 2k-1 ,y) is the coordinate of the upper left corner of the kth regular area, (x 2k ,y+L) is the coordinate of the lower right corner of the kth regular area;

[0045] Then, the coordinates in the regular field set R are removed from the field boundary point set P, and the area composed of the other coordinates is the irregular field.

[0046] Furthermore, the bat algorithm is used in step 21 to obtain the sprinkler irrigation starting ridge in the irregular field, specifically:

[0047] A1. Randomly initialize N starting ridge candidate points (x i' ,y i' ); Initialize the current time t = 0, and use the initialized starting ridge candidate point as the bat's initial position X i' (t) ; Initialize bat speed V i' (t) =0;

[0048] Among them, (x i' ,y i' ) is the position of the i'th starting ridge candidate point, i' is the number of the starting ridge candidate point, x min ≤x i' ≤x max ,y min ≤y i' ≤y max ;

[0049] The position of the candidate starting ridge point of the i'th ridge is the midpoint of the i'th ridge;

[0050] A2. Obtain the adaptability value of each bat and take the bat with the smallest fitness value as the optimal bat X best , and record the position of the optimal bat (x best ,y best );

[0051] A3. Randomly generate a pulse frequency for each bat, and obtain the bat's flight speed based on the pulse frequency, thereby updating the bat's position according to the bat's flight speed:

[0052] First, generate the pulse frequency corresponding to the bat:

[0053] f i' =f min +(f max -f min )·rand()

[0054] Among them, f i' is the pulse frequency corresponding to bat i', [f min ,f max ] is the pulse search range, f min is the pulse search minimum value, f max is the pulse search maximum value, rand() is the random number generation function;

[0055] Then, update the bat's flight speed based on the pulse frequency:

[0056] V i' (t+1) =ω·V i' (t) +(X best -X i' (t) )·f i'

[0057] Among them, V i' (t+1) is the flight speed of bat i' at time t+1, ω is the inertia weight, V i' (t) is the flying speed of bat i' at time t, X best is the current optimal bat position, X i' (t) is the position of bat i' at time t;

[0058] Finally, update the bat's position based on its flight speed:

[0059] X i' (t+1) =X i' (t) +V i' (t+1)

[0060] Among them, X i' (t+1) is the position of bat i' at time t+1, X i' (t) is the position of bat i' at time t, V i' (t+1) is the flight speed of bat i' at time t+1;

[0061] A4. Calculate the fitness value of the bat after updating the bat position. If the fitness value of the bat after updating the position is less than the fitness value of the optimal bat, then the bat after the current updated bat position is regarded as the optimal bat, and the updated position is regarded as the optimal bat position and recorded; otherwise, the optimal bat and the optimal bat position remain unchanged.

[0062] A5. Randomly perturb the optimal bat and update the position of the optimal bat:

[0063] Generate a random number rand1 in the range [0,1] and compare the random number rand1 with the pulse frequency r of bat i' i' Compare, if rand1>r i' , then perform a local random perturbation on the optimal bat, use the perturbed optimal bat position as the updated optimal bat position, and then execute step A6; otherwise, update the optimal bat position using the following formula:

[0064]

[0065] in, is the updated optimal bat position, A i' is the pulse intensity of bat i', ε∈[-1,1], ε is a random perturbation;

[0066] A6. Get the fitness value of each bat after updating the bat position. If the fitness value of the bat after updating is less than the fitness value of the bat before updating, F(X i' (t+1) )<F(X i' (t) ), then the bat position X will be updated i' (t+1) As the position of the corresponding bat; otherwise, the bat position X i' (t) constant;

[0067] A7. Get the current fitness values ​​of all bats and obtain the minimum fitness value. If the minimum fitness value is less than the current optimal bat fitness value, update the pulse frequency r. i' Harmony Strong A i' , then execute step A8; otherwise, directly execute step A8;

[0068] Update pulse frequency r i' Harmony Strong A i' , using the following formula:

[0069] A i' (t+1) =α·A i' (t)

[0070] r i'(t+1) =r i' (t) ·(1-e -βt )

[0071] Among them, α is the sound intensity attenuation coefficient, β is the frequency increase coefficient, A i' (t) is the pulse intensity of bat i' at time t, r i' (t) is the pulse frequency of bat i' at time t;

[0072] A8. Update the optimal bat position to the bat position corresponding to the current minimum fitness value, and then determine whether the maximum number of iterations T has been reached. max , if t=T max The ridge corresponding to the current optimal bat position is the starting ridge for sprinkler irrigation in the irregular field; otherwise, set t=t+1 and return to step A3.

[0073] Furthermore, the step A2 of obtaining the adaptability value of each bat is as follows:

[0074]

[0075] Among them, C irregular is the geometric center point of the irregular field, j' is the point number in the irregular field, N' is the total number of points in the irregular field, (x j' ,y j' ) are the coordinates of the j'th point in the irregular field.

[0076] Furthermore, in step 3, the irrigation path in the field is planned to obtain the planned field path, and then irrigation is started from the starting ridge, specifically:

[0077] Step 3. Randomly initialize the positions in M ​​furrows and use the positions in M ​​furrows to randomly generate M' paths R m' =[(x m',1 ,y m',1 ),(x m',2 ,y m',2 ),...,(x m',M ,y m',M )], each path is a firefly;

[0078] Among them, (x m',1 ,y m',1 ) is the first point in the m'th path, (x m',M ,y m',M ) is the Mth point in the m'th path;

[0079] Step 32: Obtain the objective function value of each firefly, take the firefly with the minimum objective function value as the optimal firefly, and take the objective function value of the optimal firefly as the fluorescence brightness of the optimal firefly I m' ;

[0080] Step 3. Get the fluorescence brightness of the two fireflies I m' and Get the fluorescence brightness difference and attraction of two fireflies:

[0081] Among them, I m' is the fluorescence brightness of firefly m', is the fluorescence brightness of the firefly m*;

[0082] Step 3 and 4: According to the fluorescence brightness difference of the fireflies, the brighter firefly between the two fireflies is obtained. Firefly m' moves towards the brighter firefly m*, and the attraction between the two fireflies is used to update the position of firefly m':

[0083]

[0084] in, is the position of firefly m' at time t+1, is the position of firefly m' at time t, is the position of the firefly m* at time t, α' is the step size factor, ε'∈[-1,1], ε' is random noise, is the attraction between firefly m' and firefly m*;

[0085] Step 3.5: Randomly perturb the current optimal firefly and update the optimal firefly:

[0086]

[0087] in, is the optimal firefly position at time t+1, is the optimal firefly position at time t;

[0088] Step 36: Obtain the objective function value of the updated firefly, compare the objective function of the updated firefly with the objective function value of the optimal firefly, update the firefly corresponding to the minimum objective function value to the optimal firefly, and use the objective function value of the optimal firefly as the fluorescence brightness of the optimal firefly;

[0089] Step 37: Determine whether the number of iterations T' has been reached max , if t'=T' max Then output the path corresponding to the optimal firefly as the planned field path; otherwise, set t'=t'+1 and return to step 33;

[0090] Step 38: Using the starting ridge obtained in step 2 as the starting point, sprinkler irrigation is performed according to the planned field path obtained in step 37.

[0091] Furthermore, the objective function value of each firefly in step 32 is obtained as follows:

[0092] F(R k )=ω1·D m',total +ω2·C m',compact -ω3·E m',efficiency

[0093]

[0094] Among them, ω1 is the total path length weight, ω2 is the compaction weight, ω3 is the sprinkler efficiency weight, D k,total is the total length of the kth path, C k,compact is the compaction degree of the m'th path, E k,efficiency is the sprinkler efficiency of the m'th path, F(R m' ) is the location set in the m'th path, D total,m' is the total length of the m'th path, D row,w is the row distance of the wth operation, d return,w' is the return distance of the w'th operation, W is the total number of operations, compact(l') is the function of whether the l'th ridge is compacted, if compacted, compact(l') is 1, if not compacted, compact(l') is 0, L' is the total number of ridges, C compact,m' is the number of ridges compacted by sprinkler irrigation along the m'th path, E efficiency,m' is the sprinkler efficiency of the m'th path, A cov,m' is the coverage area of ​​the m'th path, n rows is the number of ridges covered in one operation, L is the working width of the convolution disk, and W * is the ridge width, is the floor symbol, A cov,l' is the irrigation area of ​​the l'th ridge, t end,l' is the end time of sprinkler irrigation of the l'th ridge, t start,l' is the start time of sprinkler irrigation for the l'th ridge, R dynamic (t) is the sprinkler radius, t is the time, R0 is the maximum sprinkler radius, v(t) is the variable speed of the reel sprinkler, v max is the maximum speed of the reel sprinkler;

[0095] If the sprinkler passes over the current ridge during its travel path, it means that the current ridge is compacted.

[0096] Furthermore, the step 33 of obtaining the fluorescence brightness of the two fireflies Im' and Get the fluorescence brightness difference and attractiveness of two fireflies, specifically:

[0097]

[0098] in, is the difference in fluorescence brightness between firefly m' and firefly m*, is the attraction between firefly m' and firefly m*, β0 is the maximum attraction, γ' is the attraction attenuation coefficient, is the Euclidean distance between firefly m' and firefly m*, X' m' is the location of firefly m', is the position of firefly m*.

[0099] The beneficial effects of the present invention are:

[0100] This paper proposes a method for field path planning for reel-type sprinklers to mitigate soil compaction. By distinguishing between regular and irregular areas within the field and optimizing sprinkler coverage and operating paths, this method plans an irrigation path that reduces the impact of mechanical compaction. This method uses a bat algorithm to determine the irrigation starting point and combines it with a firefly algorithm to optimize the irrigation path. This method achieves an optimal irrigation path with optimal path length, compaction level, and irrigation efficiency, improving irrigation efficiency while reducing soil compaction, maintaining soil permeability and water retention, and minimizing soil structural damage. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] Figure 1 Flowchart of the present invention;

[0102] Figure 2 It is a convex edge scanning line flow chart;

[0103] Figure 3 This is the flowchart of the bat algorithm;

[0104] Figure 4 This is the flow chart of the firefly algorithm. DETAILED DESCRIPTION

[0105] Specific implementation method 1: Figure 1 As shown, the specific process of the in-field path planning method for a reel-type sprinkler for reducing soil mechanical compaction in this embodiment is as follows:

[0106] Step 1: Figure 2 As shown in the figure, the convex edge scanning algorithm is used to classify the fields to obtain regular fields and irregular fields, specifically:

[0107] Step 11: Get the boundary segment set of each field, and use the boundary segments of the field to construct the edge table ET:

[0108] First, obtain the field boundary point set P = {(x i ,y i )|i=1,2,...,n}, and obtain the field boundary line segment set E={e j =(x i1 ,y i1 )→(x i2 ,y i2 )|j=1,2,...,m};

[0109] Among them, (x i ,y i ) is the coordinate of the i-th field boundary point, n is the total number of field boundary points, (x i1 ,y i1 ) is the coordinate of the i1th field boundary point, (x i2 ,y i2 ) is the coordinate of the i2th field boundary point, j is the field boundary segment number, m is the total number of field boundary segments, e j is the jth boundary line segment of the field, → is the connection relationship; the lower left corner vertex in the convex field is the origin of the coordinate axis, the x-axis direction is the direction perpendicular to the ridge, and the y-axis direction is the ridge direction.

[0110] Then, the field boundary segment is used to obtain the field boundary segment increment Δx j :

[0111]

[0112] Among them, || is the absolute value symbol;

[0113] Finally, using the field boundary segment increment Δx j Get the edge table ET:

[0114] ET={(y min,j ,y max,j ,x min,j ,Δx j )|j=1,2,...,m}

[0115] y min,j =min(y i1 ,y i2 )

[0116] y max,j =max(y i1 ,y i2 )

[0117] Among them, y min,j It is e j The minimum value of the ordinates of the two endpoints, ymax,j It is e j The maximum value of the horizontal coordinates of the two endpoints, x min,j It is y min,j The horizontal coordinate of the corresponding endpoint.

[0118] Step 1 and 2: Use the working width L of the reel sprinkler as the moving step length of the scan line, scan each field, and construct the active edge table AET based on the scan results and the edge table ET, specifically:

[0119] First, obtain the field circumscribed rectangle R = [(x min ,y min ),(x max ,y max )];

[0120] x min =min(x i ),x max =max(x i ),y min =min(y i ),y max =max(y i )

[0121] Among them, x min is the minimum value of the horizontal coordinate of the field boundary point, x max is the maximum value of the horizontal coordinate of the field boundary point, y min is the minimum value of the vertical coordinate of the field boundary point, y max is the maximum value of the vertical coordinate of the field boundary point, (x min ,y min ) is the coordinate of the lower left corner of the circumscribed rectangle of the field boundary, (x max ,y max ) is the coordinate of the upper right corner of the circumscribed rectangle of the field boundary;

[0122] Then, y min As the starting ordinate of the scanning line y', with L as the moving step of the scanning line, each field is scanned along the y-axis direction until y'>y max , get y min to y max All boundary segments within the range, and add the minimum abscissa value, maximum ordinate value and corresponding boundary segment increment of each boundary segment to the active edge table AET:

[0123] AET={(x min,j ,Δx j ,y max,j )|j=1,2,...,m}

[0124] Step 13: Use the intersection of each boundary line segment in the active edge table AET and the scan line to obtain a set of intersection pairs, specifically:

[0125] First, obtain the horizontal coordinates of the intersection of each boundary line segment in the active edge table AET and the scan line:

[0126] x current,j =x min,j +(y'-y min,j )·Δx j

[0127] Among them, x current,j is the horizontal coordinate of the intersection of the jth boundary segment and the scan line;

[0128] Then, arrange the horizontal coordinates of the intersection points of each boundary line segment and the scan line in ascending order, and form an intersection point pair with each two horizontal coordinates in order to obtain the intersection point pair set Pair, specifically:

[0129] Pair={(x 2k-1 ,x 2k ),k=1,...,K}

[0130] Among them, x 2k-1 is the horizontal coordinate of the 2k-1th intersection point, x 2k is the horizontal coordinate of the 2kth intersection point, 2K is the total number of intersections between all boundary segments and scan lines;

[0131] Step 14: Use the intersection point pair set Pair to classify the fields to obtain regular fields and irregular fields;

[0132] Regular field set R = {R k |k=1,2,...,K} is:

[0133] R k =[(x 2k-1 ,y),(x 2k ,y+L)]

[0134] Among them, R k are the coordinates of the upper left corner and lower right corner of the kth regular area, (x 2k-1 ,y) is the coordinate of the upper left corner of the kth regular area, (x 2k ,y+L) is the coordinate of the lower right corner of the kth regular area;

[0135] In the field boundary point set P, except for the coordinates of the regular field R, the area composed of other coordinates is the irregular field area.

[0136] Step 2: Figure 3 As shown, the starting ridges of the sprinkler irrigation paths for regular and irregular fields are obtained as follows:

[0137] Step 2.1: Use the bat algorithm to obtain the starting ridge for sprinkler irrigation in irregular fields:

[0138] A1. Randomly initialize N starting ridge candidate points (x i' ,y i' ); Initialize the current time t = 0, and use the initialized starting ridge candidate point as the bat's initial position X i' (t) ; Initialize bat speed V i' (t) =0;

[0139] Among them, (x i' ,y i' ) is the position of the i'th starting ridge candidate point (the starting ridge candidate position refers to the midpoint of the current ridge), i' is the number of the starting ridge candidate point, x min ≤x i' ≤x max ,y min ≤y i' ≤y max ;

[0140] A2. Obtain the adaptability value of each bat and take the bat with the smallest fitness value as the optimal bat X best , and record the position of the optimal bat (x best ,y best ):

[0141] The adaptability value of the bat is obtained by:

[0142]

[0143] Among them, (x center ,y center ) are the coordinates of the geometric centre of the irregular field;

[0144]

[0145] Among them, C irregular is the geometric center point of the irregular field, j' is the point number in the irregular field, N' is the total number of points in the irregular field, (x j' ,y j' ) are the coordinates of the j'th point in the irregular field;

[0146] A3. Randomly generate a pulse frequency for each bat, and obtain the bat's flight speed based on the pulse frequency, thereby updating the bat's position according to the bat's flight speed:

[0147] First, generate the pulse frequency corresponding to the bat:

[0148] fi' =f min +(f max -f min )·rand()

[0149] Among them, f i' is the pulse frequency corresponding to bat i', [f min ,f max ] is the pulse search range (such as 0 to 2), f min is the pulse search minimum value, f max is the pulse search maximum value, rand() is the random number generation function;

[0150] Then, update the bat's flight speed based on the pulse frequency:

[0151] V i' (t+1) =ω·V i' (t) +(X best -X i' (t) )·f i'

[0152] Among them, V i' (t+1) is the flight speed of bat i' at time t+1, ω is the inertia weight used to balance global and local search (0.9-0.4 dynamically decreasing), V i' (t) is the flying speed of bat i' at time t, X best is the current optimal bat position, X i' (t) is the position of bat i' at time t;

[0153] Finally, update the bat's position based on its flight speed:

[0154] X i' (t+1) =X i' (t) +V i' (t+1)

[0155] Among them, X i' (t+1) is the position of bat i' at time t+1, X i' (t) is the position of bat i' at time t, V i' (t+1) is the flight speed of bat i' at time t+1;

[0156] A4. Calculate the fitness value of the bat after updating the bat position. If the fitness value of the bat after updating the position is less than the fitness value of the optimal bat, then the bat after the current updated bat position is regarded as the optimal bat, and the updated position is regarded as the optimal bat position and recorded; otherwise, the optimal bat and the optimal bat position remain unchanged.

[0157] A5. Randomly perturb the optimal bat and update the position of the optimal bat:

[0158] Generate a random number rand1∈[0,1] and compare the random number rand1 with the pulse frequency r of bat i' i' Compare, if rand1>r i' , then perform a local random perturbation on the optimal bat, use the perturbed optimal bat position as the updated optimal bat position, and then execute step A6; otherwise, update the optimal bat position using the following formula:

[0159]

[0160] in, is the updated optimal bat position, A i' is the pulse intensity of bat i', ε∈[-1,1], ε is a random perturbation;

[0161] A6. Get the fitness value of each bat after updating the bat position. If the fitness value of the bat after updating is less than the fitness value of the bat before updating, F(X i' (t+1) )<F(X i' (t) ), then the bat position X will be updated i' (t+1) As the position of the corresponding bat; otherwise, the bat position X i' (t) constant;

[0162] A7. Get the current fitness values ​​of all bats and obtain the minimum fitness value. If the minimum fitness value is less than the current optimal bat fitness value, update the pulse frequency r. i' Harmony Strong A i' , then execute step A8; otherwise, directly execute step A8;

[0163] A i' (t+1) =α·A i' (t)

[0164] r i' (t+1) =r i' (t) ·(1-e -βt )

[0165] Among them, α = 0.98 is the sound intensity attenuation coefficient, β = 0.02 is the frequency increase coefficient, A i' (t) is the pulse intensity of bat i' at time t, r i' (t) is the pulse frequency of bat i' at time t;

[0166] A8. Update the optimal bat position to the bat position corresponding to the current minimum fitness value, and then determine whether the maximum number of iterations T has been reached. max , if t=T max The ridge corresponding to the current optimal bat position is the starting ridge for sprinkler irrigation in the irregular field; otherwise, set t=t+1 and return to step A3.

[0167] Step 22: The starting ridge of the regular field is obtained by the following method:

[0168] Take boundary lines A and B as tangents of the working circle to obtain the working circle. The ridge where the center of the working circle is located is the starting ridge of the regular field.

[0169] The starting point of boundary line A and boundary line B is the same boundary point.

[0170] Step 3: Plan the irrigation path within the field, obtain the planned field path, and then start irrigation from the starting ridge, such as Figure 4 As shown, specifically:

[0171] Step 3. Randomly initialize the positions in M ​​furrows and use the positions in M ​​furrows to randomly generate M' paths R m' =[(x m',1 ,y m',1 ),(x m',2 ,y m',2 ),...,(x m',M ,y m',M )], each path is a firefly;

[0172] Among them, (x m',1 ,y m',1 ) is the first point in the m'th path, (x m',M ,y m',M ) is the Mth point in the m'th path;

[0173] Step 32: Obtain the objective function value of each firefly, take the firefly with the minimum objective function value as the optimal firefly, and take the objective function value of the optimal firefly as the fluorescence brightness of the optimal firefly I m' ;

[0174] The specific objective function value for each firefly is obtained as follows:

[0175] F(R k )=ω1·D m',total +ω2·C m',compact -ω3·E m',efficiency

[0176]

[0177] Among them, ω1 is the total path length weight, ω2 is the compaction weight, ω3 is the sprinkler efficiency weight, D k,total is the total length of the kth path, C k,compact is the compaction degree of the m'th path, E k,efficiency is the sprinkler efficiency of the m'th path, F(R m' ) is the location set in the m'th path, D total,m' is the total length of the m'th path, D row,w is the row distance of the wth operation, d return,w' is the return distance of the w'th operation, W is the total number of operations, compact(l') is the function of whether the l'th ridge is compacted, if it is compacted it is 1 (walking once in the ridge is considered compacted), if it is not compacted it is 0, L' is the total number of ridges, C compact,m' is the number of ridges compacted by sprinkler irrigation along the m'th path, E efficiency,m' is the sprinkler efficiency of the m'th path, A cov,m' is the coverage area of ​​the m'th path, n rows is the number of ridges covered in one operation, L is the working width of the convolution disk, and W * is the ridge width, is the floor symbol, A cov,l' is the irrigation area of ​​the l'th ridge, t end,l' is the end time of sprinkler irrigation of the l'th ridge, t start,l' is the start time of sprinkler irrigation for the l'th ridge, R dynamic (t) is the sprinkler radius, t is the time, R0 is the maximum sprinkler radius, v(t) is the variable speed of the reel sprinkler, v max is the maximum speed of the reel sprinkler.

[0178] Step 3. Get the fluorescence brightness of the two fireflies I m' and Get the fluorescence brightness difference and attraction of two fireflies:

[0179]

[0180] in, is the difference in fluorescence brightness between firefly m' and firefly m*, is the attraction between firefly m' and firefly m*, β0 = 1 is the maximum attraction, γ' = 1 is the attenuation coefficient of attraction with distance, is the Euclidean distance between firefly m' and firefly m*, X' m' is the location of firefly m', is the position of firefly m*;

[0181] Step 3 and 4: Determine the brighter firefly between each pair of fireflies based on the fluorescence brightness difference between the two fireflies. Firefly m' moves toward the brighter firefly m*, and the attraction between the two fireflies is used to update the position of firefly m':

[0182]

[0183] in, is the position of firefly m' at time t+1, is the position of firefly m' at time t, is the position of firefly m* at time t, α' = 0.2 is the step size factor, and ε'∈[-1,1] is uniform random noise;

[0184] Step 3.5: Randomly perturb the current optimal firefly and update the optimal firefly:

[0185]

[0186] in, is the optimal firefly position at time t+1, is the optimal firefly position at time t;

[0187] Step 36: Obtain the objective function value of the updated firefly, compare the objective function value of the updated firefly with the objective function value of the optimal firefly, update the firefly corresponding to the minimum objective function value to the optimal firefly, and use the objective function value of the optimal firefly as the brightness of the optimal firefly;

[0188] Step 37: Determine whether the number of iterations T' has been reached max , if t'=T' max Then output the path corresponding to the optimal firefly as the planned field path; otherwise, set t'=t'+1 and return to step 33.

[0189] Step 38: Using the starting ridge obtained in step 2 as the starting point, sprinkler irrigation is performed according to the planned field path obtained in step 37.

Claims

1. A method for field path planning of a reel-type sprinkler for reducing soil mechanical compaction, characterized by The specific process of the method is: Step 1: Use the convex edge scanning algorithm to divide the fields into regular fields and irregular fields. Specifically: Step 11: Obtain the boundary segment set of each field, and use the boundary segments of the field to construct the edge table ET; Step 1 and 2: Use the working width L of the reel sprinkler as the moving step length of the scanning line to scan each field, and use the scanning results and the edge table ET to construct the active edge table AET; Step 13: Obtain the intersection of each boundary line segment in the active edge table AET and the scan line, thereby obtaining a set of intersection pairs Pair; Step 14: Use the intersection point pair set Pair to divide the field into regular fields and irregular fields; Step 2: Get the starting ridges of the irrigation paths for regular and irregular fields, specifically: Step 21: Use the bat algorithm to obtain the starting ridge of sprinkler irrigation in the irregular field; Step 22: Get the starting ridge of the regular field: The boundary lines A and B are used as tangents of the working circle to obtain the working circle. The ridge where the center of the working circle is located is the starting ridge of the regular field. The starting points of boundary lines A and B are the same boundary point; Step 3: Plan the irrigation path within the field, obtain the planned field path, and then start irrigation from the starting ridge.

2. The method for field path planning of a reel-type sprinkler for reducing soil mechanical compaction according to claim 1, characterized in that: In the step 11, the boundary line segment set of each field is obtained, and the edge table ET is constructed using the boundary line segments of the field, specifically: First, obtain the field boundary point set P = {(x i ,y i )|i=1,2,...,n}, and obtain the field boundary line segment set E={e j =(x i1 ,y i1 )→(x i2 ,y i2 )|j=1,2,...,m}; Among them, (x i ,y i ) is the coordinate of the i-th field boundary point, n is the total number of field boundary points, (x i1 ,y i1 ) is the coordinate of the i1th field boundary point, (x i2 ,y i2 ) is the coordinate of the i2th field boundary point, j is the field boundary segment number, m is the total number of field boundary segments, e j is the jth boundary segment of the field, → is the connection relationship; Then, the field boundary segment is used to obtain the field boundary segment increment Δx j : Among them, || is the absolute value symbol; Finally, use the field boundary segment increment Δx j Get the edge table ET: ET={(y min,j ,y max,j ,x min,j ,Δx j )|j=1,2,...,m} and min,j =min(y i1 ,and i2 ) and max,j =max(y i1 ,and i2 ) Among them, y min,j It is e j The minimum value of the two endpoints of the ordinate, y max,j It is e j The maximum value of the two endpoints of the horizontal coordinate, x min,j It is y min,j The horizontal coordinate of the corresponding endpoint.

3. The method for field path planning for a reel-type sprinkler for reducing soil mechanical compaction according to claim 2, characterized in that: In the steps 1 and 2, the working width L of the reel sprinkler is used as the moving step length of the scanning line to scan each field, and the active edge table AET is constructed using the scanning results and the edge table ET, specifically: First, obtain the field's circumscribed rectangle R = [(x min ,y min ),(x max ,y max )]; x min =min(x i ),x max =max(x i ),y min =min(y i ),y max =max(y i ) Among them, x min is the minimum value of the horizontal coordinate of the field boundary point, x max is the maximum value of the horizontal coordinate of the field boundary point, y min is the minimum value of the vertical coordinate of the field boundary point, y max is the maximum value of the vertical coordinate of the field boundary point, (x min ,y min ) is the coordinate of the lower left corner of the circumscribed rectangle of the field boundary, (x max ,y max ) is the coordinate of the upper right corner of the circumscribed rectangle of the field boundary; Then, y min As the starting ordinate of the scanning line y', with L as the moving step of the scanning line, each field is scanned along the y-axis direction until y'>y max , get y min to y max All boundary segments within the range, and add the minimum abscissa value, maximum ordinate value and corresponding boundary segment increment of each boundary segment to the active edge table AET: AET={(x min,j ,Δx j ,y max,j )|j=1,2,...,m}。 4. The method for field path planning for a reel-type sprinkler for reducing soil mechanical compaction according to claim 3, characterized in that: The step 13 obtains the intersection of each boundary line segment in the active edge table AET and the scan line, thereby obtaining a set of intersection pairs Pair, specifically: First, obtain the horizontal coordinates of the intersection of each boundary line segment in the active edge table AET and the scan line: x current,j =x min,j +(y'-y min,j )·Δx j Among them, x current,j is the horizontal coordinate of the intersection of the jth boundary segment and the scan line; Then, arrange the horizontal coordinates of the intersection points of each boundary line segment and the scan line in ascending order, and form an intersection point pair with each two horizontal coordinates in order to obtain the intersection point pair set Pair, specifically: Pair={(x 2k-1 ,x 2k ),k=1,...,K} Among them, x 2k-1 is the horizontal coordinate of the 2k-1th intersection point, x 2k is the horizontal coordinate of the 2kth intersection point, and 2K is the total number of intersection points between all boundary segments and scan lines.

5. The method for field path planning of a reel-type sprinkler for reducing soil mechanical compaction according to claim 4, characterized in that: In step 14, the field is divided using the intersection point pair set Pair to obtain regular field plots and irregular field plots, specifically: First, we use the intersection point pair set Pair to obtain the regular field set R = {R k |k=1,2,...,K}: R k =[(x 2k-1 ,y),(x 2k ,y+L)] Among them, R k is the kth regular region, (x 2k-1 ,y) is the coordinate of the upper left corner of the kth regular area, (x 2k ,y+L) is the coordinate of the lower right corner of the kth regular area; Then, the coordinates in the regular field set R are removed from the field boundary point set P, and the area composed of the other coordinates is the irregular field.

6. The method for field path planning for a reel-type sprinkler for reducing soil mechanical compaction according to claim 5, characterized in that: The bat algorithm is used in step 21 to obtain the sprinkler irrigation starting ridge in the irregular field, specifically: A1. Randomly initialize N starting ridge candidate points (x i' ,y i' ); Initialize the current time t = 0, and use the initialized starting ridge candidate point as the bat's initial position X i' (t) ; Initialize bat speed V i' (t) =0; Among them, (x i' ,y i' ) is the position of the i'th starting ridge candidate point, i' is the number of the starting ridge candidate point, x min ≤x i' ≤x max ,y min ≤y i' ≤y max ; The position of the candidate starting ridge point of the i'th ridge is the midpoint of the i'th ridge; A2. Obtain the adaptability value of each bat and take the bat with the smallest fitness value as the optimal bat X best , and record the position of the optimal bat (x best ,y best ); A3. Randomly generate a pulse frequency for each bat, and obtain the bat's flight speed based on the pulse frequency, thereby updating the bat's position according to the bat's flight speed: First, generate the pulse frequency corresponding to the bat: f i' =f min +(f max -f min )·rand() Among them, f i' is the pulse frequency corresponding to bat i', [f min ,f max ] is the pulse search range, f min is the pulse search minimum, f max is the pulse search maximum value, rand() is the random number generation function; Then, update the bat's flight speed based on the pulse frequency: V i' (t+1) =ω·V i' (t) +(X best -X i' (t) )·f i' Among them, V i' (t+1) is the flight speed of bat i' at time t+1, ω is the inertia weight, V i' (t) is the flying speed of bat i' at time t, X best is the current optimal bat position, X i' (t) is the position of bat i' at time t; Finally, update the bat's position based on its flight speed: X i' (t+1) =X i' (t) +V i' (t+1) Δt Among them, X i' (t+1) is the position of bat i' at time t+1, X i' (t) is the position of bat i' at time t, V i' (t+1) is the flight speed of bat i' at time t+1, Δt is the time step; A4. Calculate the fitness value of the bat after updating the bat position. If the fitness value of the bat after updating the position is less than the fitness value of the optimal bat, then the bat after the current updated bat position is regarded as the optimal bat, and the updated position is regarded as the optimal bat position and recorded; otherwise, the optimal bat and the optimal bat position remain unchanged. A5. Randomly perturb the optimal bat and update the position of the optimal bat: Generate a random number rand1 in the range [0,1] and compare the random number rand1 with the pulse frequency r of bat i' i' Compare, if rand1>r i' , then perform a local random perturbation on the optimal bat, use the perturbed optimal bat position as the updated optimal bat position, and then execute step A6; otherwise, update the optimal bat position using the following formula: in, is the updated optimal bat position, A i' is the pulse intensity of bat i', ε∈[-1,1], ε is a random perturbation; A6. Get the fitness value of each bat after updating the bat position. If the fitness value of the bat after updating is less than the fitness value of the bat before updating, F(X i' (t+1) )<F(X i' (t) ), then the bat position X will be updated i' (t+1) As the position of the corresponding bat; otherwise, the bat position X i' (t) constant; A7. Get the current fitness values ​​of all bats and obtain the minimum fitness value. If the minimum fitness value is less than the current optimal bat fitness value, update the pulse frequency r. i' Harmony Strong A i' , then execute step A8; otherwise, directly execute step A8; Update pulse frequency r i' Harmony Strong A i' , using the following formula: A i' (t+1) =α·A i' (t) r i' (t+1) ＝r i' (t) ·(1-e -βt ) Among them, α is the sound intensity attenuation coefficient, β is the frequency increase coefficient, A i' (t) is the pulse intensity of bat i' at time t, r i' (t) is the pulse frequency of bat i' at time t; A8. Update the optimal bat position to the bat position corresponding to the current minimum fitness value, and then determine whether the maximum number of iterations T has been reached. max , if t=T max The ridge corresponding to the current optimal bat position is the starting ridge for sprinkler irrigation in the irregular field; otherwise, set t=t+1 and return to step A3.

7. The method for field path planning of a reel-type sprinkler for reducing soil mechanical compaction according to claim 6, characterized in that: The step A2 of obtaining the adaptability value of each bat is as follows: Among them, C irregular is the geometric center point of the irregular field, j' is the point number in the irregular field, N' is the total number of points in the irregular field, (x j' ,y j' ) are the coordinates of the j'th point in the irregular field.

8. The method for field path planning of a reel-type sprinkler for reducing soil mechanical compaction according to claim 7, characterized in that: In step 3, the irrigation path in the field is planned to obtain the planned field path, and then irrigation is started from the starting ridge, specifically: Step 3. Randomly initialize the positions in M ​​furrows and use the positions in M ​​furrows to randomly generate M' paths R m' =[(x m',1 ,y m',1 ),(x m',2 ,y m',2 ),...,(x m',M ,y m',M )], each path is a firefly; Among them, (x m',1 ,y m',1 ) is the first point in the m'th path, (x m',M ,y m',M ) is the Mth point in the m'th path; Step 32: Obtain the objective function value of each firefly, take the firefly with the minimum objective function value as the optimal firefly, and take the objective function value of the optimal firefly as the fluorescence brightness of the optimal firefly I m' ; Step 3. Get the fluorescence brightness of the two fireflies I m' and I m* , get the fluorescence brightness difference and attraction of the two fireflies: Among them, I m' is the fluorescence brightness of firefly m', I m* is the fluorescence brightness of the firefly m*; Step 3 and 4: According to the fluorescence brightness difference of the fireflies, the brighter firefly between the two fireflies is obtained. Firefly m' moves towards the brighter firefly m*, and the attraction between the two fireflies is used to update the position of firefly m': in, is the position of firefly m' at time t+1, is the position of firefly m' at time t, is the position of the firefly m* at time t, α' is the step size factor, ε'∈[-1,1], ε' is random noise, β m'm* is the attraction between firefly m' and firefly m*; Step 3.5: Randomly perturb the current optimal firefly and update the optimal firefly: in, is the optimal firefly position at time t+1, is the optimal firefly position at time t; Step 36: Obtain the objective function value of the updated firefly, compare the objective function of the updated firefly with the objective function value of the optimal firefly, update the firefly corresponding to the minimum objective function value to the optimal firefly, and use the objective function value of the optimal firefly as the fluorescence brightness of the optimal firefly; Step 37: Determine whether the number of iterations T' has been reached max , if t'=T' max Then output the path corresponding to the optimal firefly as the planned field path; otherwise, set t'=t'+1 and return to step 33; Step 38: Using the starting ridge obtained in step 2 as the starting point, sprinkler irrigation is performed according to the planned field path obtained in step 37.

9. The method for field path planning for a reel-type sprinkler for reducing soil mechanical compaction according to claim 8, characterized in that: The objective function value of each firefly in step 32 is obtained as follows: F(R k )=ω1·D m',total +ω2·C m',compact -ω3·E m',efficiency Among them, ω1 is the total path length weight, ω2 is the compaction weight, ω3 is the sprinkler efficiency weight, D k,total is the total length of the kth path, C k,compact is the compaction degree of the m'th path, E k,efficiency is the sprinkler efficiency of the m'th path, F(R m' ) is the location set in the m'th path, D total,m' is the total length of the m'th path, D row,w is the row distance of the wth operation, d return,w' is the return distance of the w'th operation, W is the total number of operations, compact(l') is the function of whether the l'th ridge is compacted, if compacted, compact(l') is 1, if not compacted, compact(l') is 0, L' is the total number of ridges, C compact,m' is the number of ridges compacted by sprinkler irrigation along the m'th path, E efficiency,m' is the sprinkler efficiency of the m'th path, A cov,m' is the coverage area of ​​the m'th path, n rows is the number of ridges covered in one operation, L is the working width of the convolution disk, and W * is the ridge width, is the floor symbol, A cov,l' is the irrigation area of ​​the l'th ridge, t end,l' is the end time of sprinkler irrigation of the l'th ridge, t start,l' is the start time of sprinkler irrigation for the l'th ridge, R dynamic (t) is the sprinkler radius, t is the time, R0 is the maximum sprinkler radius, v(t) is the variable speed of the reel sprinkler, v max is the maximum speed of the reel sprinkler; If the sprinkler passes over the current ridge during its travel path, it means that the current ridge is compacted.

10. The method for field path planning of a reel-type sprinkler for reducing soil mechanical compaction according to claim 9, characterized in that: The fluorescence brightness of the two fireflies in step 33 is obtained. m' and I m* , get the fluorescence brightness difference and attraction of the two fireflies, specifically: ΔI m'm* =I m' -I m* r m'm* =||X' m' -X' m* || Among them, ΔI m'm* is the fluorescence brightness difference between firefly m' and firefly m*, β m'm* is the attraction between firefly m' and firefly m*, β0 is the maximum attraction, γ' is the attraction attenuation coefficient, r m'm* is the Euclidean distance between firefly m' and firefly m*, X' m' is the position of firefly m', X' m* is the position of firefly m*.

Images