A lateral displacement fertilization control method based on crown layer information perception

Abstract

The application discloses a kind of transverse displacement fertilization control methods based on crown information perception, it is related to agricultural fertilization technical field, based on the sensor installed on double helix fertilizer distributor obtains crop canopy growth information;According to the crop canopy growth information obtained, the water dripping line position of crop is fitted, and the water dripping line accurate curve of crop is extracted.The present application can accurately identify the nutrient requirement of crops through canopy information perception, and provide control strategy for the adjustment of fertilization amount, avoid the waste of fertilizer caused by blind fertilization, appropriately reduce the amount of fertilization in the area of vigorous growth, prevent nutrient excess;In the area of weak growth, increase the amount of fertilization, meet the growth demand of crops, reduce the amount of fertilization, and also reduce soil and water pollution caused by excessive fertilization, which is conducive to the protection of ecological environment, at the same time, by real-time monitoring and adjusting the speed of fertilization, the uniform distribution of fertilizer can be ensured, and the utilization efficiency of fertilizer is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural fertilization, and in particular to a lateral displacement fertilization control method based on canopy information perception. Background Art

[0002] In agronomy, crop growth is closely linked to its nutritional needs. Therefore, understanding crop growth is a prerequisite for precision fertilization. The crop canopy serves as a crucial interface between the crop and the external environment, and its information can reflect the crop's growth, nutritional, and health status. Automatic control technology is key to achieving precision fertilization. Sensors capture real-time crop canopy information, analyze it, and then use automatic control technology to adjust the fertilizer application rate and placement, achieving precise fertilization.

[0003] Fertilization operations require that fertilizer be applied near the crop root system, where the crop absorbs the fertilizer best. Existing fertilizer spreaders mostly apply fertilizer in straight furrows between rows, which affects fertilizer absorption. Studies have shown that the root system at the vertical projection of the drip line along the crop canopy is the place where fertilizer is most actively absorbed. Therefore, how to obtain crop growth data in different areas through the perception of crop canopy information, analyze the growth status of crops, and then determine the area and amount of fertilization is our problem to solve. To this end, a lateral displacement fertilization control method based on canopy information perception is proposed. Summary of the Invention

[0004] The object of the present invention is to provide a lateral displacement fertilization control method based on canopy information perception to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A lateral displacement fertilization control method based on canopy information perception includes the following steps:

[0007] Step 1: obtaining crop canopy growth information based on sensors installed on a double-screw fertilizer spreader;

[0008] Step 2: Based on the acquired crop canopy growth information, the drip line position of the crop is fitted and the precise drip line curve of the crop is extracted. The drip line is the trace left by evaporation of water at the edge of the crop canopy and is usually the area where the root system is most concentrated.

[0009] Step 3, according to the distance from the trunk to the drip line and the crown structure of the crop, a theoretical curve of lateral fertilization position is established by using geometric principles, which provides a theoretical basis for accurate fertilization of the fertilizer applicator, wherein, for fruit trees, the trunk is taken as the center and the lateral distance from the trunk to the drip line of the tree crown is taken as the radius, so as to calculate the most suitable position for fertilization, the theoretical curve of lateral fertilization position represents the position where the fertilizer should be applied to ensure that the fertilizer can be as close to the root system of the crop as possible, thereby improving the absorption efficiency of the fertilizer;

[0010] Step 4, according to the established theoretical curve of lateral fertilization position, the fertilization area is judged, and the operation path of the double-helix fertilizer applicator is planned, when there is a spacing between two fruit trees or crops, the double-helix fertilizer applicator walks in a straight line at the spacing and does not fertilize, and only when it reaches the theoretical fertilization position of the next fruit tree or crop, the fertilization operation continues, thereby avoiding fertilization in the invalid area;

[0011] Step 5, the actual position of the double-helix fertilizer applicator during the fertilization operation is monitored in real time, the actual position is compared with the theoretical curve position, the lateral deviation distance and deviation speed of the fertilizer applicator are adjusted according to the deviation value, the left and right deviation of the double-helix fertilizer applicator is controlled, and the fertilizer applicator is ensured to operate along the theoretical curve;

[0012] Step 6, according to the growth state and fertilization demand of the crop, the fertilization amount and fertilization speed are dynamically adjusted to ensure efficient utilization of the fertilizer, through real-time monitoring of the crop canopy, the nutrient demand and growth state of the crop are analyzed, the fertilization amount is adjusted in real time during the fertilization process, the fertilization amount is reduced in the area where the crop grows vigorously, and the fertilization amount is increased in the area where the crop grows weakly.

[0013] The further improvement of the technical scheme of the present application is that in step 1, the acquisition process of the crop canopy growth information is as follows:

[0014] Step 11, a hyperspectral sensor and a machine vision sensor are deployed on the double-helix fertilizer applicator, so that the sensors can cover most of the area of the crop canopy to obtain comprehensive canopy information, and the sensors are calibrated, and then according to the crop type and growth cycle, the parameters of the sensor sampling frequency and data transmission format are configured to adapt to different working environments and crop types;

[0015] Step 12, when the double-helix fertilizer applicator passes through the crop, the sensor collects image information and spectral information of the crop canopy in real time, obtains the crop canopy growth information including canopy height, canopy width, normalized vegetation index and chlorophyll content, and transmits the collected data to the control system of the double-helix fertilizer applicator in real time;

[0016] Step 13, preprocessing the crop canopy growth information, including denoising and enhancement processing steps, wherein the collected images are denoised and enhanced to improve image quality, reduce the impact of environmental factors on the images, and integrate the preprocessed crop canopy growth information into a unified data set.

[0017] A further improvement of the technical solution of the present invention is that in step 2, the process of extracting the precise curve of the crop drip line is as follows:

[0018] Step 21: Use image processing software to convert the collected color image into a grayscale image, perform binarization on the grayscale image, and divide the pixels in the image into two categories: foreground and background, where the foreground is set to white and the background is set to black;

[0019] Step 22: Processing the binarized image using the Canny edge detection algorithm, using the edge pixels output by the Canny edge detection result to identify the edge of the crop canopy and form the outline of the crop canopy. The Canny algorithm can detect strong and weak edges in the image and improve the accuracy of edge detection by suppressing false edges.

[0020] Step 23: Fit the detected canopy edge using a polynomial curve fitting method to obtain a smooth curve of the canopy contour. Based on the shape and position of the canopy contour curve, a preliminary estimate of the drip line is made. The drip line is usually located inside the canopy edge and is the trace left after water evaporation. Its position can be estimated by the inward offset of the canopy contour.

[0021] Step 24 , using a Bezier curve fitting method to fit the initially estimated drip line position, and using the Bezier curve fitting result to output a smooth and continuous curve to obtain an accurate crop drip line curve.

[0022] A further improvement of the technical solution of the present invention is that the expression of the crop drip line precise curve is:

[0023] ;

[0024] ;

[0025] Where, It is the precise curve of the crop drip line, representing the points on the Bezier curve. is the kth control point, which is a two-dimensional coordinate, and m is the order of the Bezier curve, that is, the number of control points minus one. is a parameter with a value range of [0,1]. is the number of combinations, which means the number of combinations of k elements from m different elements. and Respectively and The power of =0, the starting point of the curve is at ,when =1, the end point of the curve is at ,along with From 0 to 1, the curve starts from the starting point Smooth transition to the end point , the middle control point , ,…, Determines the shape and curvature of the curve.

[0026] A further improvement of the technical solution of the present invention is that in step 3, the process of establishing the theoretical curve of the horizontal fertilization position is as follows:

[0027] Step 31, obtain the precise position of the drip line through image processing and curve fitting method (Bezier curve fitting), and use the positioning sensor to determine the position of the trunk, and record the coordinates of the trunk as ;

[0028] Step 32, for each point on the drip line , calculate its coordinates with the trunk Euclidean distance , to obtain the distance from the trunk to the drip line, and calculate the average distance from all drip line points to the trunk ;

[0029] Step 33: With the trunk as the center and the average distance from all drip line points to the trunk as the radius, draw a circle as the preliminary theoretical curve for horizontal fertilizer application position. The theoretical curve for horizontal fertilizer application position is a circle with the trunk as the center, indicating the position where fertilizer should be applied. Then, according to the parametric equation of the circle, generate the points on the theoretical curve. ;

[0030] Step 34: Fine-tune the fertilization position based on the crop canopy structure and calculate and adjust the fertilization radius based on the canopy structure and soil conditions. If the canopy is denser in some directions, the fertilization radius can be appropriately increased; if the canopy is denser on the north side, the fertilization radius on the north side can be appropriately increased;

[0031] Step 35, adjust the fertilization radius according to and the coordinates of the tree trunk to regenerate the final theoretical curve of horizontal fertilization position.

[0032] A further improvement of the technical solution of the present invention is that in step 4, the process of planning the operating path of the double-spiral fertilizer spreader is:

[0033] Step 41, read the final horizontal fertilization position theoretical curve data, and define the fertilization area, wherein the fertilization area is a circle with the tree trunk as the center and the The circular area with a radius of 1 is calculated, and the inner and outer boundaries of the fertilization area are calculated. The inner boundary radius can be slightly smaller than , the outer boundary radius can be slightly larger than ;

[0034] Step 42, initialize the operation path of the double spiral fertilizer spreader, select a starting point on the inner boundary or outer boundary of the fertilization area as the path starting point of the double spiral fertilizer spreader, plan an operation path of the double spiral fertilizer spreader according to the boundary of the fertilization area, and obtain the coordinate points on it as , where the double helix operation path moves in a spiral along the boundary of the fertilization area to ensure that the entire fertilization area is covered;

[0035] Step 43: When there is a gap between two fruit trees or crops, a gap area is detected and the fertilizer spreader is set to turn off the fertilization function in the straight driving section and automatically turn it on when it reaches the fertilization area. The gap area refers to the blank area between the two fertilization areas. The width of the gap area is determined according to the arrangement density of the fruit trees or crops and the operating width of the fertilizer spreader. The path planning ensures that the fertilizer spreader passes quickly in the gap area without fertilizing.

[0036] Step 44: Connect the operating path of the double-auger fertilizer spreader in the fertilization area and the straight path in the spacing area to generate a final operating path. During fertilization, start the double-auger fertilizer spreader and drive it in a straight line between the crop rows. During the driving process, monitor the position of the double-auger fertilizer spreader in real time to ensure that it moves along the planned fertilization path, and control the opening and closing of the fertilization function based on the position information.

[0037] A further improvement of the technical solution of the present invention is that the expression of the operating path of the double-spiral fertilizer spreader is:

[0038] ;

[0039] ;

[0040] Where, is the coordinate point of the double helix operation path, are the coordinates of the tree trunk, is the adjusted fertilization radius, is the inner border radius adjustment value, Adjust the outer border radius. is a parameter with a value range of [0, T]. T is the total length of the spiral path, which is determined by the size of the fertilization area and the density of the spiral path. From 0 to T, the radius increases from - Gradually increase to + , and Follow the circular path with spiral motions to ensure that the entire fertilizing area is covered.

[0041] A further improvement of the technical solution of the present invention is that in step 5, the process of controlling the left and right swing of the double-helix fertilizer spreader is as follows:

[0042] Step 51: Monitor the position of the double spiral fertilizer spreader in real time using the GPS installed on the double spiral fertilizer spreader to obtain the actual position coordinates of the double spiral fertilizer spreader. ;

[0043] Step 52: Load the theoretical curve data of the horizontal fertilization position and obtain the theoretical position coordinates. , compare the actual position of the fertilizer spreader monitored in real time with the corresponding position on the theoretical curve, calculate the deviation between the actual position and the theoretical position of the fertilizer spreader, and combine the forward speed information of the machine to convert the deviation into lateral deflection distance and deflection speed;

[0044] Step 53: Determine whether the current deviation is within an acceptable range based on a preset deviation threshold. If the deviation exceeds the preset deviation threshold, adjustments are required. If the deviation is within the preset deviation threshold, continue monitoring.

[0045] Step 54, design a PID control algorithm, adjust the lateral swing distance and swing speed of the fertilizer spreader according to the deviation value, and adjust the parameters of the PID controller according to the actual operation situation. 、 and ;

[0046] Step 55: Send a control command to the control system of the twin-screw fertilizer spreader based on the control signal calculated by the PID control algorithm to adjust the lateral swing distance of the twin-screw fertilizer spreader and adjust the swing speed of the fertilizer spreader based on the lateral swing distance. Combined with the calculated lateral swing distance adjustment value and swing speed adjustment value, the lateral swing distance and swing speed of the fertilizer spreader are adjusted in real time to ensure that the fertilizer spreader operates along the theoretical curve.

[0047] Step 56 , monitor the actual position of the fertilizer spreader in real time, compare the actual position with the theoretical position, and continue to adjust the lateral swing distance and swing speed of the fertilizer spreader according to the deviation value to form a closed-loop control.

[0048] A further improvement of the technical solution of the present invention is that the deviation expression between the actual position and the theoretical position of the fertilizer applicator is:

[0049] ;

[0050] ;

[0051] Where, is the deviation between the actual position and the theoretical position of the fertilizer applicator, is the actual position coordinate, is the theoretical position coordinate;

[0052] The expressions of the lateral yaw distance and yaw speed are:

[0053] ;

[0054] ;

[0055] Where, is the lateral deflection distance, is the yaw speed, is the sampling time interval;

[0056] The expression of the PID control algorithm is:

[0057] ;

[0058] Where, is the control signal, is the deviation, 、 and are proportional, integral and differential gains respectively, t is time, and t is a continuous time variable used to represent the deviation at different time points in the control process and control signals ;

[0059] The expressions of the lateral yaw distance adjustment value and the yaw speed adjustment value are:

[0060] ;

[0061] ;

[0062] Where, is the lateral deflection distance adjustment value, is the control signal calculated according to the PID control algorithm, is the yaw speed adjustment value, is the maximum allowable deflection distance.

[0063] A further improvement of the technical solution of the present invention is that in step 6, the process of dynamically adjusting the amount and speed of fertilization is as follows:

[0064] Step 61, calculating a crop status evaluation index based on the collected crop canopy growth information including canopy height, canopy width, normalized difference vegetation index, and chlorophyll content, to assess the nutritional requirements and growth status of the crop;

[0065] Step 62 , based on the evaluation results of the crop's nutritional needs and growth status, analyze the fertilization needs of different areas, identify areas with strong growth and areas with weaker growth, and then calculate the fertilizer amount for each area according to the preset fertilization plan to ensure that fertilizer can be distributed as needed;

[0066] Step 63, based on the calculated fertilizer application amount and fertilizer application speed, a control instruction is sent to the control system of the double-screw fertilizer applicator to adjust the fertilizer application amount of the double-screw fertilizer applicator to ensure accurate fertilizer application;

[0067] Step 64: monitor the actual position of the double-screw fertilizer applicator, as well as the nutrient requirements and growth status of the crop in real time, compare the actual position with the theoretical position, and continue to adjust the fertilizer application rate and speed of the fertilizer applicator based on the deviation value. Through continuous monitoring and adjustment, ensure that the fertilizer applicator always operates according to the preset fertilization plan, thereby improving the accuracy and efficiency of fertilization;

[0068] The expression of the crop status evaluation index is:

[0069] ;

[0070] Where S is the crop status evaluation index, which represents the comprehensive evaluation value of the crop's nutritional needs and growth status; N is the number of crop canopy growth indicators; are the values ​​of the j-th crop canopy growth index, representing canopy height, canopy width, normalized vegetation index, and chlorophyll content, respectively. is the weight of the j-th crop canopy growth index. The value of S ranges from 0 to 1. When the values ​​of all crop canopy growth indicators are 0, S=0. When the values ​​of all crop canopy growth indicators are the maximum, S is close to 1.

[0071] Due to the adoption of the above technical solution, the present invention has the following technical advancements compared to the prior art:

[0072] 1. The present invention provides a lateral displacement fertilization control method based on canopy information perception. Through canopy information perception, the nutritional needs of crops can be accurately identified to avoid fertilizer waste caused by blind fertilization. In areas with vigorous growth, the amount of fertilizer applied is appropriately reduced to prevent nutrient excess. In areas with weaker growth, the amount of fertilizer applied is increased to meet the nutritional needs of crops. This not only improves the utilization rate of fertilizers, but also reduces soil and water pollution caused by excessive fertilization, which is beneficial to protecting the ecological environment. At the same time, by real-time monitoring and adjustment of the fertilization speed, the uniform distribution of fertilizers can be ensured, further improving the utilization efficiency of fertilizers.

[0073] 2. The present invention provides a lateral displacement fertilization control method based on canopy information perception. By real-time collection of crop canopy growth information, including canopy height, canopy width, normalized vegetation index and chlorophyll content, it can accurately reflect the nutritional needs and growth status of crops, and thus provide a basis for dynamically adjusting the amount and location of fertilizer application, ensuring that fertilizer can be accurately applied to crops, thereby improving fertilizer absorption efficiency, reducing fertilizer waste, reducing agricultural production costs, and reducing pollution to the environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0075] Figure 1 is a flow chart of the method of the present invention;

[0076] Figure 2 This is a flow chart of the present invention for dynamically adjusting the amount and speed of fertilizer application;

[0077] Figure 3 This is a schematic diagram of the appearance of the double-screw fertilizer spreader of the present invention. DETAILED DESCRIPTION

[0078] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0079] Example 1, as Figure 1 、 Figure 3As shown, the present invention provides a lateral displacement fertilization control method based on canopy information perception, comprising the following steps:

[0080] Step 1: Obtain crop canopy growth information based on the sensors installed on the double-auger fertilizer spreader. Deploy hyperspectral sensors and machine vision sensors on the double-auger fertilizer spreader to ensure that the sensors can cover most areas of the crop canopy in order to obtain comprehensive canopy information. Calibrate the sensors to ensure their measurement precision and accuracy. Perform spectral calibration on the hyperspectral sensor to ensure that the spectral data it measures is accurate. Perform image calibration on the machine vision sensor, including lens distortion correction and color correction, to improve image quality. Then, configure the sensor sampling frequency and data transmission format parameters according to the crop type and growth cycle. To adapt to different operating environments and crop types, when the double-auger fertilizer spreader passes through the crops, the sensor collects image information and spectral information of the crop canopy in real time, obtains crop canopy growth information including canopy height, canopy width, normalized vegetation index and chlorophyll content, and transmits the collected data to the control system of the double-auger fertilizer spreader in real time, and pre-processes the crop canopy growth information, including the steps of denoising and enhancing, wherein the collected images are denoised and enhanced to improve image quality and reduce the impact of environmental factors on the images, and the pre-processed crop canopy growth information is integrated into a unified data set;

[0081] Step 2: Based on the acquired crop canopy growth information, the drip line position of the crop is fitted and the precise curve of the crop drip line is extracted. The drip line is the trace left after the water evaporates at the edge of the crop canopy. It is usually also the area where the root system is more concentrated. The collected color image is converted into a grayscale image using image processing software. The grayscale image is binarized and the pixels in the image are divided into foreground and background. The foreground is set to white and the background is set to black. The binarized image is processed using the Canny edge detection algorithm. The edge pixels output by the Canny edge detection result are used to identify the edge of the crop canopy and form the outline of the crop canopy. The nny algorithm can detect strong and weak edges in an image and improve the accuracy of edge detection by suppressing false edges. The detected canopy edge is fitted using a polynomial fitting curve fitting method to obtain a smooth curve of the canopy contour. The drip line position is preliminarily estimated based on the shape and position of the canopy contour curve. The drip line is usually located on the inner side of the canopy edge and is the trace left after water evaporation. Its position can be estimated by the inward offset of the canopy contour. The Bezier curve fitting method is used to fit the preliminarily estimated drip line position. The Bezier curve fitting result is used to output a smooth and continuous curve to obtain the precise drip line curve of the crop.

[0082] Further, the expression of the accurate crop drip line curve is:

[0083] ;

[0084] ;

[0085] In the formula, is the accurate crop drip line curve, represents a point on the Bezier curve, is the kth control point, which is a two-dimensional coordinate, m is the order of the Bezier curve, i.e., the number of control points minus one, is a parameter, taking a value in the range [0, 1], is the combination number, representing the number of combinations of taking k elements from m different elements, and represent the power of and respectively, when = 0, the starting point of the curve is located at , when = 1, the end point of the curve is located at , as increases from 0 to 1, the curve smoothly transitions from the starting point to the end point , and the intermediate control points , , determine the shape and curvature of the curve;

[0086] Step 3: Based on the distance from the trunk to the drip line and the canopy structure of the crop, a theoretical curve for lateral fertilizer application is established using geometric principles, providing a theoretical basis for accurate fertilizer application by the fertilizer applicator. For fruit trees, the trunk is taken as the center and the lateral distance from the trunk to the drip line of the tree canopy is taken as the radius to calculate the optimal fertilizer application position. The theoretical curve for lateral fertilizer application represents the position where the fertilizer should be applied to ensure that the fertilizer is as close as possible to the crop's root system, improving fertilizer uptake efficiency. The accurate position of the drip line is obtained through image processing and curve fitting methods (Bezier curve fitting), and the position of the trunk is determined using a positioning sensor. The coordinates of the trunk are recorded as For each point on the drip line, the Euclidean distance from the trunk coordinates is calculated to obtain the distance from the trunk to the drip line, and the average distance of all drip line points to the trunk is calculated as where the distance from the trunk to the drip line is calculated as: and the average distance of all drip line points to the trunk is calculated as: ​, n is the number of points on the drip line; with the trunk as the center and the average distance from all drip line points to the trunk as the radius, draw a circle as the preliminary horizontal fertilizer position theoretical curve. The horizontal fertilizer position theoretical curve is a circle with the trunk as the center, indicating the position where fertilizer should be applied. Then, according to the parametric equation of the circle, the points on the theoretical curve are generated. ,in, , , is a parameter with a value range of [0, 2π]. It fine-tunes the fertilization position according to the crop canopy structure and calculates and adjusts the fertilization radius according to the canopy structure and soil conditions. If the canopy is denser in some directions, the fertilization radius can be appropriately increased. If the canopy is denser on the north side, the fertilization radius on the north side can be appropriately increased. The expression for adjusting the fertilization radius is: , It is a distance adjusted according to the canopy structure and soil conditions. It can be a small positive or negative number. If the canopy is denser on the north side, you can set A positive number, according to which the fertilization radius is adjusted and the coordinates of the trunk to regenerate the final theoretical curve of horizontal fertilization position. The point on the final theoretical curve of horizontal fertilization position is ;in, , ;

[0087] Step 4: Based on the established theoretical curve of horizontal fertilization position, the fertilization area is determined and the operation path of the double-spiral fertilizer spreader is planned. When there is a gap between two fruit trees or crops, the double-spiral fertilizer spreader moves in a straight line at the gap without fertilizing. Fertilization will not continue until it reaches the theoretical fertilization position of the next fruit tree or crop, thus avoiding fertilizing in an ineffective area.

[0088] Step 5: Real-time monitoring of the actual position of the double-screw fertilizer spreader during the fertilization operation, comparison of the actual position with the theoretical curve position, adjustment of the lateral swing distance and swing speed of the fertilizer spreader according to the deviation value, control of the left and right swing of the double-screw fertilizer spreader, and ensuring that the fertilizer spreader operates along the theoretical curve;

[0089] Step 6: Dynamically adjust the amount and speed of fertilizer application based on the crop's growth status and fertilizer needs to ensure efficient fertilizer utilization. By real-time monitoring of the crop canopy, analyze the crop's nutritional needs and growth status. During the fertilization process, adjust the amount of fertilizer in real time, reduce the amount of fertilizer in areas with vigorous crop growth, and increase the amount of fertilizer in areas with weaker growth.

[0090] Example 2, as Figure 2 、 Figure 3As shown, based on Example 1, the present invention provides a technical solution: Preferably, in step 4, the process of planning the operating path of the double-spiral fertilizer spreader is:

[0091] Read the final horizontal fertilization position theoretical curve data and define the fertilization area, where the fertilization area is the circle with the tree trunk as the center and the The circular area with a radius of 1 is calculated, and the inner and outer boundaries of the fertilization area are calculated. The inner boundary radius can be slightly smaller than , the outer boundary radius can be slightly larger than , initialize the operation path of the double spiral fertilizer spreader, select a starting point on the inner or outer boundary of the fertilization area as the path starting point of the double spiral fertilizer spreader, and plan an operation path of the double spiral fertilizer spreader according to the boundary of the fertilization area, and obtain the coordinate points on it as , wherein the double-helix operation path moves in a spiral along the boundary of the fertilization area to ensure that the entire fertilization area is covered. When there is a gap between two fruit trees or crops, the gap area is detected, and the fertilizer spreader is set to turn off the fertilization function in the straight driving section and automatically turn it on when it reaches the fertilization area. The gap area refers to the blank area between the two fertilization areas. The width of the gap area is determined according to the arrangement density of the fruit trees or crops and the operation width of the fertilizer spreader. The path planning ensures that the fertilizer spreader passes quickly in the gap area without performing fertilization operations. The double-helix fertilizer spreader operation path in the fertilization area and the straight path of the gap area are connected to generate the final operation path. During the fertilization operation, the double-helix fertilizer spreader is started to travel in a straight line between the crop rows. During the driving process, the position of the double-helix fertilizer spreader is monitored in real time to ensure that it travels along the planned fertilization path, and the opening and closing of the fertilization function is controlled according to the position information;

[0092] Furthermore, the expression of the double-spiral fertilizer spreader operation path is:

[0093] ;

[0094] ;

[0095] Where, is the coordinate point of the double helix operation path, are the coordinates of the tree trunk, is the adjusted fertilization radius, is the inner border radius adjustment value, Adjust the outer border radius. is a parameter with a value range of [0, T]. T is the total length of the spiral path, which is determined by the size of the fertilization area and the density of the spiral path. From 0 to T, the radius increases from - Gradually increase to + , and Perform spiral movements along the circular path to ensure that the entire fertilization area is covered;

[0096] In step 5, the process of controlling the left and right swing of the double-screw fertilizer spreader is as follows:

[0097] According to the GPS installed on the double spiral fertilizer spreader, the position information of the double spiral fertilizer spreader is monitored in real time to obtain the actual position coordinates of the double spiral fertilizer spreader. , load the theoretical curve data of the horizontal fertilization position and obtain the theoretical position coordinates , compare the actual position of the fertilizer spreader monitored in real time with the corresponding position on the theoretical curve, calculate the deviation between the actual position and the theoretical position of the fertilizer spreader, and combine the forward speed information of the machine to convert the deviation into lateral swing distance and swing speed. According to the preset deviation threshold, judge whether the current deviation is within the acceptable range. If the deviation exceeds the preset deviation threshold, it needs to be adjusted. If the deviation is within the preset deviation threshold, continue monitoring, design a PID control algorithm, adjust the lateral swing distance and swing speed of the fertilizer spreader according to the deviation value, and adjust the parameters of the PID controller according to the actual operation situation. 、 and To ensure the control effect, if the swing distance is large, the speed can be appropriately reduced to adjust the position more accurately. If the swing distance is small, the speed can be maintained or appropriately increased to improve the working efficiency. According to the control signal calculated by the PID control algorithm, a control instruction is sent to the control system of the double-spiral fertilizer spreader to adjust the lateral swing distance of the double-spiral fertilizer spreader, and according to the lateral swing distance, the swing speed of the fertilizer spreader is adjusted. Combined with the calculated lateral swing distance adjustment value and the swing speed adjustment value, the lateral swing distance and swing speed of the fertilizer spreader are adjusted in real time to ensure that the fertilizer spreader operates along the theoretical curve. The actual position of the fertilizer spreader is monitored in real time, the actual position is compared with the theoretical position, and the lateral swing distance and swing speed of the fertilizer spreader are further adjusted according to the deviation value to form a closed-loop control.

[0098] Furthermore, the deviation expression between the actual position and the theoretical position of the fertilizer applicator is:

[0099] ;

[0100] ;

[0101] Where, is the deviation between the actual position and the theoretical position of the fertilizer applicator, is the actual position coordinate, is the theoretical position coordinate;

[0102] The expression of the lateral deviation distance and the deviation speed is:

[0103] ;

[0104] ;

[0105] In the formula, is the lateral deviation distance, is the deviation speed, is the sampling time interval;

[0106] The expression of the PID control algorithm is:

[0107] ;

[0108] In the formula, is the control signal, is the deviation, , and are the proportional, integral and differential gains respectively, t is the time, and t is a continuous time variable used to represent the deviation at different time points in the control process and the control signal ;

[0109] The expression of the lateral deviation distance adjustment value and the deviation speed adjustment value is:

[0110] ;

[0111] ;

[0112] In the formula, is the lateral deviation distance adjustment value, is the control signal calculated according to the PID control algorithm, is the deviation speed adjustment value, is the maximum allowed deviation distance;

[0113] In step 6, the process of dynamically adjusting the fertilizer amount and the fertilization speed is:

[0114] In step 61, according to the collected crop canopy growth information including canopy height, canopy width, normalized vegetation index and chlorophyll content, the crop state evaluation index is calculated to evaluate the nutrient demand and growth state of the crop;

[0115] In step 62, according to the evaluation results of the crop nutrient demand and growth state, the fertilizer demand of different regions is analyzed, the regions with vigorous growth and the regions with weak growth are identified, and then according to the preset fertilization scheme, the fertilizer amount of each region is calculated to ensure that the fertilizer can be distributed as needed;

[0116] Step 63, based on the calculated fertilizer application amount and fertilizer application speed, a control instruction is sent to the control system of the double-screw fertilizer applicator to adjust the fertilizer application amount of the double-screw fertilizer applicator to ensure accurate fertilizer application;

[0117] Step 64: monitor the actual position of the double-screw fertilizer applicator, as well as the nutrient requirements and growth status of the crop in real time, compare the actual position with the theoretical position, and continue to adjust the fertilizer application rate and speed of the fertilizer applicator based on the deviation value. Through continuous monitoring and adjustment, ensure that the fertilizer applicator always operates according to the preset fertilization plan, thereby improving the accuracy and efficiency of fertilization;

[0118] Among them, the expression of crop status evaluation index is:

[0119] ;

[0120] Where S is the crop status evaluation index, which represents the comprehensive evaluation value of the crop's nutritional needs and growth status; N is the number of crop canopy growth indicators; are the values ​​of the j-th crop canopy growth index, representing canopy height, canopy width, normalized vegetation index, and chlorophyll content, respectively. is the weight of the j-th crop canopy growth index. The value range of S is between 0 and 1. When the values ​​of all crop canopy growth indexes are 0, S=0. When the values ​​of all crop canopy growth indexes are at their maximum, S is close to 1. As the crop canopy growth index increases, the comprehensive evaluation value S also increases, but the rate of increase gradually slows down. If the value of S is low, it means that the nutritional demand and growth status of the crop are poor, and the amount of fertilizer needs to be increased. If the value of S is high, it means that the nutritional demand and growth status of the crop are good, and the amount of fertilizer can be appropriately reduced.

[0121] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A lateral displacement fertilization control method based on canopy information perception, characterized in that: The following steps are involved: Step 1: obtaining crop canopy growth information based on sensors installed on a double-screw fertilizer spreader; In step 1, the process of acquiring crop canopy growth information is as follows: Step 11: Deploy a hyperspectral sensor and a machine vision sensor on the double-auger fertilizer spreader, calibrate the sensors, and then configure the sensor sampling frequency and data transmission format parameters according to the crop type and growth cycle to adapt to different operating environments and crop types; Step 12: When the double-auger fertilizer spreader passes over the crops, the sensor collects image information and spectral information of the crop canopy in real time, obtains crop canopy growth information including canopy height, canopy width, normalized vegetation index, and chlorophyll content, and transmits the collected data to the control system of the double-auger fertilizer spreader in real time; Step 13, preprocessing the crop canopy growth information, including denoising and enhancement processing steps, and integrating the preprocessed crop canopy growth information into a unified data set; Step 2: Based on the acquired crop canopy growth information, the drip line position of the crop is fitted and the precise drip line curve of the crop is extracted; In step 2, the process of extracting the precise curve of the crop drip line is as follows: Step 21: Use image processing software to convert the collected color image into a grayscale image, perform binarization on the grayscale image, and divide the pixels in the image into two categories: foreground and background, where the foreground is set to white and the background is set to black; Step 22, using the Canny edge detection algorithm to process the binarized image, using the edge pixels output by the Canny edge detection result to identify the edge of the crop canopy and form the outline of the crop canopy; Step 23, using a polynomial curve fitting method to fit the detected canopy edge to obtain a smooth curve of the canopy contour, and preliminarily estimating the position of the drip line based on the shape and position of the canopy contour curve; Step 24, using a Bezier curve fitting method to fit the initially estimated drip line position, and using the Bezier curve fitting result to output a smooth and continuous curve to obtain an accurate crop drip line curve; Step 3: Based on the distance from the tree trunk to the drip line and the canopy structure of the crop, a theoretical curve for lateral fertilization position is established using geometric principles; Step 4: Determine the fertilization area based on the established theoretical curve of horizontal fertilization position and plan the operation path of the double-screw fertilizer spreader; Step 5: Real-time monitoring of the actual position of the double-screw fertilizer spreader during the fertilization operation, comparison of the actual position with the theoretical curve position, and adjustment of the lateral swing distance and swing speed of the fertilizer spreader according to the deviation value to control the left and right swing of the double-screw fertilizer spreader; Step 6: Dynamically adjust the amount and speed of fertilizer application based on the growth status and fertilizer requirements of the crops.

2. The method for controlling lateral displacement fertilization based on canopy information perception according to claim 1, characterized in that: The expression of the crop drip line precise curve is: ； ； Where, It is the precise curve of the crop drip line, representing the points on the Bezier curve. is the kth control point, which is a two-dimensional coordinate, and m is the order of the Bezier curve, that is, the number of control points minus one. is a parameter with a value range of [0,1]. is the number of combinations, which means the number of combinations of k elements from m different elements. and Respectively and The power of =0, the starting point of the curve is at ,when =1, the end point of the curve is at ,along with From 0 to 1, the curve starts from the starting point Smooth transition to the end point .

3. The method for controlling lateral displacement fertilization based on canopy information perception according to claim 2, characterized in that: In step 3, the process of establishing the theoretical curve of the horizontal fertilization position is as follows: Step 31, obtain the precise position of the drip line through image processing and curve fitting method, and use the positioning sensor to determine the position of the trunk, and record the coordinates of the trunk as ; Step 32, for each point on the drip line , calculate its coordinates with the trunk Euclidean distance , to obtain the distance from the trunk to the drip line, and calculate the average distance from all drip line points to the trunk , where the distance from the trunk to the drip line is calculated as: , the average distance from all drip line points to the tree trunk is calculated as: , n is the number of points on the drip line; Step 33: With the trunk as the center and the average distance from all drip line points to the trunk as the radius, draw a circle as the preliminary theoretical curve for horizontal fertilization position, and then generate the points on the theoretical curve according to the parametric equation of the circle. ; Step 34: Fine-tune the fertilization position based on the crop canopy structure and calculate and adjust the fertilization radius based on the canopy structure and soil conditions. ; Step 35, adjust the fertilization radius according to and the coordinates of the tree trunk to regenerate the final theoretical curve of horizontal fertilization position.

4. The method for controlling lateral displacement fertilization based on canopy information perception according to claim 3, characterized in that: In step 4, the process of planning the operation path of the double-screw fertilizer spreader is as follows: Step 41, read the final horizontal fertilization position theoretical curve data, and define the fertilization area, wherein the fertilization area is a circle with the tree trunk as the center and the The circular area with a radius of 1 is calculated, and the inner and outer boundaries of the fertilization area are calculated; Step 42, initialize the operation path of the double spiral fertilizer spreader, select a starting point on the inner boundary or outer boundary of the fertilization area as the path starting point of the double spiral fertilizer spreader, plan an operation path of the double spiral fertilizer spreader according to the boundary of the fertilization area, and obtain the coordinate points on it as ; Step 43: When there is a gap between two fruit trees or crops, a gap area is detected, and the fertilizer spreader is set to turn off the fertilization function in the straight driving section and automatically turn it on when it reaches the fertilization area. The gap area refers to the blank area between the two fertilization areas, and the width of the gap area is determined according to the arrangement density of the fruit trees or crops. Step 44: Connect the operating path of the double-auger fertilizer spreader in the fertilization area and the straight path in the spacing area to generate a final operating path. During fertilization, start the double-auger fertilizer spreader and drive it in a straight line between the crop rows. During the driving process, monitor the position of the double-auger fertilizer spreader in real time to ensure that it moves along the planned fertilization path, and control the opening and closing of the fertilization function based on the position information.

5. The method for controlling lateral displacement fertilization based on canopy information perception according to claim 4, characterized in that: The expression of the double-spiral fertilizer spreader operation path is: ； ； Where, is the coordinate point of the double helix operation path, are the coordinates of the tree trunk, is the adjusted fertilization radius, is the inner border radius adjustment value, Adjust the outer border radius. is a parameter whose value range is [0,T], T is the total length of the spiral path, as From 0 to T, the radius increases from - Gradually increase to + .

6. The method for controlling lateral displacement fertilization based on canopy information perception according to claim 5, characterized in that: In step 5, the process of controlling the left and right swing of the double-helix fertilizer spreader is as follows: Step 51: Monitor the position of the double spiral fertilizer spreader in real time using the GPS installed on the double spiral fertilizer spreader to obtain the actual position coordinates of the double spiral fertilizer spreader. ; Step 52: Load the theoretical curve data of the horizontal fertilization position and obtain the theoretical position coordinates. , compare the actual position of the fertilizer spreader monitored in real time with the corresponding position on the theoretical curve, calculate the deviation between the actual position and the theoretical position of the fertilizer spreader, and combine the forward speed information of the machine to convert the deviation into lateral deflection distance and deflection speed; Step 53: Determine whether the current deviation is within an acceptable range based on a preset deviation threshold. If the deviation exceeds the preset deviation threshold, adjustments are required. If the deviation is within the preset deviation threshold, continue monitoring. Step 54, design a PID control algorithm, adjust the lateral swing distance and swing speed of the fertilizer spreader according to the deviation value, and adjust the parameters of the PID controller according to the actual operation situation. 、 and ; Step 55: Send a control command to the control system of the twin-screw fertilizer spreader based on the control signal calculated by the PID control algorithm to adjust the lateral swing distance of the twin-screw fertilizer spreader and adjust the swing speed of the fertilizer spreader based on the lateral swing distance. Combined with the calculated lateral swing distance adjustment value and swing speed adjustment value, the lateral swing distance and swing speed of the fertilizer spreader are adjusted in real time. Step 56 , monitor the actual position of the fertilizer spreader in real time, compare the actual position with the theoretical position, and continue to adjust the lateral swing distance and swing speed of the fertilizer spreader according to the deviation value to form a closed-loop control.

7. The method for controlling lateral displacement fertilization based on canopy information perception according to claim 6, characterized in that: The deviation expression between the actual position and the theoretical position of the fertilizer applicator is: ； ； Where, is the deviation between the actual position and the theoretical position of the fertilizer applicator, is the actual position coordinate, is the theoretical position coordinate; The expressions of the lateral yaw distance and yaw speed are: ； ； Where, is the lateral deflection distance, is the yaw speed, is the sampling time interval; The expressions of the lateral yaw distance adjustment value and the yaw speed adjustment value are: ; ； Where, is the lateral deflection distance adjustment value, is the control signal calculated according to the PID control algorithm, is the yaw speed adjustment value, is the maximum allowable deflection distance.

8. The method for controlling lateral displacement fertilization based on canopy information perception according to claim 7, characterized in that: In step 6, the process of dynamically adjusting the amount and speed of fertilization is as follows: Step 61, calculating a crop status evaluation index based on the collected crop canopy growth information including canopy height, canopy width, normalized difference vegetation index, and chlorophyll content, to assess the nutritional requirements and growth status of the crop; Step 62 , based on the evaluation results of the crop's nutritional needs and growth status, analyze the fertilization requirements of different regions, identify areas with vigorous growth and areas with weaker growth, and then calculate the fertilizer amount for each region according to a preset fertilization plan; Step 63, sending a control instruction to the control system of the double-screw fertilizer applicator based on the calculated fertilizer application amount and fertilizer application speed to adjust the fertilizer application amount of the double-screw fertilizer applicator; Step 64: monitor the actual position of the double-screw fertilizer applicator, as well as the nutritional needs and growth status of the crop in real time, compare the actual position with the theoretical position, and continue to adjust the fertilizer application rate and speed of the fertilizer applicator based on the deviation value; The expression of the crop status evaluation index is: ； Where S is the crop status evaluation index, N is the number of crop canopy growth indicators, are the values ​​of the j-th crop canopy growth index, representing canopy height, canopy width, normalized vegetation index, and chlorophyll content, respectively. is the weight of the j-th crop canopy growth index. The value of S ranges from 0 to 1. When the values ​​of all crop canopy growth indicators are 0, S=0. When the values ​​of all crop canopy growth indicators are the maximum, S is close to 1.

Images