Intelligent dicing path planning method based on bud eye position information

Abstract

The invention discloses an intelligent dicing path planning method based on bud eye position information, relates to the technical field of bud eye dicing planning, and aims to solve the problems of unreasonable cutting position, decreased breeding success rate and waste of cutting resources. Detecting the water content, analyzing a target cutting path, generating a driving signal, determining a dicing area, calculating the bud eye gravity center shift degree and the polymerization degree, and generating the cutting utilization rate. After a driving signal is detected, cutting operation is started, movement and stress data of a cutting tool are collected in real time, the offset rate and the abnormal impedance value are calculated, and whether a path switching mechanism is started or not is judged; if so, detecting the cutting rate, judging adjacent cutting points by combining the cutting utilization rate, and adjusting the path. The cutting utilization rate is calculated by integrating the activity weight, the water content, the bud eye deviation degree and the polymerization degree, and the cutting efficiency is maximized while the crop integrity is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of eyelet cutting planning, and more specifically, to an intelligent cutting path planning method based on eyelet position information. Background Art

[0002] Currently, during the planting and processing of agricultural crops (such as potatoes and sugarcane), there is a demand for pre-cutting of tuber crops to improve breeding efficiency or speed up subsequent processing. With the continuous development of intelligent agricultural equipment, some equipment has begun to introduce visual recognition, force control feedback, laser mapping and other technologies for crop identification and processing path planning.

[0003] The existing technology has the following deficiencies:

[0004] At present, it is impossible to perform fine processing on the structural characteristics of crops, especially the distribution characteristics of buds. There is a lack of active identification and quantitative evaluation of bud information, and the influence of crop moisture content, texture and biological properties on cutting behavior is ignored, resulting in unreasonable cutting positions, reduced breeding success rate and waste of cutting resources. Therefore, an intelligent cutting path planning method based on bud position information is proposed.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent cutting path planning method based on eyelet position information, which solves the problems raised in the above-mentioned background technology by applying three-dimensional image recognition, eyelet activity modeling, moisture content analysis and cutting feedback control fusion mechanism.

[0007] To achieve the above object, the present invention provides the following technical solution: an intelligent segmentation path planning method based on eye position information, comprising the following steps:

[0008] Step S1: Obtain a three-dimensional image of the crop to be cut using a laser radar scanner, find the eye point after pre-processing the three-dimensional image and mark the location, collect attribute data of the marked location, and calculate the activity weight of the marked location based on the attribute data, wherein the attribute data includes eye area point density and eye depth;

[0009] Step S2: Detecting the moisture content of the marked position, analyzing the target cutting path of the crop to be cut based on the activity weight of the marked position and generating a drive signal, determining the cutting area based on the target cutting path, and calculating the eye center of gravity offset and aggregation degree of the cutting area, and calculating the cutting utilization rate of the cutting area by combining the eye center of gravity offset and aggregation degree;

[0010] Step S3: After the driving signal is detected, the cutting operation is performed, the motion data and force data of the cutting tool are collected in real time, and the offset rate and abnormal impedance value are calculated respectively. The offset rate and abnormal impedance value are combined to determine whether the path switching mechanism is activated;

[0011] Step S4: When the path switching mechanism is activated, the cutting rate of the cutting tool is detected, and the adjacent cutting points are determined in combination with the cutting utilization rate of the cutting area, and the target cutting path is adjusted based on the adjacent cutting points.

[0012] In a preferred embodiment, in step S1, obtaining a three-dimensional image of the crop to be cut by a laser radar scanner includes:

[0013] The product of the laser beam transmission speed and the time interval from the laser beam emission to the laser beam reception is divided by 2 to obtain the distance between the laser beam emission point and the pixel point on the surface of the crop to be cut;

[0014] Based on the distance between the laser beam emission point and the pixel point on the surface of the crop to be cut, the three-dimensional coordinates of the pixel point on the surface of the crop to be cut are calculated by trigonometric function;

[0015] The three-dimensional coordinates of the pixel points on the surface of the crop to be cut are processed by the Poisson reconstruction algorithm to obtain a three-dimensional image of the crop to be cut.

[0016] In a preferred embodiment, in step S1, after pre-processing the three-dimensional image, searching for the eye point and marking the position thereof include:

[0017] The three-dimensional image of the crop to be cut is preprocessed by statistical outlier elimination method;

[0018] Traverse each pixel point of the pre-processed 3D image of the crop to be cut as the target point, construct a spatial domain with a preset radius with the target point as the center, and calculate the average Z-axis height of all pixels in the spatial domain;

[0019] The difference between the mean Z-axis height of the target point and the mean Z-axis height of all pixels in the spatial domain is taken as the eye feature value;

[0020] If the eyelet feature value is greater than or equal to 0, the target point is determined to be not an eyelet point;

[0021] If the eyelet feature value is less than 0 and the absolute value of the eyelet feature value is greater than the average Z-axis height of all pixels in the spatial domain, the target point is determined to be the eyelet point and the position is marked.

[0022] In a preferred embodiment, in step S1, the attribute data of the marked position is collected, and the activity weight of the marked position is calculated based on the attribute data, including:

[0023] Traverse all pixel points on the pre-processed three-dimensional image of the crop to be cut, and calculate the Euclidean distance from all pixel points to the eye position;

[0024] If the Euclidean distance is less than or equal to the preset radius and The absolute value of the axis coordinate is greater than the eye position The absolute value of the axis coordinate value determines that the pixel point belongs to the eye area;

[0025] Otherwise, it is determined that the pixel does not belong to the eye area;

[0026] Count the total number of pixels belonging to the eye area as the number of eye area points, and calculate the eye area point density based on the preset radius;

[0027] The absolute value of the eye feature value at the marked position is taken as the eye depth;

[0028] The activity weight of the eye was calculated after normalizing the eye area point density and eye depth using the Max-Min normalization method.

[0029] In a preferred embodiment, in step S2, the moisture content of the marked position is detected, and the target cutting path of the crop to be cut is analyzed in combination with the activity weight of the marked position, including:

[0030] The water content at the marked position is obtained by detecting the marked position with a plant-suitable capacitive moisture sensor;

[0031] The target cutting path of the crop to be cut is obtained based on the A-star algorithm combined with the target cutting path constraint conditions; the target cutting path constraint conditions are:

[0032] If the water content of the marked location is less than or equal to the preset water content threshold and the activity weight of the marked location is less than or equal to the preset activity weight threshold, the marked location is determined to be an impassable area;

[0033] If the water content of the marked position is greater than the preset water content threshold and the activity weight of the marked position is greater than the preset activity weight threshold, the marked position is determined to be a priority selection area;

[0034] If the water content of the marked position is less than or equal to the preset water content threshold and the activity weight of the marked position is greater than the preset activity weight threshold, the marked position is determined to be an avoid-passage area.

[0035] In a preferred embodiment, in step S2, determining the cutting area according to the target cutting path and calculating the eye center of gravity offset and aggregation degree of the cutting area include:

[0036] According to the target cutting path of the crop to be cut, combined with the cutting width and cutting depth of the cutting tool, the portion of the crop to be cut cut by the cutting tool is used as the cutting area;

[0037] The three-dimensional coordinates of the geometric center of the cut area are obtained by averaging the three-dimensional coordinates of each pixel point in the cut area;

[0038] The three-dimensional coordinates of the eyelet gravity center of the cut area are calculated by integrating the three-dimensional coordinates of all eyelet positions in the cut area;

[0039] Calculating the eye gravity center offset of the cut area according to the three-dimensional coordinates of the geometric center of the cut area and the three-dimensional coordinates of the eye gravity center of the cut area;

[0040] The degree of polymerization of the cut area was calculated by the distance between all bud eye positions within the cut area.

[0041] In a preferred embodiment, in step S3, the motion data and force data of the cutting tool are collected in real time and the deflection rate and abnormal impedance value are calculated respectively, including:

[0042] The absolute encoder of the cutting tool monitors the position change of the guide shaft in real time and reads the position change value to convert it into the actual position of the cutting tool. The offset is calculated by combining the actual position of the cutting tool and the ideal position of the cutting tool at the same time, and then normalized to obtain the offset rate.

[0043] The real-time feed speed of the cutting tool is obtained through the servo motor controller, the ratio of the difference between the real-time feed speeds of the cutting tool at two adjacent moments and the real-time feed speed at the previous moment is calculated, and the absolute value is taken to obtain the abnormal impedance value.

[0044] In a preferred embodiment, in step S3, the determination of whether to initiate a path switching mechanism is based on the integrated offset rate and the abnormal impedance value, including:

[0045] The product of the deviation rate and the abnormal impedance value is used as the path switching characteristic value;

[0046] If the path switching characteristic value is greater than the preset path switching threshold, the path switching mechanism is activated;

[0047] If the path switching characteristic value is less than or equal to the preset path switching threshold, the path switching mechanism is not activated.

[0048] In a preferred embodiment, step S4 detects the cutting rate of the cutting tool, determines adjacent cutting points based on the cutting utilization rate of the cutting area, and adjusts the target cutting path based on the adjacent cutting points, including:

[0049] The real-time feed speed of the cutting tool is used as the cutting rate of the cutting tool;

[0050] Taking the current cutting tool position as the center point, obtain the positions of all adjacent cutting points within a preset radius and calculate the distances between the positions of all adjacent cutting points within the preset radius and the current cutting tool position;

[0051] The cutting rate of the cutting tool and the cutting utilization of the cutting area are compared with the preset cutting rate threshold and the preset cutting utilization threshold respectively, and different adjacent cutting points are selected as the next cutting point of the current path to adjust the target cutting path.

[0052] Technical effects and advantages of the present invention:

[0053] The present invention obtains a three-dimensional image of the crop through a laser radar scanner, searches for the eye point after pre-processing and marks the position, collects attribute data to calculate the activity weight, detects the moisture content and analyzes the target cutting path to generate a driving signal, determines the cutting area according to the target path, calculates the eye center of gravity offset and the degree of aggregation, generates the cutting utilization rate, starts the cutting operation after detecting the driving signal, collects the motion data and force data of the cutting tool in real time, calculates the offset rate and the abnormal impedance value, determines whether to enable the path switching mechanism, and if the path switching mechanism is enabled, detects the cutting rate, determines the adjacent cutting points in combination with the cutting utilization rate, and adjusts the target path based on the adjacent cutting points; by calculating the cutting utilization rate in combination with multiple factors such as the activity weight, moisture content, eye center of gravity offset, degree of aggregation, etc., it is possible to maximize the cutting efficiency while ensuring the integrity of the crop, thereby improving the accuracy of the cutting operation and the crop utilization rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a flow chart for implementing the intelligent slicing path planning method based on eye position information of the present invention.

[0055] Figure 2 Schematic diagram of the steps of the intelligent slicing path planning method based on eye position information of the present invention. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0057] The present invention obtains a three-dimensional image of the crop through a laser radar scanner, marks the position after pre-processing, collects attribute data to calculate the activity weight, detects the moisture content and analyzes the target cutting path to generate a driving signal, determines the cutting area according to the target path, calculates the eye center of gravity offset and the degree of aggregation, and generates a cutting utilization rate. After detecting the driving signal, the cutting operation is started, the motion data and force data of the cutting tool are collected in real time, the offset rate and the abnormal impedance value are calculated, and it is determined whether the path switching mechanism is enabled. If the path switching mechanism is enabled, the cutting rate is detected, and the adjacent cutting points are determined in combination with the cutting utilization rate, and the target path is adjusted based on the adjacent cutting points. By calculating the cutting utilization rate based on multiple factors such as the activity weight, moisture content, eye center of gravity offset, and degree of aggregation, the cutting efficiency can be maximized while ensuring the integrity of the crop.

[0058] Example 1, intelligent segmentation path planning method based on eye position information, such as Figures 1 to 2 As shown, the following steps are included:

[0059] Step S1: Obtain a three-dimensional image of the crop to be cut using a laser radar scanner, find the eye point after pre-processing the three-dimensional image and mark the location, collect attribute data of the marked location, and calculate the activity weight of the marked location based on the attribute data;

[0060] Step S2: Detecting the moisture content of the marked position, analyzing the target cutting path of the crop to be cut based on the activity weight of the marked position and generating a drive signal, determining the cutting area based on the target cutting path, and calculating the eye center of gravity offset and aggregation degree of the cutting area, and calculating the cutting utilization rate of the cutting area by combining the eye center of gravity offset and aggregation degree;

[0061] Step S3: After the driving signal is detected, the cutting operation is performed, the motion data and force data of the cutting tool are collected in real time, and the offset rate and abnormal impedance value are calculated respectively. The offset rate and abnormal impedance value are combined to determine whether the path switching mechanism is activated;

[0062] Step S4: When the path switching mechanism is activated, the cutting rate of the cutting tool is detected, and the adjacent cutting points are determined in combination with the cutting utilization rate of the cutting area, and the target cutting path is adjusted based on the adjacent cutting points.

[0063] The specific implementation is as follows:

[0064] In step S1, the laser radar scanner continuously emits a laser beam toward the crop to be cut in a rotating scanning manner. When the laser beam encounters the surface of the crop to be cut, it is reflected and returns to the receiving unit of the laser radar scanner. The receiving unit records the time interval from the emission to the reception of the laser beam. The product of the transmission speed of the laser beam and the time interval is divided by 2 to obtain the distance between the laser beam emission point of the laser radar scanner and the pixel point on the surface of the crop to be cut. The three-dimensional coordinates of the pixel point on the surface of the crop to be cut are calculated by trigonometric functions based on the preset laser radar scanner installation information and the distance between the laser beam emission point of the laser radar scanner and the pixel point on the surface of the crop to be cut.

[0065] It should be explained that the laser radar scanner is a device that determines the position of an object by emitting a laser beam and measuring the time difference of the laser being reflected back, and is used to obtain the three-dimensional coordinates of the surface of the crop to be cut; the rotational scanning method is a scanning method in the laser radar, which is often used to obtain all-round three-dimensional information of the environment; the receiving unit is the core component of the laser radar system, which is used to receive the laser signal reflected from the surface of the object. The preset installation information of the laser radar scanner refers to the installation position, angle, direction and other installation-related parameters of the laser radar sensor in the system, which are set by professionals according to the on-site implementation situation; trigonometric functions are a very important type of function in mathematics, which are mainly used to describe the relationship between angles and right triangles. In the present invention, the sine function is used to calculate the distance between the laser beam emission point of the laser radar scanner and the pixel point on the surface of the crop to be cut and the vertical scanning angle of the laser radar to obtain the z-axis coordinate of the pixel point, and the cosine function is used to calculate the distance between the laser beam emission point of the laser radar scanner and the pixel point on the surface of the crop to be cut, the vertical scanning angle of the laser radar and the horizontal scanning angle of the laser radar to obtain the x-axis and y-axis coordinates of the pixel point;

[0066] The three-dimensional coordinates of the pixel points on the surface of the crop to be cut are processed by the Poisson reconstruction algorithm to obtain a three-dimensional image of the crop to be cut;

[0067] The three-dimensional image of the crop to be cut is preprocessed by statistical outlier elimination method;

[0068] Traverse each pixel point of the pre-processed 3D image of the crop to be cut as the target point, construct a spatial domain with a preset radius with the target point as the center, and calculate the average Z-axis height of all pixels in the spatial domain;

[0069] The difference between the mean Z-axis height of the target point and the mean Z-axis height of all pixels in the spatial domain is taken as the eye feature value;

[0070] If the eyelet feature value is greater than or equal to 0, the target point is determined to be not an eyelet point;

[0071] If the eyelet feature value is less than 0 and the absolute value of the eyelet feature value is greater than the average Z-axis height of all pixels in the spatial domain, the target point is determined to be an eyelet point and the position is marked;

[0072] Collect attribute data of the marked position, including eye field point density and eye depth;

[0073] Traverse all the pixels on the pre-processed 3D image of the crop to be cut and calculate the Euclidean distance from all pixels to the eye position. The specific calculation formula is:

[0074] ;

[0075] in, 、 and The first The three-dimensional coordinates of the pixel points, 、 and is the three-dimensional coordinate of the eye position, is the Euclidean distance;

[0076] Judging by Euclidean distance and preset radius:

[0077] If the Euclidean distance is less than or equal to the preset radius and The absolute value of the axis coordinate is greater than the eye position The absolute value of the axis coordinate value determines that the pixel point belongs to the eye area;

[0078] Otherwise, it is determined that the pixel does not belong to the eye area;

[0079] The total number of pixels belonging to the eye area is counted as the number of eye area points and the eye area point density is calculated based on the preset radius. The specific calculation formula is: ,in is the preset radius, Point to the eye area. is the point density of the eye area;

[0080] The absolute value of the eye feature value at the marked position is taken as the eye depth;

[0081] The eye area point density and eye depth are normalized using the Max-Min normalization method, and the calculation formula is:

[0082] , ;

[0083] in, is the eye area point density, and is the maximum and minimum point density in the eye area, is the eye depth, and is the maximum and minimum eye depth, is the normalized value of the eye area point density, is the normalized value of eye depth;

[0084] The greater the eye point density, the flatter the area around the eye, and the better the growth of the crop to be cut. The greater the eye depth, the farther the eye is from the surface of the crop to be cut, and the worse the growth of the crop to be cut. The activity weight of the eye is calculated by combining the eye point density and eye depth. The calculation formula is: ,in, is the normalized value of the eye area point density, is the normalized value of the eye depth, is the activity weight of the eye;

[0085] It should be explained that the Poisson reconstruction algorithm is an algorithm for generating three-dimensional surfaces from point cloud data, which is used to obtain three-dimensional images of crops to be cut; the statistical outlier removal method is a common method based on statistical principles for identifying and removing outliers from data sets, and is used to pre-process three-dimensional images of crops to be cut; the preset radius is used to construct the spatial domain and is set by professionals; the Max-Min normalization method is a commonly used data preprocessing method, which is used to scale data to a specific range.

[0086] In step S2, the marked position is detected by a plant-suitable capacitive moisture sensor to obtain the water content of the marked position;

[0087] The target cutting path of the crop to be cut is obtained based on the A-star algorithm combined with the target cutting path constraint conditions;

[0088] The target cutting path constraints are:

[0089] If the water content of the marked location is less than or equal to the preset water content threshold and the activity weight of the marked location is less than or equal to the preset activity weight threshold, the marked location is determined to be an impassable area;

[0090] If the water content of the marked position is greater than the preset water content threshold and the activity weight of the marked position is greater than the preset activity weight threshold, the marked position is determined to be a priority selection area;

[0091] If the water content of the marked position is less than or equal to the preset water content threshold and the activity weight of the marked position is greater than the preset activity weight threshold, the marked position is determined to be an avoidance zone;

[0092] It should be explained that the plant-suitable capacitive moisture sensor is a sensor that monitors the moisture content in the soil and is used to obtain the moisture content of the marked location; A-star is a classic algorithm widely used in path planning and graph search problems, and is used to obtain the target cutting path of the crop to be cut; the preset moisture content threshold and activity weight threshold are used to determine the type of marked location and are set by professionals; the impassable area refers to the area that cannot be crossed or passed in the field of path planning; the priority selection area refers to the priority selection of some areas or paths for exploration among multiple available areas or paths based on certain specific standards or goals; the avoidance area refers to the area that should be avoided in path planning, that is, the area that needs to be avoided when performing the task;

[0093] Analyze the target cutting path of the crop to be cut to obtain specific control instructions for the cutting tool, and convert the specific control instructions for the cutting tool into a drive signal through a signal converter;

[0094] It should be noted that the specific analysis process is as follows: the target cutting path of the crop to be cut is converted into the motion path of the cutting tool within the crop area to be cut through a coordinate transformation algorithm; the angle requirement of the cutting tool is obtained based on the geometric characteristics of the target cutting path of the crop to be cut, such as turns and straight segments; finally, the control instruction generation unit generates specific control instructions for the cutting tool based on the motion path of the cutting tool within the crop area to be cut and the angle of the cutting tool;

[0095] According to the target cutting path of the crop to be cut, combined with the cutting width and cutting depth of the cutting tool, the portion of the crop to be cut cut by the cutting tool is used as the cutting area;

[0096] The three-dimensional coordinates of the geometric center of the cut area are obtained by averaging the three-dimensional coordinates of each pixel point in the cut area;

[0097] The three-dimensional coordinates of the eyelet positions in the cut area are integrated to calculate the three-dimensional coordinates of the eyelet gravity center in the cut area. The three-dimensional coordinates of the eyelet gravity center in the cut area are: , , ,in, 、 and In the cutting area The three-dimensional coordinates of the eye position, is the number of buds in the cutting area, 、 and is the three-dimensional coordinate of the eye center of the cutting area;

[0098] The bud eye center of gravity offset of the cut area is calculated based on the three-dimensional coordinates of the geometric center of the cut area and the three-dimensional coordinates of the bud eye center of gravity of the cut area. The calculation formula is:

[0099] ;

[0100] in, 、 and is the three-dimensional coordinate of the eye center of the cutting area, 、 and is the three-dimensional coordinate of the geometric center of the cutting area, 、 、 、 、 and For cutting area axis, Axis and The corresponding maximum and minimum values ​​on the axis, The bud eye center of gravity offset in the cutting area;

[0101] The degree of aggregation of the cut area is calculated by the distance between all eye positions in the cut area. The degree of aggregation of the cut area is: ,in, The subscript is The bud position and subscript are The Euclidean distance of the eye position, is the number of buds in the cutting area, is the degree of polymerization of the cut area;

[0102] The cutting utilization rate of the cutting area is calculated by combining the eye center of gravity deviation and the aggregation degree of the cutting area. The cutting utilization rate of the cutting area is: ,in is the bud eye center of gravity offset in the cutting area, is the degree of polymerization of the cut area, and is the preset weighting coefficient, is the cutting utilization rate of the cutting area;

[0103] The greater the deviation of the eyelet center of gravity in the cutting area, the more uneven the eyelet distribution in the cutting area, and the lower the cutting utilization rate of the cutting area;

[0104] The greater the degree of aggregation of the cutting area, the more concentrated the buds are and the closer the distance between adjacent buds is, the higher the cutting utilization rate of the cutting area is.

[0105] It needs to be explained that the control instruction is a set of instructions required to perform a certain operation. In the intelligent cutting system, the command used to direct the cutting tool to perform a specific task is obtained through the analysis of the target cutting path, the evaluation of crop characteristics and the performance matching of the cutting tool, combined with real-time feedback data. The signal converter is an electronic device that converts one type of signal into another type, and is used to convert the control instruction into a drive signal; the Euclidean distance is a method of measuring the straight-line distance between two points in space; the preset weighting coefficient is used to adjust the influence of the bud eye center of gravity offset and the aggregation degree of the cutting area on the cutting utilization rate of the cutting area, and is set by professionals.

[0106] In step S3, the motion data of the cutting tool refers to the physical parameters used to characterize the spatial motion state of the tool during the cutting operation, including the actual position of the cutting tool;

[0107] The absolute encoder of the cutting tool monitors the position change of the guide rail axis in real time and reads the position change value and converts it into the actual position of the cutting tool;

[0108] The offset is calculated by combining the actual position of the cutting tool and the ideal position of the cutting tool at the same time. The offset is: ,in, 、 and is the actual position coordinate value of the cutting tool, 、 and is the ideal position coordinate value of the cutting tool, is the offset;

[0109] The offset is normalized as the offset rate, which is: ,in, is the offset at time t, is the total number of collection moments, is the deviation rate at time t;

[0110] The force data of the cutting tool refers to the numerical values ​​of various physical forces that the cutting tool is subjected to during the cutting process, including the real-time feed speed of the cutting tool. The real-time feed speed of the cutting tool is obtained through the servo motor controller, and the ratio of the difference between the real-time feed speeds of the cutting tool at two adjacent moments to the real-time feed speed at the previous moment is calculated. The absolute value is taken to obtain the feed speed change rate, and the feed speed change rate is used as the abnormal impedance value. It should be noted that there is a close internal connection between the feed speed and the force data. The speed change rate of the real-time feed speed of the cutting tool is a direct reflection of the force change. The real-time feed speed can be used as a proxy value for the force data to help evaluate the load condition and performance of the tool during the cutting process.

[0111] The product of the deviation rate and the abnormal impedance value is used as the path switching characteristic value and compared with the preset path switching threshold to determine:

[0112] If the path switching characteristic value is greater than the preset path switching threshold, the path switching mechanism is activated;

[0113] If the path switching characteristic value is less than or equal to the preset path switching threshold, the path switching mechanism is not activated;

[0114] It needs to be explained that the absolute encoder of the cutting tool is a device that accurately measures the position of the cutting tool and is used to obtain the actual position of the cutting tool; the preset path switching threshold refers to a limit or standard set in advance in the intelligent control system, which is used to determine when the target path needs to be adjusted or switched, and is set by professionals; the path switching mechanism refers to when the cutting equipment deviates from the predetermined path or encounters an inaccessible area, the system adjusts the path or motion trajectory of the equipment according to certain rules and algorithms to ensure that the task is successfully completed as planned.

[0115] In step S4, the real-time feed speed of the cutting tool is used as the cutting rate of the cutting tool;

[0116] Taking the current cutting tool position as the center point, obtain the positions of all adjacent cutting points within a preset radius and calculate the distances between the positions of all adjacent cutting points within the preset radius and the current cutting tool position;

[0117] comparing a cutting rate of the cutting tool and a cutting utilization rate of the dicing area with a preset cutting rate threshold and a preset cutting utilization rate threshold, respectively;

[0118] If the cutting rate of the cutting tool is greater than a preset cutting rate threshold and the cutting utilization rate of the cutting area is greater than a preset cutting utilization rate threshold, the adjacent cutting point with the largest distance between the adjacent cutting point position and the current cutting tool position is selected as the next cutting point of the current path to adjust the target cutting path;

[0119] If the cutting rate of the cutting tool is less than or equal to a preset cutting rate threshold and the cutting utilization rate of the cutting area is less than or equal to a preset cutting utilization rate threshold, the target cutting path is adjusted by selecting the adjacent cutting point with the smallest distance between the adjacent cutting point position and the current cutting tool position as the next cutting point of the current path;

[0120] When the cutting rate of the cutting tool is high and the cutting utilization rate of the cutting area is high, the cutting tool can advance quickly and can select the farthest adjacent cutting point, thereby reducing the frequency of path adjustment and continuing to advance along the target path;

[0121] When the cutting rate of the cutting tool is low and the cutting utilization rate of the cutting area is low, the progress of the cutting tool is slow, and it is necessary to control the path more accurately, select the nearest adjacent cutting point, and ensure cutting accuracy and efficient utilization of resources by fine-tuning the path;

[0122] It should be explained that the adjacent cutting point refers to the point on the target cutting path that is close to the current cutting position and is about to be cut during the cutting process; the preset cutting rate threshold and the preset cutting utilization threshold are standards for judging whether the cutting operation meets the predetermined requirements, which are set by professionals and will not be elaborated here.

[0123] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0124] Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0125] As used herein, the singular forms "a," "an," and "the" may also include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "include," "comprising," "having," and the like specify the presence of stated features, integers, steps, operations, components, parts, or combinations thereof, but do not preclude the presence or addition of one or more other features, integers, steps, operations, components, parts, or combinations thereof. Furthermore, the term "and / or" as used in this specification includes any and all combinations of the relevant listed items.

[0126] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referenced to each other.

[0127] The above description of the disclosed embodiments will enable those skilled in the art to implement or use various modifications of these embodiments, and it will be apparent to those skilled in the art that the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent segmentation path planning method based on eye position information, characterized by: The following steps are involved: Step S1: Obtain a three-dimensional image of the crop to be cut using a laser radar scanner, find the eye point after pre-processing the three-dimensional image and mark the location, collect attribute data of the marked location, and calculate the activity weight of the marked location based on the attribute data, wherein the attribute data includes eye area point density and eye depth; Step S2: Detecting the moisture content of the marked position, analyzing the target cutting path of the crop to be cut based on the activity weight of the marked position and generating a drive signal, determining the cutting area based on the target cutting path, and calculating the eye center of gravity offset and aggregation degree of the cutting area, and calculating the cutting utilization rate of the cutting area by combining the eye center of gravity offset and aggregation degree; Step S3: After the driving signal is detected, the cutting operation is performed, the motion data and force data of the cutting tool are collected in real time, and the offset rate and abnormal impedance value are calculated respectively. The offset rate and abnormal impedance value are combined to determine whether the path switching mechanism is activated; Step S4: When the path switching mechanism is activated, the cutting rate of the cutting tool is detected, and the adjacent cutting points are determined in combination with the cutting utilization rate of the cutting area, and the target cutting path is adjusted based on the adjacent cutting points.

2. The intelligent segmentation path planning method based on eye position information according to claim 1 is characterized in that: In step S1, a three-dimensional image of the crop to be cut is obtained by a laser radar scanner, including: The product of the laser beam transmission speed and the time interval from the laser beam emission to the laser beam reception is divided by 2 to obtain the distance between the laser beam emission point and the pixel point on the surface of the crop to be cut; Based on the distance between the laser beam emission point and the pixel point on the surface of the crop to be cut, the three-dimensional coordinates of the pixel point on the surface of the crop to be cut are calculated by trigonometric function; The three-dimensional coordinates of the pixel points on the surface of the crop to be cut are processed by the Poisson reconstruction algorithm to obtain a three-dimensional image of the crop to be cut.

3. The intelligent segmentation path planning method based on eye position information according to claim 1 is characterized in that: In step S1, after pre-processing the three-dimensional image, the eye point is found and the position is marked, including: The three-dimensional image of the crop to be cut is preprocessed by statistical outlier elimination method; Traverse each pixel point of the pre-processed 3D image of the crop to be cut as the target point, construct a spatial domain with a preset radius with the target point as the center, and calculate the average Z-axis height of all pixels in the spatial domain; The difference between the mean Z-axis height of the target point and the mean Z-axis height of all pixels in the spatial domain is taken as the eye feature value; If the eyelet feature value is greater than or equal to 0, the target point is determined to be not an eyelet point; If the eyelet feature value is less than 0 and the absolute value of the eyelet feature value is greater than the average Z-axis height of all pixels in the spatial domain, the target point is determined to be the eyelet point and the position is marked.

4. The intelligent segmentation path planning method based on eye position information according to claim 1 is characterized in that: In step S1, attribute data of the marked position is collected, and the activity weight of the marked position is calculated based on the attribute data, including: Traverse all pixel points on the pre-processed three-dimensional image of the crop to be cut, and calculate the Euclidean distance from all pixel points to the eye position; If the Euclidean distance is less than or equal to the preset radius and The absolute value of the axis coordinate is greater than the eye position The absolute value of the axis coordinate value determines that the pixel point belongs to the eye area; Otherwise, it is determined that the pixel does not belong to the eye area; Count the total number of pixels belonging to the eye area as the number of eye area points, and calculate the eye area point density based on the preset radius; The absolute value of the eye feature value at the marked position is taken as the eye depth; The activity weight of the eye was calculated after normalizing the eye area point density and eye depth using the Max-Min normalization method.

5. The intelligent segmentation path planning method based on eye position information according to claim 1 is characterized in that: In step S2, the moisture content of the marked position is detected, and the target cutting path of the crop to be cut is analyzed in combination with the activity weight of the marked position, including: The water content at the marked position is obtained by detecting the marked position with a plant-suitable capacitive moisture sensor; The target cutting path of the crop to be cut is obtained based on the A-star algorithm combined with the target cutting path constraint conditions; the target cutting path constraint conditions are: If the water content of the marked location is less than or equal to the preset water content threshold and the activity weight of the marked location is less than or equal to the preset activity weight threshold, the marked location is determined to be an impassable area; If the water content of the marked position is greater than the preset water content threshold and the activity weight of the marked position is greater than the preset activity weight threshold, the marked position is determined to be a priority selection area; If the water content of the marked position is less than or equal to the preset water content threshold and the activity weight of the marked position is greater than the preset activity weight threshold, the marked position is determined to be an avoid-passage area.

6. The intelligent segmentation path planning method based on eye position information according to claim 1 is characterized in that: In step S2, the cutting area is determined according to the target cutting path and the eye center of gravity offset and aggregation degree of the cutting area are calculated, including: According to the target cutting path of the crop to be cut, combined with the cutting width and cutting depth of the cutting tool, the portion of the crop to be cut cut by the cutting tool is used as the cutting area; The three-dimensional coordinates of the geometric center of the cut area are obtained by averaging the three-dimensional coordinates of each pixel point in the cut area; The three-dimensional coordinates of the eyelet gravity center of the cut area are calculated by integrating the three-dimensional coordinates of all eyelet positions in the cut area; Calculating the eye gravity center offset of the cut area according to the three-dimensional coordinates of the geometric center of the cut area and the three-dimensional coordinates of the eye gravity center of the cut area; The degree of polymerization of the cut area was calculated by the distance between all bud eye positions within the cut area.

7. The intelligent segmentation path planning method based on eye position information according to claim 1 is characterized in that: In step S3, the motion data and force data of the cutting tool are collected in real time and the deflection rate and abnormal impedance value are calculated respectively, including: The absolute encoder of the cutting tool monitors the position change of the guide shaft in real time and reads the position change value to convert it into the actual position of the cutting tool. The offset is calculated by combining the actual position of the cutting tool and the ideal position of the cutting tool at the same time, and then normalized to obtain the offset rate. The real-time feed speed of the cutting tool is obtained through the servo motor controller, the ratio of the difference between the real-time feed speeds of the cutting tool at two adjacent moments and the real-time feed speed at the previous moment is calculated, and the absolute value is taken to obtain the abnormal impedance value.

8. The intelligent segmentation path planning method based on eye position information according to claim 1 is characterized in that: In step S3, the offset rate and the abnormal impedance value are combined to determine whether to enter the path switching mechanism, including: The product of the deviation rate and the abnormal impedance value is used as the path switching characteristic value; If the path switching characteristic value is greater than the preset path switching threshold, the path switching mechanism is activated; If the path switching characteristic value is less than or equal to the preset path switching threshold, the path switching mechanism is not activated.

9. The intelligent segmentation path planning method based on eye position information according to claim 1 is characterized in that: Step S4 detects the cutting rate of the cutting tool, determines adjacent cutting points based on the cutting utilization rate of the cutting area, and adjusts the target cutting path based on the adjacent cutting points, including: The real-time feed speed of the cutting tool is used as the cutting rate of the cutting tool; Taking the current cutting tool position as the center point, obtain the positions of all adjacent cutting points within a preset radius and calculate the distances between the positions of all adjacent cutting points within the preset radius and the current cutting tool position; The cutting rate of the cutting tool and the cutting utilization of the cutting area are compared with the preset cutting rate threshold and the preset cutting utilization threshold respectively, and different adjacent cutting points are selected as the next cutting point of the current path to adjust the target cutting path.

Images