Mushroom reciprocating pushing non-destructive picking method and path planning method thereof

Abstract

The application discloses a mushroom reciprocating pushing nondestructive picking mode and a path planning method thereof, and comprises the following steps: A. designing a reciprocating pushing picking mode for separating mushrooms from a culture medium; B. determining a nondestructive picking optimal pushing direction of mature mushrooms; and C. a double-target optimization mature mushroom picking path planning algorithm considering the nondestructive picking pushing direction and an optimal path. The method optimizes the mushroom picking path problem. The application realizes the optimization of the multi-target problem after the combination of the mushroom reciprocating pushing picking and the optimal path, and through the determination of the nondestructive picking direction and the picking path planning, the damage to the current to-be-picked mushroom and the surrounding mushrooms during the picking of the mushroom is avoided, so that the success rate of the nondestructive picking of the mushroom is effectively improved, and the picking efficiency is improved. The application is not only limited to the picking of edible fungi such as mushrooms and straw mushrooms, but also suitable for the directional picking and trajectory planning of other kinds of spherical and cluster fruits.

Description

Technical Field

[0001] This invention relates to the field of harvesting machinery technology, particularly to the field of mushroom harvesting, and specifically to a non-destructive mushroom harvesting method using a reciprocating pushing mechanism and its path planning method. Background Technology

[0002] With the continuous development of mushroom cultivation technology, mushroom cultivation has evolved into factory-style production. Because mushroom harvesting has a long cycle and requires a significant investment of labor to ensure timely harvesting, it is the most time-consuming and arduous link in the entire production chain. Manual labor is currently the primary harvesting method. Mushroom harvesting is physically demanding and takes place in humid environments, causing substantial depletion of the harvesters' energy and impacting their health. Furthermore, the shortage of young and middle-aged laborers in rural areas has made traditional manual mushroom harvesting a major bottleneck restricting the development of the mushroom cultivation industry. Therefore, automation of mushroom harvesting is urgently needed.

[0003] In existing automated mushroom harvesting methods, mushroom picking mainly involves using an end effector to grip or hold the cap and then rotate it to separate the fruiting body from the culture medium. However, field experiments have shown that, on the one hand, this rotational separation method is prone to separation between the stipe and cap (i.e., root breakage) when the bond between the fruiting body and the culture medium is strong. On the other hand, due to the anisotropy and uneven density of mushroom growth, multiple mushrooms may adhere and stick together around a single mushroom. In such cases, picking by gripping or rotating is difficult to adapt to the clustered growth of mushrooms and easily damages surrounding mushrooms, seriously affecting the quality of harvesting.

[0004] Therefore, intelligent and non-destructive mushroom harvesting has become a key technical problem that urgently needs to be solved for fruit and vegetable harvesting robots. It is necessary to propose a non-destructive mushroom harvesting method using reciprocating propulsion and its path planning mechanism. Summary of the Invention

[0005] The purpose of this invention is to address the problem of high fruiting body damage rate in existing automated intelligent mushroom harvesting technologies by providing a non-destructive mushroom harvesting method and its path planning method. This method effectively improves the success rate of non-destructive harvesting of mature mushrooms, especially clustered mushrooms, and optimizes the harvesting path length to increase harvesting efficiency, thereby achieving non-destructive mushroom harvesting while obtaining the optimal harvesting path.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows:

[0007] A non-destructive mushroom harvesting method using a reciprocating pushing mechanism and its path planning method mainly include:

[0008] A: Design a reciprocating pushing harvesting method to separate mushrooms from the culture medium;

[0009] B: Method for determining the optimal pushing direction for non-destructive harvesting of mature mushrooms;

[0010] C: A dual-objective optimization algorithm for mature mushroom harvesting path planning, considering both the direction of non-destructive harvesting and the optimal path.

[0011] Preferably, a non-destructive mushroom harvesting method using reciprocating pushing and its path planning method specifically includes:

[0012] A: Design a reciprocating pushing harvesting method to separate mushrooms from the culture medium;

[0013] Step A1: Use an end effector to grip or adhere to the mushroom;

[0014] Step A2: Push the mushroom back and forth along the optimal pushing direction to separate the mushroom from the culture medium;

[0015] B: Method for determining the optimal pushing direction for non-destructive harvesting of mature mushrooms;

[0016] Step B1: Obtain the center coordinates of the mushrooms, the radius of the mushroom cap, and the height of the mushrooms from the mushroom distribution image. Save this data as mushroom data features.

[0017] Step B2: Use mushroom data features to search for neighboring mushrooms and next-nearest mushrooms around the target mushroom;

[0018] Step B3: Calculate the solution set of the neighboring mushrooms and the solution set of the next neighboring mushrooms, and determine the optimal direction for pushing the mushrooms;

[0019] C: A dual-objective optimization algorithm for mature mushroom harvesting path planning, considering both the direction of non-destructive harvesting and the optimal path;

[0020] Step C1: Construct a multi-objective optimization model for mushroom harvesting routes:

[0021] Step C1.1: Construct a mushroom reciprocating push picking failure rate function and calculate the picking failure rate of the picking sequence;

[0022] Step C1.1.1: Judging the reciprocating movement of the target mushroom for picking;

[0023] Step C1.1.2: Calculate the harvesting failure rate of the mushroom harvesting sequence;

[0024] Step C1.2: Construct a mushroom path length calculation function to calculate the path length of the picking sequence;

[0025] Step C1.3: Establish a multi-objective optimization model based on the established mushroom reciprocating push picking failure rate function and path length function;

[0026] Step C2: After the model is built in step C1, the improved NSGA-II algorithm is used to solve the model to obtain the Pareto non-dominated set;

[0027] Step C2.1: Individual chromosome encoding within the population;

[0028] Step C2.2: Initialize the population;

[0029] Step C2.3: Inject the extreme value of the single-target picking path length into the initial population and calculate p for all individuals in the population. failure dist sum Function value;

[0030] Step C2.4: Select the parent population using an elite strategy;

[0031] Step C2.5: Selection, crossover, and mutation operations:

[0032] Step C2.6: Merging parent and child populations and deduplication;

[0033] Step C2.7: Select the offspring population using an elitist strategy incorporating a cyclic crowding sorting algorithm;

[0034] Step C2.7.1: Connect the two boundary individuals 1 in the non-dominated hierarchy. d n d The crowding level is set to inf. Individuals in the non-dominated hierarchy are sorted from highest to lowest crowding level, and the individual with the lowest crowding level is deleted.

[0035] Step C2.7.2: Recalculate the crowding degree of the remaining individuals in this level, and delete the individual with the lowest crowding degree again;

[0036] Step C2.7.3: Iterate sequentially until a specified number of solutions remain.

[0037] Step C2.8: Repeat steps C2.5 to C2.8 until the target genetic generation, then exit the current loop;

[0038] Step C2.9: Select the optimal solution from the non-dominated solution set according to the weight coefficients and output it.

[0039] Preferably, step A specifically includes:

[0040] Step A1: Use an end effector to grip or adhere to the mushroom;

[0041] Step A2: Push the mushroom back and forth along the optimal pushing direction to separate the mushroom from the culture medium.

[0042] Preferably, the optimal pushing direction for mature mushrooms in step A2 is determined based on the method used in step B for determining the optimal pushing direction for non-destructive harvesting of mature mushrooms. This method specifically includes:

[0043] Step B1: Obtain the center coordinates of the mushrooms, the radius of the mushroom cap, and the height of the mushrooms from the mushroom distribution image. Save this data as mushroom data features.

[0044] Step B2: Use mushroom data features to search for neighboring mushrooms and next-nearest mushrooms around the target mushroom; mushrooms that intersect or are tangent to the target mushroom are called neighboring mushrooms, and mushrooms that are far from the target mushroom but still interfere with the target mushroom's reciprocating pushing space are called next-nearest mushrooms.

[0045] The evaluation criteria for determining whether a suspected mushroom belongs to a neighboring mushroom family are defined as follows:

[0046] 0 oi ≤r oi

[0047] Where a oi r represents the center distance between the target mushroom and the i-th suspected mushroom; oi Let represent the sum of the cap radii of the target and the i-th suspicious mushroom; i∈[1,…,n-1], where n is the total number of mushrooms;

[0048] The evaluation criteria for determining whether a suspected mushroom belongs to the next-neighborhood mushroom are as follows:

[0049] r oi oi ≤d o +r i

[0050] Where r i Let d be the radius of the cap of the suspected mushroom. o The reciprocating range of the target mushroom is represented by the following formula:

[0051]

[0052] In the formula h o With r o represents the height of the target mushroom and the radius of the mushroom cap, respectively, and k is the coefficient for pushing and picking the range, k∈(0,1];

[0053] Step B3: Calculate the solution set of the neighboring mushrooms and the solution set of the next-nearest mushrooms, and determine the optimal pushing direction of the mushroom; whether the target mushroom is surrounded by its neighboring mushrooms is analogous to the relationship between a point and a polygon. An improved cross product criterion is used to determine the relationship between a point and a polygon, thus determining whether the target mushroom is surrounded by its neighboring mushrooms; assuming the number of neighboring mushrooms is m, at the target mushroom mass point p... o With each neighboring mushroom particle p​​i Make vectors between Where i∈[1,2,3,…,m], and m is the total number of mushrooms in the neighborhood; select any vector as shown in the example. As basis vectors, the remaining vectors are called neighborhood vectors. Neighboring vectors are then multiplied by the basis vectors in turn by cross product and dot product, as shown in the following formulas:

[0054]

[0055]

[0056] Where f is the direction coefficient, θ 2,i Angle between vectors with direction;

[0057] After performing cross product and dot product operations on all neighborhood vectors and basis vectors in sequence, sort the results in ascending order and select the maximum value θ among the results. max With minimum value θ min The specific formula for determining the relationship between a point and a polygon is as follows:

[0058]

[0059] When a point is on or outside the polygon, it is considered that the target mushroom is not surrounded by its neighboring mushrooms; the solution set A of the neighboring mushrooms is recorded, and its specific form is as follows:

[0060] A = 360° - (|θ) max |+|θ min |)

[0061] The formula for calculating the internal common tangent angle between the target mushroom and a second-neighbor mushroom using the internal common tangent theorem is as follows:

[0062]

[0063] Where n is the number of mushrooms in the second neighborhood, α oi Let α be the angle between the internal common tangents of the target mushroom and the next neighboring mushroom. oi Let r be the center distance between the target mushroom and the next neighboring mushroom. o r i These are the radius of the target mushroom cap and the radius of the mushroom cap in the next neighborhood, respectively.

[0064] The supplementary angle β of the common internal angle in the range of 0-360° oi This is called the second-neighbor mushroom individual solution set; supplementary angle β oi The calculation formula is as follows:

[0065] β oi =360°-α oi , i∈[1,…,n]

[0066] The presence of neighboring and next-neighboring mushrooms around the target mushroom can be mainly divided into three situations:

[0067] 1) There are no neighboring mushrooms or second-neighboring mushrooms around the target mushroom;

[0068] 2) The target mushroom is surrounded by both neighboring mushrooms and second-neighboring mushrooms;

[0069] 3) Only one type of mushroom exists around the target mushroom: either a neighboring mushroom or a second-neighboring mushroom.

[0070] For the first case, to avoid invalid calculations, the target mushroom's picking direction is assumed to be 180°. For the second case, the neighboring mushroom solution set A should be sequentially compared with the next-nearest mushroom individual solution set β. oi Perform an intersection operation; if an intersection occurs, the set is empty. If the target mushroom fails to be picked, the picking algorithm is terminated. If the intersection is not empty after all intersection operations are completed, the intersection is retained and used to determine the feasible pushing direction. For the third case, when only neighboring mushrooms exist around the target mushroom, the solution set of neighboring mushrooms obtained from the relationship determination between the target mushroom and neighboring mushrooms is used to determine the feasible pushing direction. When only second-neighboring mushrooms exist around the target mushroom, the solution set β of all second-neighboring mushroom individuals is used. oi The intersection of these intersections forms the final feasible push direction solution set for the target mushroom after interference from all second-neighbor mushrooms. This set is called the second-neighbor mushroom solution set B, and its specific form is as follows:

[0071]

[0072] When the solution set B of the next neighboring mushrooms is empty, the target mushroom will not be picked by reciprocating push, and the picking is considered a failure. When the solution set B of the next neighboring mushrooms is not empty, the target mushroom will be picked by reciprocating push, and the picking is considered a success.

[0073] From the obtained set of feasible thrust direction solutions, select the solution set with the largest angle range as the solution set for the optimal thrust direction; the formula for solving the optimal thrust direction is:

[0074]

[0075] Where θ down θ up These are the lower and upper bounds of the solution set, respectively.

[0076] Preferably, especially considering the actual situation of mushrooms growing in clusters, in order to achieve the non-destructive harvesting of mature mushrooms, in conjunction with step B, step C specifically includes the following steps:

[0077] Step C1: Construct a multi-objective optimization model for the mushroom reciprocating harvesting path:

[0078] Step C2: After the model is built in step C1, the improved NSGA-II algorithm is used to solve the model to obtain the Pareto non-dominated set;

[0079] Step C3: Based on the results obtained in step C2, sort the resulting Pareto non-dominated set according to the weights.

[0080] Preferably, step C1 specifically includes the following steps:

[0081] Step C1.1: Construct a mushroom reciprocating push picking failure rate function and calculate the picking failure rate of the picking sequence;

[0082] Step B is used to search for and record the total number of mushrooms that failed to be harvested in a complete harvesting sequence; the harvesting failure rate of the mushroom harvesting sequence is calculated, and its determination form is as follows:

[0083]

[0084] Where n failure n represents the total number of mature mushrooms that failed to be harvested due to repeated pushing along this harvesting path. sum This represents the total number of mature mushrooms.

[0085] Step C1.2: Construct a mushroom path length calculation function to calculate the path length of the picking sequence; the specific form of the path length calculation function is as follows:

[0086]

[0087] Where n is the total number of mature mushrooms, and x and y are the x and y coordinates of the mushrooms, respectively;

[0088] Step C1.3: Based on the established mushroom reciprocating harvesting failure rate function and path length function, establish a multi-objective optimization model; the optimization objective of the multi-objective optimization model is:

[0089]

[0090] The multi-objective optimization model sets two optimization objectives: the first objective is to minimize the harvesting failure rate, and the second objective is to minimize the harvesting path length.

[0091] Preferably, step C2, the process of solving the multi-objective optimization model based on the NSGA-II method, specifically includes the following steps:

[0092] Step C2.1: Individual chromosome encoding within the population;

[0093] Step C2.2: Initialize the population and calculate p for each individual in the population. failure dist sum Function value;

[0094] Step C2.3: Inject the extreme values ​​of the single-target harvesting path length into the initial population;

[0095] Step C2.4: Select the parent population using an elite strategy;

[0096] Step C2.5: Selection, crossover, and mutation operations;

[0097] Step C2.6: Merging parent and child populations and deduplication;

[0098] Step C2.7: Select the offspring population using an elitist strategy incorporating a cyclic crowding sorting algorithm;

[0099] Step C2.8: Repeat steps C2.5 to C2.7 until the target genetic generation, then exit the current loop;

[0100] Step C2.9: Select the optimal solution from the non-dominated solution set according to the weight coefficients and output it.

[0101] Preferably, in step C2.7, the main steps of the cyclic crowding sort algorithm are as follows:

[0102] Step C2.7.1: Connect the two boundary individuals 1 in the non-dominated hierarchy. d n d The crowding level is set to inf. Individuals in the non-dominated hierarchy are sorted from highest to lowest crowding level, and the individual with the lowest crowding level is deleted.

[0103] Step C2.7.2: Recalculate the crowding degree of the remaining individuals in this level, and delete the individual with the lowest crowding degree again;

[0104] Step C2.7.3: Iterate sequentially until a specified number of solutions remain.

[0105] In a further technical solution, the optimal pushing direction in step A2 of step A, which involves reciprocating the mushroom along the optimal pushing direction to separate the mushroom from the culture medium, is determined using the method for determining the optimal pushing direction for the non-destructive harvesting of mature mushrooms in step B.

[0106] A further technical solution, in step C1.1.1 below, the basis for judging the reciprocating pushing and picking of the target mushroom is as follows:

[0107] The method for determining the optimal pushing direction of mature mushrooms (B.) is used to determine whether the target mushroom can be harvested by reciprocating pushing.

[0108] In a further technical solution, the specific method of encoding the chromosomes of individual populations in step C2.1 under step C2 is as follows:

[0109] The encoding method uses integer permutation. If a total of M mature mushrooms are selected, the chromosome is divided into M segments, where each segment corresponds to the number of a mature mushroom. For 10 mature mushrooms {1,2,3,4,5,6,7,8,9,10}, then {10,5,3,4,7,6,9,8,2,1} is a valid chromosome.

[0110] A further technical solution, the specific description of the crossover method of the particle swarm algorithm in step C2.3 under step C2, is as follows:

[0111] The particle swarm optimization algorithm uses sequential crossover (OX), which involves randomly selecting start and end positions on the two parent chromosomes, copying the genes from that region on parent chromosome 1 to the same positions on offspring chromosome 1, and then sequentially filling in the missing genes from offspring chromosome 1 on parent chromosome 2. The other offspring is obtained in a similar manner.

[0112] In a further technical solution, step C2.4 of step C2 involves selecting the parent population using an elite strategy primarily based on two criteria: non-dominant hierarchy and crowding. Individuals with higher priority are selected sequentially, and crowding is used among individuals of the same hierarchy for selection. The formula for calculating crowding is as follows:

[0113]

[0114] Where n is the number of individuals in the current level, f m (i) represents the m-th objective function value of the i-th chromosome in the current level. These are the maximum and minimum values ​​of the m-th objective function in the current level, respectively;

[0115] In a further technical solution, the specific formula for adjusting the crossover probability in step C2.6 under step C2 is as follows:

[0116]

[0117] Where T is the maximum number of generations, T1 = αT, T2 = (1-α)T, and α is usually 0.382 or 0.258; β is the adjustment coefficient for the crossover probability in the later stage of evolution, β∈(0,1]. The setting of this coefficient can ensure that the crossover probability asymptotically approaches a non-zero value in the later stage of evolution.

[0118] A further technical solution, the detailed description of the repeated individual control strategy in step C2.7 under step C2 is as follows:

[0119] In the NSGA-II algorithm, the new population S(t) is generated by applying an elitist strategy to the population resulting from merging the parent population P(t) and the offspring population R(t). During the selection, crossover, and mutation processes in the parent population P(t), some individuals will not undergo crossover or mutation when the crossover and mutation probabilities are not 1. Therefore, duplicate individuals will appear after the parent and offspring populations are merged. When the new population S(t) is then sorted using non-dominated methods, these duplicate individuals will not dominate each other, and some of these duplicate individuals will be selected into the new parent population P(t+1). After several generations, the new parent population becomes increasingly dominated by duplicate solutions, resulting in a very small number of non-dominated solutions in the final Pareto solution set.

[0120] The specific steps of the repeated individual control strategy are as follows:

[0121] (1) Remove duplicate individuals from the population S(t) after merging the parent generation P(t) and the offspring generation R(t);

[0122] (2) Determine whether the number of individuals in the deduplicated population is less than the number of individuals in the population P(t). If it is less, continue with tournament selection, crossover, mutation, and population merging, and return to (1). Otherwise, perform fast non-dominated sorting on the population and select N individuals as the new parent population P(t+1).

[0123] Compared with the prior art, the technical solution of the present invention has outstanding substantive features and significant advantages:

[0124] 1. To achieve non-destructive harvesting of mushrooms, this invention proposes a reciprocating non-destructive harvesting method, and in particular for densely packed mushrooms, proposes a trajectory planning method for determining the optimal pushing direction and harvesting sequence suitable for non-destructive reciprocating harvesting of mushrooms;

[0125] 2. This invention uses the relationship between points and polygons in graph theory to analogize the relationship between the mushrooms being picked and the surrounding mushrooms in the actual growth environment. It uses the improved cross product criterion and the internal common tangent theorem to calculate the results to determine whether the mushrooms being picked can be picked by reciprocating pushing and the optimal pushing direction when they can be picked. Then, it uses the improved NSGA-II algorithm to optimize the non-destructive picking success rate and path length of the mature mushroom picking path.

[0126] 3. In order to improve the optimization efficiency of the algorithm, this invention adopts methods such as improving the quality of the initial population and injecting single-objective extrema into the initial population, which effectively improves the convergence and solution set distribution of the algorithm, enabling the algorithm to obtain a more uniform and wider-covering approximate Pareto optimal solution set while ensuring convergence to the true Pareto front.

[0127] 4. This invention optimizes the multi-objective problem of mushroom reciprocating harvesting combined with optimal path, overcomes the phenomenon of mushroom cap and stem separation in existing harvesting techniques, and effectively avoids damage to other mushrooms during mushroom harvesting by determining the non-destructive harvesting direction and planning the harvesting path, thereby effectively improving the success rate of non-destructive mushroom harvesting.

[0128] 5. This invention and method are not limited to the harvesting of edible fungi such as mushrooms and straw mushrooms, but are also applicable to the directional harvesting and trajectory planning of other types of round, clustered fruits. Attached Figure Description

[0129] Figure 1 This is a flowchart of a preferred embodiment of the method of the present invention.

[0130] Figure 2 This is a schematic diagram of the reciprocating pushing and picking process in Embodiment 2 of the present invention.

[0131] Figure 3 This is a flowchart illustrating the process of determining whether a mature mushroom is ready to be picked, as described in Embodiment 2 of the present invention.

[0132] Figure 4 This is a flowchart of the mushroom picking path planning algorithm according to Embodiment 3 of the present invention.

[0133] Figure 5 This is an image showing the actual distribution of mushrooms in Example 3.

[0134] Figure 6 This is the optimal harvesting route diagram for mature mushrooms in Example 3.

[0135] Figure 7 This is a schematic diagram of the direction of pushing when picking mature mushrooms in Example 3. Detailed Implementation

[0136] To better understand the present invention, it will be described in detail below with reference to the accompanying drawings and examples.

[0137] Example 1

[0138] In this embodiment, see Figure 1 A non-destructive mushroom harvesting method using a reciprocating pushing mechanism and its path planning method, specifically including:

[0139] A: Design a reciprocating pushing harvesting method to separate mushrooms from the culture medium;

[0140] Step A1: Use an end effector to grip or adhere to the mushroom;

[0141] Step A2: Push the mushroom back and forth along the optimal pushing direction to separate the mushroom from the culture medium;

[0142] B: Method for determining the optimal pushing direction for non-destructive harvesting of mature mushrooms;

[0143] Step B1: Obtain the center coordinates of the mushrooms, the radius of the mushroom cap, and the height of the mushrooms from the mushroom distribution image. Save this data as mushroom data features.

[0144] Step B2: Use mushroom data features to search for neighboring mushrooms and next-nearest mushrooms around the target mushroom;

[0145] Step B3: Calculate the solution set of the neighboring mushrooms and the solution set of the next neighboring mushrooms, and determine the optimal direction for pushing the mushrooms;

[0146] C: A dual-objective optimization algorithm for mature mushroom harvesting path planning, considering both the direction of non-destructive harvesting and the optimal path;

[0147] Step C1: Construct a multi-objective optimization model for mushroom harvesting routes:

[0148] Step C1.1: Construct a mushroom reciprocating push picking failure rate function and calculate the picking failure rate of the picking sequence;

[0149] Step C1.1.1: Judging the reciprocating movement of the target mushroom for picking;

[0150] Step C1.1.2: Calculate the harvesting failure rate of the mushroom harvesting sequence;

[0151] Step C1.2: Construct a mushroom path length calculation function to calculate the path length of the picking sequence;

[0152] Step C1.3: Establish a multi-objective optimization model based on the established mushroom reciprocating push picking failure rate function and path length function;

[0153] Step C2: After the model is built in step C1, the improved NSGA-II algorithm is used to solve the model to obtain the Pareto non-dominated set;

[0154] Step C2.1: Individual chromosome encoding within the population;

[0155] Step C2.2: Initialize the population;

[0156] Step C2.3: Inject the extreme value of the single-target picking path length into the initial population and calculate p for all individuals in the population. failure dist sum Function value;

[0157] Step C2.4: Select the parent population using an elite strategy;

[0158] Step C2.5: Selection, crossover, and mutation operations:

[0159] Step C2.6: Merging parent and child populations and deduplication;

[0160] Step C2.7: Select the offspring population using an elitist strategy incorporating a cyclic crowding sorting algorithm;

[0161] Step C2.7.1: Connect the two boundary individuals 1 in the non-dominated hierarchy. d n d The crowding level is set to inf. Individuals in the non-dominated hierarchy are sorted from highest to lowest crowding level, and the individual with the lowest crowding level is deleted.

[0162] Step C2.7.2: Recalculate the crowding degree of the remaining individuals in this level, and delete the individual with the lowest crowding degree again;

[0163] Step C2.7.3: Iterate sequentially until a specified number of solutions remain.

[0164] Step C2.8: Repeat steps C2.5 to C2.8 until the target genetic generation, then exit the current loop;

[0165] Step C2.9: Select the optimal solution from the non-dominated solution set according to the weight coefficients and output it.

[0166] This embodiment of the mushroom reciprocating non-destructive harvesting method and its path planning method not only effectively improves the success rate of non-destructive harvesting of mature mushrooms, especially clustered mushrooms, but also optimizes the harvesting path length and improves harvesting efficiency, thereby achieving non-destructive mushroom harvesting while obtaining the optimal harvesting path.

[0167] Example 2

[0168] This embodiment is basically the same as the above embodiments, except that:

[0169] In this embodiment, as Figures 1-3 As shown, the optimal pushing direction for mature mushrooms in step A2 is determined based on the method used in step B for determining the optimal pushing direction for non-destructive harvesting of mature mushrooms. This method specifically includes:

[0170] Step B1: Obtain the center coordinates of the mushrooms, the radius of the mushroom cap, and the height of the mushrooms from the mushroom distribution image. Save this data as mushroom data features.

[0171] Step B2: Use mushroom data features to search for neighboring mushrooms and next-nearest mushrooms around the target mushroom; mushrooms that intersect or are tangent to the target mushroom are called neighboring mushrooms, and mushrooms that are far from the target mushroom but still interfere with the target mushroom's reciprocating pushing space are called next-nearest mushrooms.

[0172] The evaluation criteria for determining whether a suspected mushroom belongs to a neighboring mushroom family are defined as follows:

[0173] 0 oi ≤r​oi

[0174] Where a oi r represents the center distance between the target mushroom and the i-th suspected mushroom; oi Let represent the sum of the cap radii of the target and the i-th suspicious mushroom; i∈[1,…,n-1], where n is the total number of mushrooms;

[0175] The evaluation criteria for determining whether a suspected mushroom belongs to the next-neighborhood mushroom are as follows:

[0176] r oi oi ≤d o +r i

[0177] Where r i Let d be the radius of the cap of the suspected mushroom. o The reciprocating range of the target mushroom is represented by the following formula:

[0178]

[0179] In the formula h o With r o represents the height of the target mushroom and the radius of the mushroom cap, respectively, and k is the coefficient for pushing and picking the range, k∈(0,1];

[0180] Step B3: Calculate the solution set of the neighboring mushrooms and the solution set of the next-nearest mushrooms, and determine the optimal pushing direction of the mushroom; whether the target mushroom is surrounded by its neighboring mushrooms is analogous to the relationship between a point and a polygon. An improved cross product criterion is used to determine the relationship between a point and a polygon, thus determining whether the target mushroom is surrounded by its neighboring mushrooms; assuming the number of neighboring mushrooms is m, at the target mushroom mass point p... o With each neighboring mushroom particle p i Make vectors between Where i∈[1,2,3,…,m], and m is the total number of mushrooms in the neighborhood; select any vector as shown in the example. As basis vectors, the remaining vectors are called neighborhood vectors. Neighboring vectors are then multiplied by the basis vectors in turn by cross product and dot product, as shown in the following formulas:

[0181]

[0182]

[0183] Where f is the direction coefficient, θ 2,i Angle between vectors with direction;

[0184] After performing cross product and dot product operations on all neighborhood vectors and basis vectors in sequence, sort the results in ascending order and select the maximum value θ among the results. max ​With minimum value θ min The specific formula for determining the relationship between a point and a polygon is as follows:

[0185]

[0186] When a point is on or outside the polygon, it is considered that the target mushroom is not surrounded by its neighboring mushrooms; the solution set A of the neighboring mushrooms is recorded, and its specific form is as follows:

[0187] A = 360° - (|θ) max |+|θ min |)

[0188] The formula for calculating the internal common tangent angle between the target mushroom and a second-neighbor mushroom using the internal common tangent theorem is as follows:

[0189]

[0190] Where n is the number of mushrooms in the second neighborhood, α oi Let a be the angle between the internal common tangents of the target mushroom and the next neighboring mushroom. oi Let r be the center distance between the target mushroom and the next neighboring mushroom. o r i These are the radius of the target mushroom cap and the radius of the mushroom cap in the next neighborhood, respectively.

[0191] The supplementary angle β of the common internal angle in the range of 0-360° oi This is called the second-neighbor mushroom individual solution set; supplementary angle β oi The calculation formula is as follows:

[0192] β oi =360°-α oi , i∈[1,…,n]

[0193] The presence of neighboring and next-neighboring mushrooms around the target mushroom can be mainly divided into three situations:

[0194] 1) There are no neighboring mushrooms or second-neighboring mushrooms around the target mushroom;

[0195] 2) The target mushroom is surrounded by both neighboring mushrooms and second-neighboring mushrooms;

[0196] 3) Only one type of mushroom exists around the target mushroom: either a neighboring mushroom or a second-neighboring mushroom.

[0197] For the first case, to avoid invalid calculations, the target mushroom's picking direction is assumed to be 180°. For the second case, the neighboring mushroom solution set A should be sequentially compared with the next-nearest mushroom individual solution set β. oi Perform an intersection operation; if an intersection occurs, the set is empty. If the target mushroom fails to be picked, the picking algorithm is terminated. If the intersection is not empty after all intersection operations are completed, the intersection is retained and used to determine the feasible pushing direction. For the third case, when only neighboring mushrooms exist around the target mushroom, the solution set of neighboring mushrooms obtained from the relationship determination between the target mushroom and neighboring mushrooms is used to determine the feasible pushing direction. When only second-neighboring mushrooms exist around the target mushroom, the solution set β of all second-neighboring mushroom individuals is used. oi The intersection of these intersections forms the final feasible push direction solution set for the target mushroom after interference from all second-neighbor mushrooms. This set is called the second-neighbor mushroom solution set B, and its specific form is as follows:

[0198]

[0199] When the solution set B of the next neighboring mushrooms is empty, the target mushroom will not be picked by reciprocating push, and the picking is considered a failure. When the solution set B of the next neighboring mushrooms is not empty, the target mushroom will be picked by reciprocating push, and the picking is considered a success.

[0200] From the obtained set of feasible thrust direction solutions, select the solution set with the largest angle range as the solution set for the optimal thrust direction; the formula for solving the optimal thrust direction is:

[0201]

[0202] Where θ down θ up These are the lower and upper bounds of the solution set, respectively.

[0203] To achieve non-destructive mushroom harvesting, this embodiment utilizes a reciprocating non-destructive harvesting method. It uses the relationship between points and polygons in graph theory as an analogy to the relationship between the mushroom to be harvested and its surrounding mushrooms in the actual growing environment. An improved cross product criterion and the internal common tangent theorem are used to calculate whether the mushroom can be harvested using reciprocating pushing, and if so, the optimal pushing direction. This method effectively determines the direction for non-destructive mushroom harvesting, significantly avoiding damage to the mushroom to be harvested and its surrounding mushrooms, thereby improving the success rate and efficiency of non-destructive mushroom harvesting.

[0204] Example 3

[0205] This embodiment is basically the same as the above embodiments, except that:

[0206] In this embodiment, as Figures 4-7 As shown, in step C, considering the direction of non-destructive harvesting and the optimal path, a dual-objective optimization algorithm for mature mushroom harvesting path planning is adopted.

[0207] To achieve the non-destructive harvesting of mature mushrooms, in conjunction with step B, step C specifically includes the following steps:

[0208] Step C1: Construct a multi-objective optimization model for mushroom harvesting routes:

[0209] Step C1.1: Construct a mushroom reciprocating push picking failure rate function and calculate the picking failure rate of the picking sequence;

[0210] Here, step C1.1 specifically includes the following steps:

[0211] Step C1.1.1: Judging the reciprocating movement of the target mushroom during harvesting

[0212] Based on the characteristics of mushroom data obtained from the mushroom distribution map, search for the neighboring mushrooms and the next-nearest mushrooms around the target mushroom.

[0213] The evaluation criteria for determining whether a suspected mushroom belongs to a neighboring mushroom family are defined as follows:

[0214] 0 < a 0i ≤r oi

[0215] Where a oi r represents the center distance between the target mushroom and the i-th mushroom; oi Let represent the sum of the cap radii of the target and the i-th mushroom; i∈(1,...,n-1), where n is the total number of mushrooms.

[0216] The evaluation criteria for determining whether a suspected mushroom belongs to the next-neighborhood mushroom are as follows:

[0217] r oi <a 0i ≤d o +r i

[0218] Where r i Let d be the radius of the cap of the suspected mushroom. o The formula representing the range of the target mushroom's movement is as follows:

[0219]

[0220] In the formula h o With r o denoted by and representing the height of the target mushroom and the radius of the mushroom cap, respectively, and k is the coefficient for pushing and picking the mushroom, k∈(0,1).

[0221] This paper uses the relationship between points and polygons to determine whether a target mushroom is surrounded by its neighboring mushrooms. The target mushroom and its neighboring mushrooms are simplified into individual points, with the center coordinates of the mushroom representing the position of each point. Except for the target mushroom, neighboring mushrooms are connected in pairs clockwise to form a closed polygon. Thus, the problem of whether a target mushroom is surrounded by its neighboring mushrooms is abstracted into the problem of the relationship between points and polygons in graph theory. An improved cross product criterion is used to determine the relationship between a point and a polygon. The point p to be judged... o With each vertex p of the polygon i Make vectors between (m is the total number of mushrooms in the neighborhood), select any vector such as... As basis vectors, the remaining vectors are called neighborhood vectors. Neighboring vectors are then multiplied by the basis vectors in turn by cross product and dot product, as shown in the following formulas:

[0222]

[0223]

[0224] Where f is the direction coefficient, θ 2,i It represents the angle between vectors with direction.

[0225] After performing cross product and dot product operations on all neighborhood vectors and basis vectors in sequence, sort the results in ascending order and select the maximum value θ. max With minimum value θ min If the maximum value θ max The absolute value and minimum value θ min If the sum of the absolute values ​​of the points is greater than 180°, the point is inside the polygon; if the sum is less than 180°, the point is outside the polygon; and if the sum is 180°, the point is on the polygon. The specific form is as follows:

[0226]

[0227] When a point is on or outside the polygon, it is considered that the target mushroom is not surrounded by its neighboring mushrooms. The solution set A of the neighboring mushrooms is recorded, and its specific form is as follows:

[0228] A = 360° - (|θ) max |+|θ min |)

[0229] The internal common tangent theorem is used to calculate the internal common tangent angle between the target mushroom and its second-nearest neighbor mushrooms. Assuming there are n second-nearest neighbor mushrooms, the formula for calculating the internal common tangent angle between the target mushroom and its second-nearest neighbor mushrooms using the internal common tangent theorem is as follows:

[0230]

[0231] Where α oiLet a be the angle between the internal common tangents of the target mushroom and the next neighboring mushroom. oi Let r be the center distance between the target mushroom and the next neighboring mushroom. o r i These are the radius of the target mushroom cap and the radius of the next neighboring mushroom cap, respectively.

[0232] The supplementary angle β of the common internal angle in the range of 0-360° oi This is called the second-neighbor mushroom individual solution set. Supplementary angle β oi The calculation formula is as follows:

[0233] β oi =360°-α oi , i∈[1,…,n]

[0234] The presence of neighboring and next-neighboring mushrooms around the target mushroom can be mainly divided into three situations:

[0235] 1) There are no neighboring mushrooms or second-neighboring mushrooms around the target mushroom;

[0236] 2) The target mushroom is surrounded by both neighboring mushrooms and second-neighboring mushrooms;

[0237] 3) Only one of the neighboring mushrooms or the next neighboring mushrooms exists around the target mushroom.

[0238] For the first case, to avoid invalid calculations, the target mushroom is pushed in a 180° direction by default.

[0239] In the second case, the solution set A of the neighboring mushrooms should be sequentially compared with the solution set β of the next neighboring mushrooms. oi Perform an intersection operation; if an intersection occurs, the set is empty. If the situation is as described above, it is considered that the target mushroom has failed to be picked by pushing it back and forth, and the algorithm for judging whether it can be picked is terminated. If the intersection is not empty after all intersection operations are completed, the intersection is retained and used to determine the feasible pushing direction.

[0240] For the third scenario, when only neighboring mushrooms exist around the target mushroom, the solution set of neighboring mushrooms obtained from the relationship determination between the target mushroom and neighboring mushrooms is used to determine the feasible pushing direction; when only second-neighboring mushrooms exist around the target mushroom, the solution set β of all second-neighboring mushroom individuals is used. oi The intersection of these intersections forms the final feasible push direction solution set for the target mushroom after interference from all second-neighbor mushrooms. This set is called the second-neighbor mushroom solution set B, and its specific form is as follows:

[0241]

[0242] When the solution set B of the next neighboring mushrooms is empty, the target mushroom cannot be picked by pushing it back and forth, and is considered a picking failure. When the solution set B of the next neighboring mushrooms is not empty, the target mushroom can be picked by pushing it back and forth, and is considered a picking success.

[0243] From the obtained set of feasible thrust direction solutions, select the solution set with the largest angle range as the solution set for the optimal thrust direction. The formula for finding the optimal thrust direction is:

[0244]

[0245] Where θ down θ up These are the lower and upper bounds of the solution set, respectively.

[0246] Record the direction of mushroom pushing and the number of mushrooms that failed to be picked during reciprocating pushing in a complete picking sequence.

[0247] Step C1.1.2: Calculate the harvesting failure rate of the mushroom harvesting sequence, which is determined in the following form:

[0248]

[0249] Where n failure n represents the total number of mushrooms that failed to be picked along this path. sum This represents the total number of mature mushrooms.

[0250] Step C1.2: Construct a mushroom path length calculation function to calculate the path length of the picking sequence; the path length calculation function has the following specific form:

[0251]

[0252] Where n is the total number of mature mushrooms, and x and y are the x-coordinate and y-coordinate of the mushrooms, respectively.

[0253] Step C1.3: Establish a multi-objective optimization model based on the established mushroom reciprocating push picking failure rate function and path length function;

[0254] The multi-objective optimization model sets two optimization objectives: the first objective is to minimize the harvesting failure rate, and the second objective is to minimize the harvesting path length. The optimization objectives of the multi-objective optimization model are:

[0255]

[0256] Step C2: After the model is built in step C1, the improved NSGA-II algorithm is used to solve the model, and the Pareto non-dominated set is obtained after a specified number of generations of evolution;

[0257] Step C2.1: Individual chromosome encoding in the population; Each individual in the population is encoded using a fixed-length, non-repeating integer encoding method. Each chromosome gene represents a mature mushroom at a corresponding position, and the chromosome length M is 28.

[0258] Step C2.2 Initialize the population and calculate p for each individual in the population. failure dist sum The function value is generated randomly to form an initial population of size 2N-1, where N = 100;

[0259] Before the algorithm begins its iterations, an initial population of twice the original size is initialized randomly, and the p-value of each individual in the population is calculated. failure dist sum The objective function value. The purpose of this is to improve the quality of individuals in the initial population by selecting a suitable initial population size through subsequent fast non-dominated sorting and an elitist strategy, thereby accelerating the convergence speed of the algorithm.

[0260] Step C2.3 involves injecting the extreme values ​​of the single-objective picking path length into the initial population. The particle swarm optimization algorithm is used to search for the individual picking sequence with the optimal path length, and this individual is then injected into the initial population of size 2N-1 (199) from step C2.2. In the initial population of a multi-objective optimization problem, injecting the extreme values ​​of each objective can accelerate the convergence speed of the algorithm. Since reducing the picking failure rate is easier to achieve than optimizing the picking path length in the mushroom picking path optimization problem, for this unbalanced multi-objective optimization problem, only the extreme values ​​of the mushroom picking path length are injected to accelerate the convergence speed of the algorithm.

[0261] Step C2.4: Select the parent population using an elite strategy; calculate the p-value of all individuals in the initial population of size 2N (200) generated in step C2.3. failure dist sum The objective function value is calculated and a fast non-dominated sort is performed. At the same time, the crowding degree of individuals in each non-dominated layer is calculated. According to the elite strategy, the better individuals are selected to form a parent population of size N (100).

[0262] Step C2.5: Selection, crossover, and mutation operations; Based on the initial population generated in step C2.4, the parent generation for breeding is determined through tournament selection; Four individuals are randomly selected from the initial population using a method with replacement. One individual is selected as the breeding candidate based on the non-dominant level and the individual's crowding level. The other breeding candidate is selected using the same method. To prevent the same individual from being selected as the breeding candidate, the two breeding candidates must be different individuals. According to the crossover (p c ), variant (p) mGenetic operations such as crossover and mutation are performed with varying probabilities to generate offspring populations. Mutation in the population uses single-point mutation; crossover is performed using two-point crossover, with the crossover probability adaptively adjusted according to the evolutionary stage. The specific formula for adjusting the crossover probability in stages is as follows:

[0263]

[0264] Where T is the maximum number of generations, R1 = αT, T2 = (1-α)T; β is the adjustment coefficient for the crossover probability in the later stages of evolution, β∈(0,1], which ensures that the crossover probability asymptotically approaches a non-zero value in the later stages of evolution. In this example, α is 0.382 and β is 0.4.

[0265] Step C2.6: Merging and deduplicating parent and child populations to generate a population of 2N (200) chromosomes. A duplicate individual control strategy is then applied to the merged population to remove duplicate individuals and improve population diversity. The specific steps of the duplicate individual control strategy are as follows:

[0266] (1) Remove duplicate individuals from the population S(t) after merging the parent generation P(t) and the offspring generation R(t);

[0267] (2) Determine whether the size of the deduplicated population is less than the population size N(100) of population P(t). If it is less, continue with tournament selection, crossover, mutation, and population merging, and return to (1). Otherwise, calculate p for the individuals in the merged population after deduplication. failure dist sum The objective function value is used to perform fast non-dominated sorting and crowding calculation on the merged population after deduplication.

[0268] Step C2.7: Select the offspring population using an elitist strategy incorporating a cyclic crowding sorting algorithm. After deduplication, merge the individuals from the merged population and arrange them in descending order of Pareto rank, placing the non-dominated sets F1, F2, ..., F... m F is added to the new parent population P(t+1). When the size of the new parent population exceeds N(100), the cyclic crowding sorting algorithm is used to remove F. m Individuals up to F m The process stops when the sum of the remaining number of individuals and the number of individuals already stored in the new parent population equals N.

[0269] Here, step C2.7 specifically includes the following steps:

[0270] Step C2.7.1: Connect the two boundary individuals 1 in the non-dominated hierarchy. d n d Let the congestion level be inf. The formula for calculating congestion level is as follows:

[0271]

[0272] Where n is the number of individuals in the current level, f m (i) represents the m-th objective function value of the i-th chromosome in the current level. These are the maximum and minimum values ​​of the m-th objective function in the current level, respectively;

[0273] Sort individuals in the non-dominated hierarchy by crowding from high to low, and delete the individual with the lowest crowding.

[0274] Step C2.7.2: Recalculate the crowding degree of the remaining individuals in this level, and delete the individual with the lowest crowding degree again;

[0275] Step C2.7.3: Iterate sequentially until a specified number of solutions remain.

[0276] Step C2.8: Repeat steps C2.5 to C2.7. If the current generation reaches the specified target generation, exit the current loop.

[0277] Step C2.9: Select the optimal solution from the non-dominated solution set according to the weight coefficients and output it.

[0278] To achieve non-destructive harvesting of mushrooms, this embodiment employs a reciprocating push non-destructive harvesting method, particularly for densely packed mushrooms. It utilizes a trajectory planning method to determine the optimal pushing direction and harvesting sequence suitable for non-destructive reciprocating push harvesting. To improve algorithm optimization efficiency, methods such as population quality enhancement of the initial population and injection of single-objective extrema into the initial population are adopted. These effectively improve the convergence and solution set distribution of the algorithm, enabling it to obtain a more uniformly distributed and wider-coverage approximate Pareto optimal solution set while ensuring convergence to the true Pareto front. This method is not limited to the harvesting of edible fungi such as mushrooms and straw mushrooms, but is also applicable to the directed harvesting and trajectory planning of other types of spherical, clustered fruits.

[0279] The above embodiments of the present invention describe a non-destructive mushroom harvesting method and path planning approach using a reciprocating pushing mechanism. First, a reciprocating pushing harvesting method is used to separate the mushroom from the culture medium. Then, a method for determining the optimal pushing direction for non-destructive harvesting of mature mushrooms is implemented. Considering both the non-destructive harvesting pushing direction and the optimal path, a dual-objective optimization algorithm for mature mushroom harvesting path planning is used. The above embodiments employ a specific form of reciprocating pushing harvesting, and determine the optimal pushing direction for this method. This reciprocating pushing harvesting with the optimal pushing direction can significantly improve the success rate of non-destructive mushroom harvesting. Furthermore, the mushroom harvesting path problem is optimized through the methods described in the above embodiments. The above embodiments of the present invention achieve optimization of a multi-objective problem combining reciprocating pushing harvesting of mushrooms with the optimal path. By determining the non-destructive harvesting direction and planning the harvesting path, damage to the currently harvested mushroom and surrounding mushrooms is effectively avoided during harvesting, thereby effectively improving the success rate of non-destructive mushroom harvesting and increasing harvesting efficiency.

[0280] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made in accordance with the spirit and essence of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A non-destructive mushroom harvesting method using a reciprocating pushing mechanism and its path planning method, characterized in that, Includes the following steps: A. Design a reciprocating pushing harvesting method to separate mushrooms from the culture medium; the mushrooms are reciprocated along the optimal pushing direction to achieve separation of the mushrooms from the culture medium; B. Method for determining the optimal pushing direction for non-destructive harvesting of mature mushrooms: C. A dual-objective optimization algorithm for mature mushroom harvesting path planning, considering both the direction of non-destructive harvesting and the optimal path; The optimal pushing direction for mature mushrooms in step A is determined based on the method used in step B for determining the optimal pushing direction for non-destructive harvesting of mature mushrooms. This method specifically includes: Step B1: Obtain the center coordinates of the mushrooms, the radius of the mushroom cap, and the height of the mushrooms from the mushroom distribution image. Save this data as mushroom data features. Step B2: Use mushroom data features to search for neighboring mushrooms and next-nearest mushrooms around the target mushroom; mushrooms that intersect or are tangent to the target mushroom are called neighboring mushrooms, and mushrooms that are far from the target mushroom but still interfere with the target mushroom's reciprocating pushing space are called next-nearest mushrooms. The evaluation criteria for determining whether a suspected mushroom belongs to a neighboring mushroom family are defined as follows: ； in Representing the target mushroom and the first i The center-to-center distance between the suspicious mushrooms; Representing the goal and the first i The radius of the cap of a suspicious mushroom; , n The total number of mushrooms; The evaluation criteria for determining whether a suspected mushroom belongs to the next-neighborhood mushroom are as follows: ； in The radius of the cap of the suspected mushroom. The reciprocating range of the target mushroom is represented by the following formula: ； In the formula and These represent the height of the target mushroom and the radius of its cap, respectively. k To increase the picking range coefficient, ; Step B3: Calculate the solution set of the neighboring mushrooms and the solution set of the next-nearest mushrooms, and determine the optimal pushing direction of the mushrooms; whether the target mushroom is surrounded by its neighboring mushrooms is analogous to the relationship between a point and a polygon. An improved cross-product criterion is used to determine the relationship between the point and the polygon, thus determining whether the target mushroom is surrounded by its neighboring mushrooms; assuming the number of neighboring mushrooms is... m At the target mushroom particle With each neighboring mushroom particle Make vectors between ,in Let m be the total number of mushrooms in the neighborhood; select any vector such as... As basis vectors, the remaining vectors are called neighborhood vectors. Neighboring vectors are then multiplied by the basis vectors in turn by cross product and dot product, as shown in the following formulas: ； ； in f For direction coefficients, Angle between vectors with direction; After performing cross product and dot product operations on all neighborhood vectors and basis vectors in sequence, sort the results in ascending order and select the maximum value among the results. and minimum value The specific formula for determining the relationship between a point and a polygon is as follows: ； When a point is on or outside the polygon, it is considered that the target mushroom is not surrounded by its neighboring mushrooms; the solution set A of the neighboring mushrooms is recorded, and its specific form is as follows: ； The formula for calculating the internal common tangent angle between the target mushroom and a second-neighbor mushroom using the internal common tangent theorem is as follows: ； Where n is the number of mushrooms in the second neighborhood. Let the angle be the angle between the internal common tangents of the target mushroom and the next neighboring mushroom. Let the center distance be between the target mushroom and the next neighboring mushroom. These are the radius of the target mushroom cap and the radius of the mushroom cap in the next neighborhood, respectively. The supplementary angle of the common internal angle within the range of 0-360° This is called the solution set of the mushroom individuals in the second neighborhood; supplementary corner The calculation formula is as follows: ； The presence of neighboring and next-neighboring mushrooms around the target mushroom can be mainly divided into three situations: 1) There are no neighboring mushrooms or second-neighboring mushrooms around the target mushroom; 2) The target mushroom is surrounded by both neighboring mushrooms and second-neighboring mushrooms; 3) Only one type of mushroom exists around the target mushroom: either a neighboring mushroom or a second-neighboring mushroom. For the first case, to avoid invalid calculations, the target mushroom's picking direction is assumed to be 180°. For the second case, the solution set A of the neighboring mushrooms should be sequentially compared with the solution set of the next-nearest mushrooms. Perform an intersection operation; if an intersection occurs, the set is empty. If the target mushroom fails to be picked, the picking algorithm is terminated. If the intersection is not empty after all intersection operations are completed, the intersection is retained and used to determine the feasible pushing direction. For the third case, when only neighboring mushrooms exist around the target mushroom, the solution set of neighboring mushrooms obtained from the relationship determination between the target mushroom and neighboring mushrooms is used to determine the feasible pushing direction. When only second-neighboring mushrooms exist around the target mushroom, the solution set of all second-neighboring mushrooms is used. The intersection of these intersections forms the final feasible push direction solution set for the target mushroom after interference from all second-neighbor mushrooms. This set is called the second-neighbor mushroom solution set B, and its specific form is as follows: ； When the solution set B of the next neighboring mushrooms is empty, the target mushroom will not be picked by pushing back and forth, and the picking is considered a failure. When the solution set B of the next neighboring mushrooms is not empty, the target mushroom will be picked by pushing back and forth, and the picking is considered a success. From the obtained set of feasible thrust direction solutions, select the solution set with the largest angle range as the solution set for the optimal thrust direction; the formula for solving the optimal thrust direction is: ； in These are the lower and upper bounds of the solution set, respectively.

2. The mushroom reciprocating non-destructive harvesting method and its path planning method according to claim 1, characterized in that, Step A specifically includes: Step A1: Use an end effector to grip or adhere to the mushroom; Step A2: Push the mushroom back and forth along the optimal pushing direction to separate the mushroom from the culture medium.

3. The mushroom reciprocating non-destructive harvesting method and its path planning method according to claim 1, characterized in that, Especially considering the reality of mushrooms growing in clusters, in order to harvest mature mushrooms without damage, in conjunction with step B, step C specifically includes the following steps: Step C1: Construct a multi-objective optimization model for the mushroom reciprocating harvesting path: Step C2: After the model is built in step C1, the improved NSGA-II algorithm is used to solve the model to obtain the Pareto non-dominated set; Step C3: Based on the results obtained in step C2, sort the resulting Pareto non-dominated sets according to their weights.

4. The mushroom reciprocating non-destructive harvesting method and its path planning method according to claim 3, characterized in that, Step C1 specifically includes the following steps: Step C1.1: Construct a mushroom reciprocating push picking failure rate function and calculate the picking failure rate of the picking sequence; Step B is used to search for and record the total number of mushrooms that failed to be harvested in a complete harvesting sequence; the harvesting failure rate of the mushroom harvesting sequence is calculated, and its determination form is as follows: ； in This represents the total number of mature mushrooms that failed to be harvested due to repeated pushing along this harvesting path. This represents the total number of mature mushrooms. Step C1.2: Construct a mushroom path length calculation function to calculate the path length of the picking sequence; the specific form of the path length calculation function is as follows: ； Where n is the total number of mature mushrooms, and x and y are the x and y coordinates of the mushrooms, respectively; Step C1.3: Based on the established mushroom reciprocating harvesting failure rate function and path length function, establish a multi-objective optimization model; the optimization objective of the multi-objective optimization model is: ； The multi-objective optimization model sets two optimization objectives: the first objective is to minimize the harvesting failure rate, and the second objective is to minimize the harvesting path length.

5. The mushroom reciprocating non-destructive harvesting method and its path planning method according to claim 3, characterized in that, In step C2, the process of solving the multi-objective optimization model based on the NSGA-II method specifically includes the following steps: Step C2.1: Individual chromosome encoding within the population; Step C2.2: Initialize the population and calculate the population size of individuals. , Function value; Step C2.3: Inject the extreme values ​​of the single-target harvesting path length into the initial population; Step C2.4: Select the parent population using an elite strategy; Step C2.5: Selection, crossover, and mutation operations; Step C2.6: Merging parent and child populations and deduplication; Step C2.7: Select the offspring population using an elitist strategy incorporating a cyclic crowding sorting algorithm; Step C2.8: Repeat steps C2.5 to C2.7 until the target genetic generation, then exit the current loop; Step C2.9: Select the optimal solution from the non-dominated solution set according to the weight coefficients and output it.

6. The mushroom reciprocating non-destructive harvesting method and its path planning method according to claim 5, characterized in that, In step C2.7, the main steps of the cyclic crowding sort algorithm are as follows: Step C2.7.1: Connect the two boundary individuals in the non-dominated hierarchy. The crowding level is set to inf. Individuals in the non-dominated hierarchy are sorted from highest to lowest crowding level, and the individual with the lowest crowding level is deleted. Step C2.7.2: Recalculate the crowding degree of the remaining individuals in this level, and delete the individual with the lowest crowding degree again; Step C2.7.3: Iterate sequentially until a specified number of solutions remain.

Images