Intelligent transportation method for single crystal improved doping material

Abstract

The invention discloses a single crystal improved doped material intelligent transportation method which is characterized by comprising the following steps: integrating a visual sensor, a sound sensor and a vibration sensor in an intelligent sensor node so as to perform multi-modal acquisition of data to obtain multi-modal data; fusing the multi-modal data into a unified data set by using a data fusion algorithm; performing joint training and learning on the multi-modal data based on a recurrent neural network to form a comprehensive anomaly detection model; in combination with an adaptive learning rate and a regularization technology, the training process of the anomaly detection model is optimized, and the generalization ability and stability of the anomaly detection model are improved; comprising a material transport vehicle, and a path planning module, an intelligent obstacle avoidance module, an intelligent positioning module and a comprehensive management module which are installed on the material transport vehicle. The accuracy, the real-time performance and the stability of the transportation equipment are improved, the false alarm rate is reduced, and the safety and the efficiency of operation of the transportation equipment are guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of material transportation, and in particular to a single crystal improved doping material intelligent transportation method. BACKGROUND

[0002] With the rapid development of new energy technology, the application of solar power generation has shown explosive growth in recent years. Solar monocrystalline silicon wafers are divided into N-type and P-type, which are used to make solar cell wafers. The earliest doping element of P-type monocrystalline silicon wafer is boron element, but boron and oxygen in the single crystal will combine to form a boron-oxygen complex, which can easily cause cell light attenuation. N-type crystalline silicon solar cell has the advantages of high minority carrier lifetime, no light attenuation, and low temperature coefficient, and has been widely concerned in the manufacture of high-efficiency solar cells. With the market's increasing demand for solar cell efficiency, N-type phosphorus-doped monocrystalline silicon solar cells will gradually become the mainstream. The master alloy doped in N-type monocrystalline silicon is extremely volatile in the high-temperature environment of the single crystal furnace, and after multiple disconnections or a certain length of single crystal silicon is produced, the master alloy volatilizes too much, resulting in a large resistance of the single crystal silicon. However, when transporting the material, the traditional conveyor belt is still monitored on site by personnel, making the conveying device have a single function when running.

[0003] Therefore, it is necessary to provide a single crystal improved doping material intelligent transportation method to solve the above technical problems. SUMMARY

[0004] The purpose of this section is to summarize some aspects of the embodiments of the present application and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of the specification to avoid obscuring the purpose of this section, abstract and title. Such simplifications or omissions cannot be used to limit the scope of the present application.

[0005] To solve the above technical problems, the present application provides the following technical scheme: a single crystal improved doping material intelligent transportation method, characterized in that it comprises: integrating a visual sensor, a sound sensor and a vibration sensor in an intelligent sensor node to collect multi-modal data and obtain multi-modal data;

[0006] Using a data fusion algorithm, the multi-modal data is fused into a unified data set;

[0007] Based on a recurrent neural network, the multi-modal data is jointly trained and learned to form a comprehensive anomaly detection model;

[0008] The training process of the anomaly detection model is optimized by combining adaptive learning rate and regularization techniques, and the generalization ability and stability of the anomaly detection model are improved; The material transport vehicle (1) and the path planning module, intelligent obstacle avoidance module, intelligent positioning module and comprehensive management module installed on the material transport vehicle are included; The material transport vehicle, path planning module, intelligent obstacle avoidance module and intelligent positioning module are signal connected with the comprehensive management module and controlled or data information is transmitted; The intelligent obstacle avoidance module includes a structured light camera and a laser range finder installed on the material transport vehicle, and further includes an obstacle avoidance algorithm submodule installed in the server; The intelligent positioning module includes a high-definition camera, a six-axis tilt sensor and a marker block placed on the site;

[0009] As a preferred mode of the single crystal improved doping material intelligent transportation method, the transportation method comprises the following processes:

[0010] Step 1: The whole transportation system is located at the calibrated initial position, which is defined as the origin of the material transportation site world coordinate system and the origin of the BIM model; The operator establishes the BIM model of the transportation device on the three-dimensional modeling software of the server, imports the BIM submodule in the path planning module, and marks the material warehouse and the transportation destination position on the BIM model through the man-machine interface;

[0011] Step 2: The path planning module generates an initial motion path of the transportation system to the material warehouse according to the path planning algorithm, and the comprehensive management module controls the movement of the material transport vehicle (1) according to the motion path until the whole transportation system moves to the material warehouse;

[0012] During the movement, the laser range finder detects the obstacle information on the motion path in real time, and when an obstacle is encountered, the obstacle avoidance algorithm submodule generates a new local path planning based on the obstacle avoidance algorithm, and the comprehensive management module controls the material transport vehicle to avoid the obstacle according to the local path planning. After avoiding the obstacle, the path planning module updates the path, and the comprehensive management module continues to control the movement of the material transport vehicle; During the movement, the intelligent positioning module locates the actual position coordinates of the material transport vehicle in the construction site world coordinate system and the position coordinates in the BIM model based on the positioning algorithm in real time, and transmits them to the path planning module to assist the path planning module in path planning;

[0013] Step 3: After the whole transportation system reaches the material warehouse, the comprehensive management module controls the movement of the six-degree-of-freedom robot on the material transport vehicle to grab the material;

[0014] Step 4: The path planning module regenerates the motion path from the material warehouse to the transportation destination, and the comprehensive management module continues to control the movement of the material transport vehicle towards the transportation destination according to the path;

[0015] Step 5: After the transportation system reaches the transportation destination, the integrated management module controls the six-degree-of-freedom robot to place the material;

[0016] Step 6: Repeat steps 1 to 5 to continue the next material transportation work until all material transportation is completed.

[0017] As a preferred mode of the single crystal improved doped material intelligent transportation method, the specific method of path planning implemented by the path planning module is as follows:

[0018] S1: The BIM submodule extracts BIM model entity information and stores it in the path planning algorithm submodule;

[0019] S2: The path planning algorithm submodule extracts the data of the material, analyzes the quality of the doped material, divides the material into grids, then converts the boundary of the material into a combination of points and lines, takes the coordinate system of the BIM model as the coordinate system of the path planning algorithm, records the coordinate values of the inflection points, and replaces the actual boundary curve with a straight line between the inflection points; the initial attribute T of each grid in the coordinate system of the path planning algorithm is assigned a value of 0;

[0020] S3: The path planning algorithm submodule virtually establishes a plane A parallel to and above the conveyor belt, and the distance between the conveyor belt and the plane A is equal to the passing height of the single crystal silicon; the path planning algorithm submodule extracts model data with an elevation higher than the conveyor belt and lower than the plane A, converts the complex boundary of the model into a combination of points and lines, places the coordinate data of the points and lines into the BIM model coordinate system, and sets the area surrounded by the points and lines as an impassable area B; when the floor grid Wij intersects with the impassable area B and the intersection is not empty, the grid attribute T is assigned a value of 1; then all grids with attribute T of 0 are constructed into a set to form an indoor path planning base map;

[0021] S4: The path planning algorithm submodule sets the position of the material transport vehicle as the starting point, converts the starting point and the destination point into coordinate points on the indoor path planning base map, sets the current position grid of the material transport vehicle as Wi, the center point coordinate of which is (xi, yi), sets the starting point grid of the path as Wq, the center point coordinate of which is (xq, yq), and sets the destination point grid as Wz, the center point coordinate of which is (xz, yz), and constructs the priority function f(xi, yi) as shown below: f(xi, yi) = g(xi, yi) + bh(xi, yi)

[0022] Wherein:

[0023] d1 = |xi-xz| + |yi-yz|

[0024] In the formula, g(xi, yi) is a distance starting point function; h(xi, yi) is a distance end point comprehensive function; b is a selection coefficient; a1 and a2 are distance coefficients, b1, b2 and b3 are safety coefficients and b3 > b2 > b1 > 1; d1 and d2 are respectively a first distance end point function and a second distance end point function; with the current position grid Wi as the center, there are eight grids Wik, k = 1, 2…8 connected therewith, and there are sixteen grids Wil, l = 1, 2…16 connected with the eight grids; when all grid attributes T in Wik are 0 and there is a grid attribute T of 1 in Wil, it is set as the first case; when there is a grid attribute T of 1 in Wik and all grid attributes T in Wil are 0, it is set as the second case; when there is a grid attribute T of 1 in Wik and there is a grid attribute T of 1 in Wil, it is set as the third case;

[0025] S5: The path planning algorithm submodule sets two sets E1 and E2, puts the starting point into E1 as an initial grid, and takes it as a grid with the highest priority level value;

[0026] S6: It is judged whether E1 is empty, if empty, it indicates that the starting point is not put into E1 in S5, the search fails, and after the search fails, S5 is returned again, if not empty, a grid with the highest priority level is selected as a current position grid;

[0027] S7: With the current position grid Wi selected in S6 as the center, eight grids Wik connected therewith are searched, grids with attribute T of 0 are found, grid information meeting the requirements is saved in E1, the f(xi, yi) value of each grid is calculated through the formula in S4, then a grid with the smallest f(xi, yi) value is selected to remain in E1, the remaining grids are deleted, and the grid remaining in E1 is taken as a new current position grid;

[0028] S8: It is judged whether the current position grid is the end point, if not the end point, it is put into E2, then S7 is returned, and the cycle is continuously looped, if the end point, the loop is exited, then all grids from the end point to the starting point are taken out in E2, a final path is composed, and then the running path of the material transport vehicle (1) is planned;

[0029] As a preferred mode of the single crystal improved doping material intelligent transportation method, the specific method for realizing obstacle avoidance by the intelligent obstacle avoidance module is as follows:

[0030] S1: The laser range finder monitors the front obstacles in real time, when the laser range finder detects that there is an obstacle at 3RL in front of the moving transportation system, the structured light camera starts and takes a photo of the obstacle, generates three-dimensional point cloud data of the obstacle, the obstacle three-dimensional point cloud data is projected onto the conveyor belt path planning base map established by the path planning module by the obstacle avoidance algorithm submodule, and the obstacle contour line is expressed by points and lines; the passing radius of the single crystal improved doped material transport vehicle is RL;

[0031] S2: The obstacle avoidance algorithm submodule generates a path line G parallel to the obstacle contour line, the distance between the path line G and the obstacle contour line is 2RL, the intersection point H between the path planned by the path planning module and the path line G divides the path line G into two parts F1 and F2, the tangent unit vectors of F1 and F2 at the intersection point H are and respectively, and the unit direction vector from the intersection point H to the destination point is calculated respectively, and the included angles of and are and respectively; then the path line with a smaller included angle is taken as the local obstacle avoidance route of the transportation system, that is, the local path planning, and is transmitted to the comprehensive management module, the comprehensive management module controls the transportation system to walk according to the local path planning, and realizes obstacle avoidance.

[0032] As a preferred mode of the single crystal improved doped material intelligent transportation method, the multi-modal data is fused by using the data fusion algorithm, that is, the data collected by different sensors is linearly weighted and summed according to the weight, and the mathematical expression formula is as follows:

[0033]

[0034] Wherein, D 融合 is the fused data, D i is the data collected by the i-th sensor, and ω i is the weight corresponding to the data collected by the i-th sensor, and N represents the number of sensors; the multi-modal data after fusion is preprocessed, and the preprocessing at least includes denoising processing and normalization processing.

[0035] The beneficial effects of the present application are: by adopting multi-modal data acquisition, data fusion, deep learning training and optimizing the training process, the accuracy, real-time performance and stability of the transportation equipment are improved, the false alarm rate is reduced, and the safety and efficiency of the transportation equipment operation are ensured. BRIEF DESCRIPTION OF DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below, and obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creating labor intensity.

[0037] Among them:

[0038] Figure 1 A flowchart of the single crystal improved dopant material intelligent transportation method according to an embodiment of the present application is shown in the figure;

[0039] Figure 2 A schematic diagram of the operation relationship of each module of the single crystal improved dopant material intelligent transportation method according to an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0040] In order to make the above-mentioned objects, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments.

[0041] Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative efforts should fall within the scope of protection of the present application.

[0042] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application. However, the present application can be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the scope of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0043] Embodiment 1

[0044] According to an embodiment of the present application, in combination with the flowchart shown in the figure, a single crystal improved dopant material intelligent transportation method specifically includes the following steps: the transportation method includes the following processes: Figure 1

[0045] Step 1: The overall transportation system is located at a calibrated initial position, which is defined as the origin of the material transportation site world coordinate system and the origin of the BIM model. The operator establishes a transportation device BIM model on the three-dimensional modeling software of the server, imports the BIM submodule in the path planning module, and marks the material warehouse and the transportation destination position on the BIM model through the human-computer interaction interface;

[0046] ​Step 2: The path planning module first generates an initial movement path of the transportation system to the material warehouse according to the path planning algorithm, and the comprehensive management module controls the movement of the material transport vehicle according to the movement path until the whole transportation system moves to the material warehouse; during the movement, the laser range finder detects the obstacle information on the movement path in real time, and when an obstacle is encountered, the obstacle avoidance algorithm submodule generates a new local path planning based on the obstacle avoidance algorithm, and the comprehensive management module controls the material transport vehicle to avoid the obstacle according to the local path planning; after avoiding the obstacle, the path planning module updates the path again, and the comprehensive management module continues to control the movement of the material transport vehicle; during the movement, the intelligent positioning module locates the actual position coordinates of the material transport vehicle in the world coordinate system of the construction site and the position coordinates in the BIM model in real time based on the positioning algorithm, and transmits them to the path planning module to assist the path planning module in path planning;

[0047] Step 3: After the whole transportation system reaches the material warehouse, the comprehensive management module controls the movement of the six-degree-of-freedom robot on the material transport vehicle to grab the material;

[0048] Step 4: The path planning module re-generates a movement path from the material warehouse to the transportation destination, and the comprehensive management module continues to control the movement of the material transport vehicle towards the transportation destination according to the path;

[0049] Step 5: After the transportation system reaches the transportation destination, the comprehensive management module controls the movement of the six-degree-of-freedom robot to place the material;

[0050] Step 6: Repeat steps 1 to 5 to continue the next material transportation work until all the material transportation is completed;

[0051] The specific method of path planning implemented by the path planning module is as follows:

[0052] S1: The BIM submodule extracts the entity information of the BIM model and stores it in the path planning algorithm submodule;

[0053] S2: The path planning algorithm submodule extracts the data of the material, analyzes the quality of the doped material, divides the material into grids, then converts the boundary of the material into a combination of points and lines, takes the coordinate system of the BIM model as the coordinate system of the path planning algorithm, records the coordinate values of the inflection points, and replaces the actual boundary curve with a straight line between the inflection points; in the coordinate system of the path planning algorithm, the initial attribute T of each grid is assigned a value of 0;

[0054] S3: The path planning algorithm submodule virtually creates a plane A parallel to and above the conveyor belt. The distance between the conveyor belt and plane A is equal to the passage height of the monocrystalline silicon. The path planning algorithm submodule extracts model data with an elevation higher than the conveyor belt but lower than plane A, transforms the complex boundary of the model into a combination of points and lines, places the coordinate data of the points and lines into the BIM model coordinate system, and sets the area enclosed by the points and lines as the impassable area B. When the intersection of the floor grid Wij and the impassable area B is not empty, the attribute T of the grid is assigned a value of 1. Then, all grids with attribute T of 0 are combined into a set to form the basic map for indoor path planning.

[0055] S4: The path planning algorithm submodule sets the location of the material transport vehicle as the starting point, and converts the starting point and destination point into coordinate points on the indoor path planning base map. The current position grid of the material transport vehicle (1) is set as Wi, and the center point coordinates of this grid are (xi, yi). The starting point grid is set as Wq, and the center point coordinates of this grid are (xq, yq). The destination grid is set as Wz, and the center point coordinates of this grid are (xz, yz). The priority function f(xi, yi) is constructed as follows: f(xi, yi) = g(xi, yi) + bh(xi, yi)

[0056] in:

[0057] d1 = |xi - xz| + |yi - yz|

[0058] In the formula, g(xi, yi) is the distance starting point function; h(xi, yi) is the distance ending point comprehensive function; b is the selection coefficient; a1 and a2 are distance coefficients, b1, b2 and b3 are safety coefficients and b3>b2>b1>1; d1 and d2 are the first distance ending point function and the second distance ending point function, respectively; taking the current position grid Wi as the center, there are eight grids Wik connected to it, k=1,2…8, and 16 grids Wil connected to these eight grids, l=1,2…16; when all grid attributes T in Wik are 0 and some grid attributes T in Wil are 1, it is set as the first case; when Wik has grid attributes T of 1 and all grid attributes T in Wil are 0, it is set as the second case; when Wik has grid attributes T of 1 and some grid attributes T in Wil are 1, it is set as the third case;

[0059] S5: The path planning algorithm submodule sets up two sets, E1 and E2. The starting point is used as the initial grid and placed in E1, which is then used as the grid with the highest priority value.

[0060] S6: Determine if E1 is empty. If it is empty, it means that the starting point was not placed in E1 in S5, and the search failed. After the search fails, return to S5. If it is not empty, select the grid with the highest priority as the current position grid.

[0061] S7: taking the current position grid Wi selected in S6 as the center, searching eight grids Wik connected with Wi, finding the grid with attribute T being 0, saving the grid information meeting the requirement in E1, calculating the f(xi, yi) value of each grid through the formula in S4, then selecting the grid with the minimum f(xi, yi) value to remain in E1, deleting the remaining grids, and taking the grid remaining in E1 as a new current position grid;

[0062] S8: judging whether the current position grid is the terminal point, if not, putting it into E2, then returning to S7, continuously circulating, if yes, exiting the loop, then taking all the grids from the terminal point to the starting point in E2, composing a final path, and then planning the running path of the material transport vehicle.

[0063] The specific method for realizing obstacle avoidance of the intelligent obstacle avoidance module is as follows:

[0064] S1: the laser range finder monitors the front obstacles in real time, when the laser range finder detects that there is an obstacle at 3RL in front of the moving transport system, the structured light camera is started and the obstacle is photographed to generate three-dimensional point cloud data of the obstacle, the obstacle avoidance algorithm sub-module projects the three-dimensional point cloud data of the obstacle on the conveyor belt path planning base map established by the path planning module, and the obstacle contour line is expressed by points and lines; the passing radius of the single-crystal improved doped material transport vehicle is RL;

[0065] S2: the obstacle avoidance algorithm sub-module generates a path line G parallel to the obstacle contour line, the distance between the path line G and the obstacle contour line is 2RL, the intersection point H between the path planned by the path planning module and the path line G divides the path line G into two parts F1 and F2, the tangent unit vectors of F1 and F2 at the intersection point H are and respectively, and the unit direction vector from the intersection point H to the destination point is calculated, and the included angles θ1 and θ2 of and are calculated respectively; then the path line with the smaller included angle is taken as the local obstacle avoidance line of the transport system, that is, the local path planning, and is transmitted to the comprehensive management module, the comprehensive management module controls the transport system to walk according to the local path planning, and realizes obstacle avoidance.

[0066] Further, the embodiment of the present application fuses multi-modal data by using a data fusion algorithm, that is, the data collected by different sensors is linearly weighted and summed according to weights, and the mathematical expression formula is as follows:

[0067]

[0068] wherein, D 融合 is the fused data, D i is the data collected by the i th sensor, and ω iis the weight corresponding to the data collected by the i th sensor, and N represents the number of sensors;

[0069] The preprocessed multi-modal data is further preprocessed, and the pre-processing at least includes denoising processing and normalization processing.

[0070] In conclusion, the application collects data of the material transport vehicle by installing intelligent sensors on the material transport vehicle, so that the material transport is more intelligent and convenient, which not only improves the safety and accuracy of material transport, but also improves the work efficiency of material transport.

[0071] Importantly, it should be noted that the constructions and arrangements of the present application shown in the various different exemplary embodiments are merely illustrative. Although only a few embodiments have been described in detail in this disclosure, those skilled in the art who review this disclosure will readily appreciate that many modifications are possible (e.g., variations in sizes, dimensions, structures, shapes and proportions of the various elements, values of parameters, mounting arrangements, use of materials, colors, orientations, etc.) without materially departing from the novel teachings and advantages of the subject matter described in this application. For example, elements shown as integrally formed can be constructed of multiple parts or elements, the position of elements can be reversed or otherwise varied, and the nature or number of elements or positions can be altered or varied. Accordingly, all such variations are intended to be included within the scope of this application. The order or sequence of any process or method steps can be changed or re-sequenced without departing from the scope of the application. Any "means plus function" clauses are intended to cover the structures described herein as performing the recited functions and not only structural equivalents but also equivalent structures. Other substitutions, modifications, changes, and omissions can be made in the design, operating conditions, and arrangement of the exemplary embodiments without departing from the scope of the application. Accordingly, the application is not limited to the particular embodiments described but extends to any embodiments that would still fall within the scope of the claims.

[0072] Furthermore, in an effort to provide a concise description of exemplary embodiments, all features of an actual implementation can not be described (i.e., those unrelated to the best mode of practicing the application currently being considered, or those unrelated to enabling the claimed application).

[0073] It should be appreciated that in the development of any actual implementation, as in any engineering or design project, numerous implementation-specific decisions can be made. Such development efforts might be complex and time-consuming, but would nevertheless be a routine undertaking of design, fabrication, and manufacture for those of ordinary skill having the benefit of this disclosure.

[0074] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced, without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A method for intelligent transport of single-crystal modified doped materials, characterized in that, The application relates to a material transportation system and a transportation method. The system comprises a material transportation vehicle and a path planning module, an intelligent obstacle avoidance module, an intelligent positioning module and a comprehensive management module installed on the material transportation vehicle; the material transportation vehicle, the path planning module, the intelligent obstacle avoidance module and the intelligent positioning module are in signal connection with the comprehensive management module and are controlled thereby or transmit data information to the comprehensive management module; the intelligent obstacle avoidance module comprises a structured light camera and a laser range finder installed on the material transportation vehicle and an obstacle avoidance algorithm submodule installed in a server; the intelligent positioning module comprises a high-definition camera, a six-axis tilt sensor and a marker block placed on the site. The transportation method comprises the following steps: Step 1: the whole transportation system is located at a calibrated initial position, which is defined as the origin of the world coordinate system of the material transportation site and the origin of the BIM model; an operator establishes a BIM model of the transportation device on a three-dimensional modeling software of the server (2), imports the BIM submodule into the path planning module and marks the material warehouse and the transportation destination position on the BIM model through a man-machine interactive interface; Step 2: the path planning module generates an initial movement path of the transportation system to the material warehouse according to a path planning algorithm, and the comprehensive management module controls the movement of the material transportation vehicle according to the movement path until the whole transportation system moves to the material warehouse; 2. The intelligent transport method for single-crystal modified doped materials according to claim 1, characterized in that, During the movement, the laser range finder detects the obstacle information on the movement path in real time, and when an obstacle is encountered, the obstacle avoidance algorithm submodule generates a new local path planning based on an obstacle avoidance algorithm, the comprehensive management module controls the material transportation vehicle to avoid the obstacle according to the local path planning, after the obstacle is avoided, the path planning module updates the path, and the comprehensive management module continues to control the movement of the material transportation vehicle; during the movement, the intelligent positioning module locates the actual position coordinates of the material transportation vehicle in the world coordinate system of the construction site and the position coordinates in the BIM model based on a positioning algorithm in real time and transmits the position coordinates to the path planning module to assist the path planning module in path planning; Step 3: after the whole transportation system reaches the material warehouse, the comprehensive management module controls the movement of the six-degree-of-freedom robot on the material transportation vehicle to grab the material; Step 4: the path planning module generates a movement path from the material warehouse to the transportation destination again, and the comprehensive management module continues to control the movement of the material transportation vehicle towards the transportation destination according to the path; Step 5: after the transportation system reaches the transportation destination, the comprehensive management module controls the movement of the six-degree-of-freedom robot to place the material; Step 6: steps 1 to 5 are repeated to continue the next material transportation work until all the material transportation is completed. The specific method for realizing path planning by the path planning module is as follows: ​ ​ 3. The intelligent transport method for single-crystal modified doped materials according to claim 1, characterized in that, ​ S1: BIM sub-module extracts BIM model entity information and stores in path planning algorithm sub-module; S2: Path planning algorithm sub-module extracts material data, analyzes the quality of doped material, divides the material into grids, then converts the material boundary into a combination of points and lines, takes the coordinate system of BIM model as the coordinate system of path planning algorithm, records the coordinate values of inflection points, and replaces the actual boundary curve with a straight line between inflection points; in the coordinate system of path planning algorithm, the initial attribute T of each grid is valued as 0; S3: Path planning algorithm sub-module virtually establishes a plane A parallel to and above the conveyor belt, the distance between the conveyor belt and the plane A is equal to the passing height of single crystal silicon; path planning algorithm sub-module extracts model data with elevation higher than the conveyor belt and lower than the plane A, converts the complex boundary of the model into a combination of points and lines, places the coordinate data of points and lines into the coordinate system of BIM model, and sets the area surrounded by points and lines as an impassable area B; when the intersection of the material car and the impassable area B is not empty, the grid attribute T is valued as 1; then all grids with attribute T as 0 are constructed into a set to form a path planning base map; S4: Path planning algorithm sub-module sets the position of the material transport vehicle as the starting point, converts the starting point and the destination point into coordinate points on the indoor path planning base map, sets the current position grid of the material transport vehicle as Wi, the center point coordinate of which is (xi, yi), the starting point grid as Wq, the center point coordinate of which is (xq, yq), and the destination point grid as Wz, the center point coordinate of which is (xz, yz), and constructs the priority function f(xi, yi) as shown below: f(xi, yi) = g(xi, yi) + bh(xi, yi) Wherein: d1 = |xi-xz| + |yi-yz| In the formula, g(xi, yi) is the distance from the starting point function; h(xi, yi) is the distance from the end point comprehensive function; b is the selection coefficient; a1 and a2 are distance coefficients, b1, b2 and b3 are safety coefficients and b3 > b2 > b1 > 1; d1 and d2 are the first distance from the end point function and the second distance from the end point function respectively; taking the current position grid Wi as the center, there are eight grids Wik, k = 1, 2…8, connected with it, and sixteen grids Wil, l = 1, 2…16, connected with the eight grids; when all grid attributes T in Wik are 0 and there is a grid attribute T of 1 in Wil, it is set as the first case; when there is a grid attribute T of 1 in Wik and all grid attributes T in Wil are 0, it is set as the second case; when there is a grid attribute T of 1 in Wik and there is a grid attribute T of 1 in Wil, it is set as the third case; S5: Path planning algorithm sub-module sets two sets E1 and E2, takes the starting point as the initial grid and puts it into E1 as the grid with the highest priority value; S6: judging whether E1 is empty, if empty, indicating that S5 does not put the starting point into E1, the search fails, after the search fails, returning to S5 again, if not empty, selecting the grid with the highest priority as the current position grid; S7: taking the current position grid Wi selected in S6 as the center, searching the eight grids Wik connected with it, finding the grid with attribute T being 0, saving the grid information meeting the requirements in E1, calculating the f(xi, yi) value of each grid through the formula in S4, then selecting the grid with the minimum f(xi, yi) value to remain in E1, deleting the remaining grids, and taking the grid remaining in E1 as the new current position grid; S8: judging whether the current position grid is the terminal point, if not, putting it into E2, then returning to S7, continuously circulating, if yes, exiting the loop, then taking out all the grids from the terminal point to the starting point in E2, composing the final path, and then planning the running path of the material transport vehicle (1).

4. The intelligent transport method for single-crystal modified doped materials according to claim 1, characterized in that, The specific method for realizing obstacle avoidance of the intelligent obstacle avoidance module is as follows: S1: the laser range finder monitors the front obstacles in real time, when the laser range finder detects that there is an obstacle at 3RL in front of the moving transport system, the structured light camera starts and takes a photo of the obstacle, generates three-dimensional point cloud data of the obstacle, the obstacle avoidance algorithm sub-module projects the three-dimensional point cloud data of the obstacle onto the conveyor belt path planning base map established by the path planning module, and expresses the obstacle contour line with points and lines; the passing radius of the single-crystal improved doped material transport vehicle is RL; S2: the obstacle avoidance algorithm sub-module generates a path line G parallel to the obstacle contour line, the distance between the path line G and the obstacle contour line is 2RL, the intersection H between the path planned by the path planning module and the path line G divides the path line G into two parts F1 and F2, the tangent unit vectors of F1 and F2 at the intersection H are and respectively, and the unit direction vector from the intersection H to the destination point is respectively, the included angles θ1 and θ2 between and are calculated, then the path line with the smaller included angle is taken as the local obstacle avoidance line of the transport system, that is, the local path planning, and is transmitted to the comprehensive management module, the comprehensive management module controls the transport system to walk according to the local path planning, and realizes obstacle avoidance.

5. The intelligent transport method for single-crystal modified doped materials according to claim 1, characterized in that, The multi-modal data is fused by using the data fusion algorithm, that is, the data collected by different sensors is linearly weighted and summed according to the weight, and the mathematical expression formula is as follows: Wherein, D 融合 is the fused data, D i is the data collected by the i th sensor, ω i is the weight corresponding to the data collected by the i th sensor, and N represents the number of sensors; and the multi-modal data after fusion is preprocessed, and the preprocessing at least includes denoising processing and normalization processing.

Images