An AGV transportation system and method for monocrystalline silicon rods

Abstract

The present invention provides an AGV transportation system and method for monocrystalline silicon rods, wherein the system includes: determining the bulk cargo quantity and bulk cargo area based on the bulk cargo transportation request at the monocrystalline silicon rod production site; selecting the best AGV trolley from the AGV operation network based on the bulk cargo quantity and bulk cargo area; controlling the AGV trolley to go to the bulk cargo area; when the AGV trolley arrives in the bulk cargo area, assisting the user to control the AGV trolley to transport the bulk cargo. The present invention determines the bulk cargo quantity, bulk cargo area and delivery area based on the bulk cargo transportation request at the monocrystalline silicon rod production site, selects the best AGV trolley from the AGV operation network, and when the AGV trolley arrives in the bulk cargo area, assists the user to control the AGV trolley to transport the bulk cargo to the delivery area, and realizes efficient control of the AGV trolley to transport the bulk cargo that often appears at the monocrystalline silicon rod production site, improves the convenience, improves the applicability of AGV transportation, and is also more intelligent.

Description

Technical Field

[0001] The present invention relates to the technical field of AGV transportation control, and in particular to an AGV transportation system and method for monocrystalline silicon rods. Background Art

[0002] At present, when monocrystalline silicon rods are produced, a large number of objects need to be transported, such as uncut silicon rods, heads, tails and edges produced during the cutting process of monocrystalline silicon rods, and cut monocrystalline silicon wafers. Usually, in order to reduce the cost of manual handling and improve the efficiency of turnover transportation, AGV carts are often used for transportation.

[0003] However, in actual application scenarios, bulk goods often need to be transported at the production site of monocrystalline silicon rods, such as the cut monocrystalline silicon wafers sampled temporarily, waste generated during the test cutting of monocrystalline silicon rods, etc., and AGV carts can only perform fixed large-volume transportation turnover tasks. These bulk goods need to be transported manually by on-site personnel, which is less convenient, reduces the applicability of AGV transportation, and is not intelligent enough.

[0004] Therefore, a solution is urgently needed. Summary of the invention

[0005] One of the purposes of the present invention is to provide an AGV transportation system for monocrystalline silicon rods. Based on the bulk cargo transportation request at the monocrystalline silicon rod production site, the bulk cargo quantity, bulk cargo area and delivery area are determined, and the best AGV car is selected from the AGV operation network. When the AGV car arrives at the bulk cargo area, the user is assisted in controlling the AGV car to transport the bulk cargo to the delivery area, thereby realizing efficient control of the AGV car to transport bulk cargo that often appears at the monocrystalline silicon rod production site, thereby improving convenience, improving the applicability of AGV transportation, and at the same time, being more intelligent.

[0006] An AGV transportation system for a single crystal silicon rod provided by an embodiment of the present invention includes:

[0007] A determination module, for determining the bulk cargo quantity, bulk cargo area, and delivery area based on the bulk cargo transportation request of the monocrystalline silicon rod production site;

[0008] A selection module is used to select the best AGV from the AGV operation network based on bulk volume, bulk area, and delivery area;

[0009] Control module, used to control the AGV to go to the bulk cargo area;

[0010] The auxiliary module is used to assist the user in controlling the AGV to transport bulk cargo to the delivery area when the AGV arrives at the bulk cargo area.

[0011] Preferably, the auxiliary module assists the user in controlling the AGV to transport bulk cargo to the delivery area, including:

[0012] Performing a first motion control on the AGV so that the first-viewing angle image of the AGV continues to meet the first-image constraint condition;

[0013] Display the first-person perspective image to the user in real time;

[0014] When the user selects the target object in the first perspective image, the AGV is controlled to move in the second direction so that the first perspective image continues to meet the constraints of the second image.

[0015] When the user selects the target object, the AGV is controlled by the third motion so that the target object is loaded into the AGV.

[0016] The fourth motion control is performed on the AGV so that the AGV goes to the delivery area.

[0017] Preferably, the first picture constraint condition includes:

[0018] The matching degree between the object type set completely contained in the first-person perspective image and the standard object type set corresponding to the area type of the bulk cargo area is greater than or equal to the matching degree threshold;

[0019] and / or,

[0020] The upper body of the commander in the bulk cargo area is continuously and completely included in the 1 / N images on the target side of the first-person perspective image;

[0021] Among them, the steps for determining the target direction and N are as follows:

[0022] Acquire multiple screen movements of the commander's upper body generated in the first-person perspective screen within a recent preset time;

[0023] Match each screen action with multiple trigger actions in the trigger action library, and count the number of types of indication directions of the trigger actions that match the matching;

[0024] When the maximum number of types is unique, the direction indicated by the maximum number of types is taken as the target direction;

[0025] When the maximum number of types is not unique, the trigger action of the indication direction of the maximum number of types is used as the target trigger action and is associated with the indication direction corresponding to the maximum number of types;

[0026] The screen action that matches the target trigger action is used as the target screen action;

[0027] The indicated direction associated with the target trigger action that matches the target screen action generated last is used as the target direction;

[0028] N is calculated as the value corresponding to the maximum number of types in the numerical table; the larger the maximum number of types, the larger the value corresponding to the maximum number of types in the numerical table.

[0029] Preferably, the second picture constraint condition includes:

[0030] The target object is completely included in the first-person perspective picture;

[0031] and,

[0032] The postures of the target object in any two consecutive frames in the first-view picture meet the standard posture relationship corresponding to the object type of the target object;

[0033] and,

[0034] The overlap between the captured viewing angles of any two consecutive frames in the first viewing angle picture is greater than or equal to the overlap threshold.

[0035] Preferably, the AGV transportation system for monocrystalline silicon rods further comprises:

[0036] A processing module for comprising:

[0037] Detecting operational anomalies in the AGV operation network;

[0038] Display operational anomalies to the user;

[0039] Based on the exception handling knowledge base, decide on the handling solution for operation exceptions;

[0040] Handle operation anomalies based on the processing plan;

[0041] Track and obtain the progress of handling abnormal operations;

[0042] Display processing progress to the user.

[0043] An AGV transportation method for a single crystal silicon rod provided by an embodiment of the present invention includes:

[0044] Determine the bulk cargo quantity, bulk cargo area and delivery area based on the bulk cargo transportation request from the monocrystalline silicon rod production site;

[0045] Select the best AGV from the AGV operation network based on bulk volume, bulk area and delivery area;

[0046] Control the AGV to go to the bulk cargo area;

[0047] When the AGV arrives at the bulk cargo area, the auxiliary user controls the AGV to transport the bulk cargo to the delivery area.

[0048] Preferably, the auxiliary user controls the AGV to transport bulk cargo to the delivery area, including:

[0049] Performing a first motion control on the AGV so that the first-viewing angle image of the AGV continues to meet the first-image constraint condition;

[0050] Display the first-person perspective image to the user in real time;

[0051] When the user selects the target object in the first perspective image, the AGV is controlled to move in the second direction so that the first perspective image continues to meet the constraints of the second image.

[0052] When the user selects the target object, the AGV is controlled by the third motion so that the target object is loaded into the AGV.

[0053] The fourth motion control is performed on the AGV so that the AGV goes to the delivery area.

[0054] Preferably, the first picture constraint condition includes:

[0055] The matching degree between the object type set completely contained in the first-person perspective image and the standard object type set corresponding to the area type of the bulk cargo area is greater than or equal to the matching degree threshold;

[0056] and / or,

[0057] The upper body of the commander in the bulk cargo area is continuously and completely included in the 1 / N images on the target side of the first-person perspective image;

[0058] Among them, the steps for determining the target direction and N are as follows:

[0059] Acquire multiple screen movements of the commander's upper body generated in the first-person perspective screen within a recent preset time;

[0060] Match each screen action with multiple trigger actions in the trigger action library, and count the number of types of indication directions of the trigger actions that match the matching;

[0061] When the maximum number of types is unique, the direction indicated by the maximum number of types is taken as the target direction;

[0062] When the maximum number of types is not unique, the trigger action of the indication direction of the maximum number of types is used as the target trigger action and is associated with the indication direction corresponding to the maximum number of types;

[0063] The screen action that matches the target trigger action is used as the target screen action;

[0064] The indicated direction associated with the target trigger action that matches the target screen action generated last is used as the target direction;

[0065] N is calculated as the value corresponding to the maximum number of types in the numerical table; the larger the maximum number of types, the larger the value corresponding to the maximum number of types in the numerical table.

[0066] Preferably, the second picture constraint condition includes:

[0067] The target object is completely included in the first-person perspective picture;

[0068] and,

[0069] The postures of the target object in any two consecutive frames in the first-view picture meet the standard posture relationship corresponding to the object type of the target object;

[0070] and,

[0071] The overlap between the captured viewing angles of any two consecutive frames in the first viewing angle picture is greater than or equal to the overlap threshold.

[0072] Preferably, the AGV transportation method for monocrystalline silicon rods further includes:

[0073] Detecting operational anomalies in the AGV operation network;

[0074] Display operational anomalies to the user;

[0075] Based on the exception handling knowledge base, decide on the handling solution for operation exceptions;

[0076] Handle operation anomalies based on the processing plan;

[0077] Track and obtain the progress of handling abnormal operations;

[0078] Display processing progress to the user.

[0079] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0080] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0082] Figure 1 This is a schematic diagram of an AGV transportation system for single crystal silicon rods in an embodiment of the present invention;

[0083] Figure 2 It is a schematic diagram of an AGV transportation method for single crystal silicon rods in an embodiment of the present invention. DETAILED DESCRIPTION

[0084] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0085] The embodiment of the present invention provides an AGV transportation system for single crystal silicon rods, such as Figure 1 As shown, including:

[0086] Determination module 1, used to determine the bulk cargo quantity, bulk cargo area and delivery area based on the bulk cargo transportation request of the single crystal silicon rod production site;

[0087] Selection module 2, used to select the best AGV from the AGV operation network based on bulk quantity, bulk area and delivery area;

[0088] Control module 3, used to control the AGV to go to the bulk cargo area;

[0089] The auxiliary module 4 is used to assist the user in controlling the AGV to transport bulk cargo to the delivery area when the AGV arrives at the bulk cargo area.

[0090] In the above technical solution, the bulk cargo transportation request is a request for bulk cargo to be transported in the monocrystalline silicon rod production site, based on which the bulk cargo quantity, bulk cargo area and delivery area can be determined. The bulk cargo quantity is the quantity, weight, etc. of the bulk cargo to be transported, the bulk cargo area is the current location of the bulk cargo to be transported, and the delivery area is the area to which the bulk cargo to be transported needs to be transported; the AGV operation network is a site map network of the monocrystalline silicon production site, on which the real-time position of each AGV car, the remaining transportable volume, the current transportation destination, the current transportation route, etc. are marked; Based on the bulk cargo volume, bulk cargo area and delivery area, the best AGV is selected. The best means that all AGVs meet the selection conditions such as the remaining transportable volume exceeds the bulk cargo volume, and the current transportation route of the AGV passes through the bulk cargo area and the delivery area. The selection conditions can be set by technical personnel according to actual needs; the AGV is controlled to go to the bulk cargo area; since the location of the bulk cargo to be transported in the bulk cargo area is not fixed or the location is relatively scattered, when the AGV arrives at the bulk cargo area, the auxiliary user controls the AGV to transport the bulk cargo to the delivery area.

[0091] This application is based on the bulk cargo transportation request at the single crystal silicon rod production site, determines the bulk cargo quantity, bulk cargo area and delivery area, selects the best AGV car from the AGV operation network, and when the AGV car arrives at the bulk cargo area, it assists the user to control the AGV car to transport the bulk cargo to the delivery area, thereby realizing efficient control of the AGV car to transport bulk cargo that often appears at the single crystal silicon rod production site, improving the convenience and applicability of AGV transportation, and at the same time, making it more intelligent.

[0092] In one embodiment, the auxiliary module assists the user in controlling the AGV to transport bulk cargo to the delivery area, including:

[0093] Performing a first motion control on the AGV so that the first-viewing angle image of the AGV continues to meet the first-image constraint condition;

[0094] Display the first-person perspective image to the user in real time;

[0095] When the user selects the target object in the first perspective image, the AGV is controlled to move in the second direction so that the first perspective image continues to meet the constraints of the second image.

[0096] When the user selects the target object, the AGV is controlled by the third motion so that the target object is loaded into the AGV.

[0097] The fourth motion control is performed on the AGV so that the AGV goes to the delivery area.

[0098] In the above technical solution, the AGV car is equipped with an operation recorder, which can shoot the first-perspective picture; when the first-perspective picture of the AGV car continues to meet the first-picture constraint conditions, the first-perspective picture can be used to facilitate the user to select the target object that he wants to further confirm whether it needs to be transported by the AGV car; after the AGV car is subjected to the first motion control, the first-perspective picture is displayed to the user in real time, and the user can select the target object that he wants to further confirm whether it needs to be transported by the AGV car by viewing the first-perspective picture in real time; when the first-perspective picture continues to meet the second-picture constraint conditions, the first-perspective picture can be used to facilitate the user to determine the target object and further confirm whether it needs to be transported by the AGV car; when the user selects the target object, it means that the user wants the AGV car to transport the target object, and the AGV car is subjected to the third motion control so that the target object is loaded into the AGV car. The AGV car is equipped with a transporting device and a containing device, and the transporting device can transport the target object into the containing device to complete the loading; finally, the AGV car is subjected to the fourth motion control so that the AGV car goes to the delivery area. When assisting a user in controlling an AGV to transport bulk cargo to a delivery area, the embodiment of the present invention helps the user to select a target object for which they want to further confirm whether it needs to be transported by the AGV through the first-person perspective screen after the AGV arrives in the bulk cargo area. When the user selects the target object in the first-person perspective screen, it also helps the user to determine whether the target object needs to be transported by the AGV, thereby greatly improving the accuracy, applicability and comprehensiveness of the assistance.

[0099] In one embodiment, the first screen constraint condition includes:

[0100] The matching degree between the object type set completely included in the first-perspective picture and the standard object type set corresponding to the area type of the bulk cargo area is greater than or equal to the matching degree threshold; in this technical solution, the object type set includes the object types of each object completely included in the first-perspective picture; different area types of the bulk cargo area correspond to standard object type sets, and the standard object type set includes object types of objects that should be completely included in the first-perspective picture in order to make it convenient for users to select the target object that they want to further confirm whether it needs to be transported by the AGV trolley in the first-perspective picture, for example: the area type of the bulk cargo area is a temporary storage area, and the object types in the standard object type set are racks, objects placed on the racks, etc.; the matching degree threshold can be 80%; when this first picture constraint condition is met, the first-perspective picture can be convenient for users to select the target object that they want to further confirm whether it needs to be transported by the AGV trolley;

[0101] and / or,

[0102] The upper body of the commander in the bulk cargo area is continuously and completely included in the 1 / N screen on the target direction side of the first-view screen; in this technical solution, the target direction can be left, right, etc., and N is a positive integer, such as: 2, 3, 4, etc.; for example: when the target direction is right and N is 2, the 1 / N screen on the target direction side is the right 1 / 2 screen, which is the right 1 / 2 screen in the first-view screen; generally, there will be a commander in the bulk cargo area. When the AGV enters the bulk cargo area, the commander will command the AGV to find the objects to be transported, etc., and the commander will express it through the body movements of the upper body when commanding, such as waving, pointing the face, etc.; when this first-view constraint condition is met, the AGV will not only follow the commander to shoot, but also leave a part of the screen space in the first-view screen (i.e., the remaining screen except the 1 / N screen on the target direction side) to display the on-site situation, so that the user can see the commander, the on-site situation, etc. at the same time; so that the first-view screen is convenient for users to select the target object that they want to further confirm whether it needs to be transported by the AGV;

[0103] Among them, the steps for determining the target direction and N are as follows:

[0104] Acquire multiple screen movements of the commander's upper body generated in the first-person perspective screen within a recent preset time;

[0105] Match each screen action with multiple trigger actions in the trigger action library, and count the number of types of indication directions of the trigger actions that match the matching;

[0106] When the maximum number of types is unique, the direction indicated by the maximum number of types is taken as the target direction;

[0107] When the maximum number of types is not unique, the trigger action of the indication direction of the maximum number of types is used as the target trigger action and is associated with the indication direction corresponding to the maximum number of types;

[0108] The screen action that matches the target trigger action is used as the target screen action;

[0109] The indicated direction associated with the target trigger action that matches the target screen action generated last is used as the target direction;

[0110] N is the value corresponding to the maximum number of types in the numerical table; the larger the maximum number of types, the larger the value corresponding to the maximum number of types in the numerical table. In this technical solution, the preset time can be 30 seconds; the screen action can be a facial action, a hand action, etc.; the trigger action is an action representing the indicating person indicating where to look for the object to be transported, for example: the face is slightly raised in the same direction twice in a row, the hand is waved in a certain direction, etc. Different trigger actions have corresponding indication directions, and the indication direction is the direction on which side of the first-person perspective picture the area where the indicating person representing the trigger action indicates to look for the object to be transported should be placed, for example: the trigger action is the face is slightly raised twice in a row towards the left border of the first-person perspective picture, then the corresponding indication direction is the left side; the number of types of indication directions that match the trigger action refers to different types of indications. The number of directions; the more directions of the same type there are, the clearer the commander indicates where to find the objects to be transported. When the maximum number of types is unique, the direction of the maximum number of types is used as the target direction; when the maximum number of types is not unique, the target direction can be selected based on the commander's last indication of where to find the objects to be transported, and the trigger action of the direction of the maximum number of types is used as the target trigger action, and is associated with the corresponding direction of the maximum number of types, and the screen action that matches the target trigger action is used as the target screen action, and the direction associated with the target trigger action that matches the last generated target screen action is used as the target direction. The larger the maximum number of types, the clearer the commander indicates where to find the objects to be transported, the smaller the screen size where the commander's lower body needs to be placed, and the larger N is. By determining the target direction and N, the user can continue to see the area indicated by the commander where to find the objects to be transported when viewing the first-person perspective. The picture will also dynamically and adaptively adjust the picture layout reasonably according to the clarity of the commander's instructions, which greatly improves the applicability of the system. At the same time, it is also very intelligent, and improves the efficiency and effectiveness of assisting users.

[0111] In one embodiment, the second picture constraint condition includes:

[0112] The target object is completely included in the first-person perspective picture;

[0113] and,

[0114] The postures of the target object in any two consecutive frames in the first-view picture meet the standard posture relationship corresponding to the object type of the target object; the first-view picture is in the form of multiple frames, and the posture of the target object in the frame picture refers to the posture of the target object in the frame picture; different object types correspond to standard posture relationships, and when the postures in any two consecutive frames meet the standard posture relationship, the first-view picture can facilitate users to determine whether the target object needs to be transported by an AGV car. For example, if the object type is a storage box with an upper opening, the standard posture relationship is that the postures in two consecutive frames indicate that the storage box is about to be viewed from the upper opening of the storage box;

[0115] and,

[0116] The overlap between the captured perspectives of any two consecutive frames in the first-perspective image is greater than or equal to the overlap threshold. The captured perspective of a frame refers to the perspective of the frame captured by the AGV; the overlap threshold can be 90%; when the overlap between the captured perspectives of any two consecutive frames in the first-perspective image is greater than or equal to the overlap threshold, the captured perspective of the frame changes less, causing the lens to move slowly, so that the first-perspective image can facilitate users to determine the target object and further confirm whether it needs to be transported by the AGV.

[0117] In one embodiment, the AGV transportation system for single crystal silicon rods further includes:

[0118] A processing module for comprising:

[0119] Detecting operational anomalies in the AGV operation network; operational anomalies may be, for example, AGV trolley failures, etc.;

[0120] Display operational anomalies to the user;

[0121] Based on the exception handling knowledge base, decide on the handling solutions for operation exceptions; the exception handling knowledge base contains different handling solutions for operation exceptions, which can be set in advance by technical personnel according to actual needs;

[0122] Handle operation anomalies based on the processing plan;

[0123] Track and obtain the progress of handling abnormal operations;

[0124] Display processing progress to the user.

[0125] The embodiment of the present invention provides an AGV transportation method for single crystal silicon rods, such as Figure 2 As shown, including:

[0126] S1. Determine the bulk cargo quantity, bulk cargo area and delivery area based on the bulk cargo transportation request of the monocrystalline silicon rod production site;

[0127] S2, based on the bulk volume, bulk area and delivery area, select the best AGV from the AGV operation network;

[0128] S3, control the AGV to go to the bulk cargo area;

[0129] S4. When the AGV arrives at the bulk cargo area, the auxiliary user controls the AGV to transport the bulk cargo to the delivery area.

[0130] The auxiliary user controls the AGV to transport bulk cargo to the delivery area, including:

[0131] Performing a first motion control on the AGV so that the first-viewing angle image of the AGV continues to meet the first-image constraint condition;

[0132] Display the first-person perspective image to the user in real time;

[0133] When the user selects the target object in the first perspective image, the AGV is controlled to move in the second direction so that the first perspective image continues to meet the constraints of the second image.

[0134] When the user selects the target object, the AGV is controlled by the third motion so that the target object is loaded into the AGV.

[0135] The fourth motion control is performed on the AGV so that the AGV goes to the delivery area.

[0136] The first screen constraint condition includes:

[0137] The matching degree between the object type set completely contained in the first-person perspective image and the standard object type set corresponding to the area type of the bulk cargo area is greater than or equal to the matching degree threshold;

[0138] and / or,

[0139] The upper body of the commander in the bulk cargo area is continuously and completely included in the 1 / N images on the target side of the first-person perspective image;

[0140] Among them, the steps for determining the target direction and N are as follows:

[0141] Acquire multiple screen movements of the commander's upper body generated in the first-person perspective screen within a recent preset time;

[0142] Match each screen action with multiple trigger actions in the trigger action library, and count the number of types of indication directions of the trigger actions that match the matching;

[0143] When the maximum number of types is unique, the direction indicated by the maximum number of types is taken as the target direction;

[0144] When the maximum number of types is not unique, the trigger action of the indication direction of the maximum number of types is used as the target trigger action and is associated with the indication direction corresponding to the maximum number of types;

[0145] The screen action that matches the target trigger action is used as the target screen action;

[0146] The indicated direction associated with the target trigger action that matches the target screen action generated last is used as the target direction;

[0147] N is calculated as the value corresponding to the maximum number of types in the numerical table; the larger the maximum number of types, the larger the value corresponding to the maximum number of types in the numerical table.

[0148] The second picture constraint condition includes:

[0149] The target object is completely included in the first-person perspective picture;

[0150] and,

[0151] The postures of the target object in any two consecutive frames in the first-view picture meet the standard posture relationship corresponding to the object type of the target object;

[0152] and,

[0153] The overlap between the captured viewing angles of any two consecutive frames in the first viewing angle picture is greater than or equal to the overlap threshold.

[0154] The AGV transportation method for monocrystalline silicon rods also includes:

[0155] Detecting operational anomalies in the AGV operation network;

[0156] Display operational anomalies to the user;

[0157] Based on the exception handling knowledge base, decide on the handling solution for operation exceptions;

[0158] Handle operation anomalies based on the processing plan;

[0159] Track and obtain the progress of handling abnormal operations;

[0160] Display processing progress to the user.

[0161] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An AGV transportation system for single crystal silicon rods, characterized in that: include: A determination module, for determining the bulk cargo quantity, bulk cargo area, and delivery area based on the bulk cargo transportation request of the monocrystalline silicon rod production site; A selection module is used to select the best AGV from the AGV operation network based on bulk volume, bulk area, and delivery area; Control module, used to control the AGV to go to the bulk cargo area; The auxiliary module is used to assist the user in controlling the AGV to transport bulk cargo to the delivery area when the AGV arrives in the bulk cargo area; The auxiliary module assists the user in controlling the AGV to transport bulk cargo to the delivery area, including: Performing a first motion control on the AGV so that the first-viewing angle image of the AGV continues to meet the first-image constraint condition; Display the first-person perspective image to the user in real time; When the user selects the target object in the first perspective image, the AGV is controlled to move in the second direction so that the first perspective image continues to meet the constraints of the second image. When the user selects the target object, the AGV is controlled by the third motion so that the target object is loaded into the AGV. Perform the fourth motion control on the AGV so that the AGV moves to the delivery area; The first screen constraint condition includes: The matching degree between the object type set completely contained in the first-person perspective image and the standard object type set corresponding to the area type of the bulk cargo area is greater than or equal to the matching degree threshold; and / or, The upper body of the commander in the bulk cargo area is continuously and completely included in the 1 / N images on the target side of the first-person perspective image; Among them, the steps for determining the target direction and N are as follows: Acquire multiple screen movements of the commander's upper body generated in the first-person perspective screen within a recent preset time; Match each screen action with multiple trigger actions in the trigger action library, and count the number of types of indication directions of the trigger actions that match the matching; When the maximum number of types is unique, the direction indicated by the maximum number of types is taken as the target direction; When the maximum number of types is not unique, the trigger action of the indication direction of the maximum number of types is used as the target trigger action and is associated with the indication direction corresponding to the maximum number of types; The screen action that matches the target trigger action is used as the target screen action; The indicated direction associated with the target trigger action that matches the target screen action generated last is used as the target direction; N is the value corresponding to the maximum number of types in the value table; the larger the maximum number of types, the larger the value corresponding to the maximum number of types in the value table; The second picture constraint condition includes: The target object is completely included in the first-person perspective picture; and, The postures of the target object in any two consecutive frames in the first-view picture meet the standard posture relationship corresponding to the object type of the target object; and, The overlap between the captured viewing angles of any two consecutive frames in the first viewing angle picture is greater than or equal to the overlap threshold.

2. The AGV transportation system for single crystal silicon rods according to claim 1, characterized in that: Also includes: A processing module for comprising: Detecting operational anomalies in the AGV operation network; Display operational anomalies to the user; Based on the exception handling knowledge base, decide on the handling solution for operation exceptions; Handle operation anomalies based on the processing plan; Track and obtain the progress of handling abnormal operations; Display processing progress to the user.

3. An AGV transportation method for single crystal silicon rods, characterized in that: include: Determine the bulk cargo quantity, bulk cargo area and delivery area based on the bulk cargo transportation request from the monocrystalline silicon rod production site; Select the best AGV from the AGV operation network based on bulk volume, bulk area and delivery area; Control the AGV to go to the bulk cargo area; When the AGV arrives at the bulk cargo area, the auxiliary user controls the AGV to transport the bulk cargo to the delivery area; The auxiliary user controls the AGV to transport bulk cargo to the delivery area, including: Performing a first motion control on the AGV so that the first-viewing angle image of the AGV continues to meet the first-image constraint condition; Display the first-person perspective image to the user in real time; When the user selects the target object in the first perspective image, the AGV is controlled to move in the second direction so that the first perspective image continues to meet the constraints of the second image. When the user selects the target object, the AGV is controlled by the third motion so that the target object is loaded into the AGV. Perform the fourth motion control on the AGV so that the AGV moves to the delivery area; The first screen constraint condition includes: The matching degree between the object type set completely contained in the first-person perspective image and the standard object type set corresponding to the area type of the bulk cargo area is greater than or equal to the matching degree threshold; and / or, The upper body of the commander in the bulk cargo area is continuously and completely included in the 1 / N images on the target side of the first-person perspective image; Among them, the steps for determining the target direction and N are as follows: Acquire multiple screen movements of the commander's upper body generated in the first-person perspective screen within a recent preset time; Match each screen action with multiple trigger actions in the trigger action library, and count the number of types of indication directions of the trigger actions that match the matching; When the maximum number of types is unique, the direction indicated by the maximum number of types is taken as the target direction; When the maximum number of types is not unique, the trigger action of the indication direction of the maximum number of types is used as the target trigger action and is associated with the indication direction corresponding to the maximum number of types; The screen action that matches the target trigger action is used as the target screen action; The indicated direction associated with the target trigger action that matches the target screen action generated last is used as the target direction; N is the value corresponding to the maximum number of types in the value table; the larger the maximum number of types, the larger the value corresponding to the maximum number of types in the value table; The second picture constraint condition includes: The target object is completely included in the first-person perspective picture; and, The postures of the target object in any two consecutive frames in the first-view picture meet the standard posture relationship corresponding to the object type of the target object; and, The overlap between the captured viewing angles of any two consecutive frames in the first viewing angle picture is greater than or equal to the overlap threshold.

4. The AGV transportation method for single crystal silicon rods according to claim 3, characterized in that: Also includes: Detecting operational anomalies in the AGV operation network; Display operational anomalies to the user; Based on the exception handling knowledge base, decide on the handling solution for operation exceptions; Handle operation anomalies based on the processing plan; Track and obtain the progress of handling abnormal operations; Display processing progress to the user.

Images