Simulation charging pile training device

Abstract

The invention relates to a simulation charging pile training device. The device comprises a charging pile used for providing a product structure, a mixed reality terminal used for projecting a graph on the product structure provided by the charging pile according to a received instruction, and a voice module used for sending a voice prompt according to the received instruction. According to the simulation charging pile training device disclosed by the invention, personnel are trained in combination with real equipment and a mode of guiding in a visual field range, and the training mode can guide the personnel to quickly understand an equipment structure, a fault type and a solution method and can also perform practical operation at the same time, so that a better training effect is obtained.

Description

Technical Field

[0001] The invention relates to the technical field of charging piles, and in particular to a simulation charging pile training device. Background Art

[0002] A charging pile is a device that provides electricity for electric vehicles. With the rapid development of new energy vehicles in China, the supporting technologies for charging piles for new energy vehicles are also developing rapidly. The current main technical routes include high-voltage and high-power charging, liquid cooling, and high power density.

[0003] The rapid iteration of new technologies enables charging piles to provide better charging services, but it also brings about the dilemma of using multiple generations of products at the same time. This dilemma is mainly manifested in the insufficient follow-up speed of the personnel supporting system, and each generation of product updates requires staff to re-learn.

[0004] Back to the current personnel training system, it is still divided into two parts: basic content and practical operation. Practical operation requires direct operation of the product, but due to the serious shortage of teaching staff, the training personnel need to learn through self-study during the practical operation, and the training progress cannot keep up with market demand.

[0005] There are currently attempts to use virtual reality equipment for training. Although this training method has solved the problem of insufficient teaching staff to a certain extent, it lacks authenticity and cannot directly contact real equipment. How to solve this problem requires further research. Summary of the invention

[0006] The present invention provides a simulation charging pile training device, which combines real equipment with a guidance method within the field of vision to train personnel. This training method can guide personnel to quickly understand the equipment structure, fault types and solutions while also being able to perform practical operations, thereby achieving better training effects.

[0007] The above object of the present invention is achieved through the following technical solutions: The present invention provides a simulation charging pile training device, comprising: Charging piles, used to provide product structure; A mixed reality terminal for projecting graphics on the product structure provided by the charging pile according to the received instructions; The voice module is used to issue voice prompts according to the received instructions.

[0008] In a possible implementation of the present invention, the mixed reality terminal projects a graphic on a product structure provided by a charging pile according to a received instruction, including: In response to the received command instruction for finding a characteristic object, acquiring an image within a field of view; Finding a characteristic object in the image, and when the characteristic object does not exist in the image, issuing a moving instruction according to the position of the characteristic object in the storage content until the characteristic object appears in the image; Extract the edges of feature objects; The edge of the feature object is displayed on the display screen, or an image acquired by an image sensor is used on the display screen and the edge of the feature object is loaded on the image.

[0009] In a possible implementation of the present invention, searching for a characteristic object in an image includes: Acquire positioning objects within the field of view, including housings, device shells, and circuit boards; Create a recognition grid using the edges of the located objects; Adjusting a first three-dimensional model in the feature library according to the image within the field of view and projecting the first three-dimensional model onto a two-dimensional plane to obtain a first projection model; Moving the identification grid onto the first projection model and comparing them until the identification grid coincides with the first projection model; Determine the position of the positioning object according to the coincidence result between the recognition grid and the first projection model; Determine the position of the characteristic object in the image according to the position of the positioning object; Search at the location of the feature object and find the feature object.

[0010] In a possible implementation of the present invention, searching for a characteristic object in an image includes: Extract the complete edge of the image and obtain the complete edge contour; Using the complete edge profile, searching for a matching second three-dimensional model in the three-dimensional models in the feature library; Projecting the second three-dimensional model onto a two-dimensional plane to obtain a second projection model; adjusting the second projection model until the second projection model coincides with the complete edge contour; Select at least two position points on the complete edge contour; Create a reference circle at the selected position point and obtain the intersection point or intersection area of ​​the reference circle; Search in intersection points or intersection areas to find feature objects.

[0011] In a possible implementation of the present invention, searching is performed at the location of the feature object or at the intersection point or in the intersection area, and finding the feature object includes: Establish reference lines at the locations of feature objects or at intersection points or intersection areas; Use the pixels on the reference line to create an analysis curve; The analysis curve is transferred into the frequency domain and the time domain for processing to obtain a first curve group having a starting point and a cut-off point, and the number of the curves is multiple; Process the adjusted first projection model or the second model in the same manner to obtain a second curve group; Match the first curve group and the second curve group to find the feature object; Wherein, when matching the first curve group and the second curve group, it also includes moving the reference line.

[0012] In a possible implementation of the present invention, matching the first curve group and the second curve group includes: Determine a curve associated with the feature object in the second curve group according to the adjusted first projection model or the feature object on the second model, and record it as a positioning curve; Find a curve in the first curve group that is consistent with the starting point and the end point of the positioning curve, and record it as the reference positioning curve; The starting point and the end point of the reference positioning curve are used to define the range, and the feature object is determined within the defined range.

[0013] In a possible implementation of the present invention, issuing a movement instruction according to the position of the feature object in the storage content until the feature object appears in the image includes: Acquire positioning objects within the field of view, including housings, device shells, and circuit boards; Create a recognition grid using the edges of the located objects; Adjusting a first three-dimensional model in the feature library according to the image within the field of view and projecting the first three-dimensional model onto a two-dimensional plane to obtain a first projection model; Moving the identification grid onto the first projection model and comparing them until the identification grid coincides with the first projection model; Determine the position of the positioning object according to the coincidence result between the recognition grid and the first projection model; Generate movement instructions based on the position of the located object.

[0014] In a possible implementation of the present invention, extracting the edge of the feature object includes: After determining the reference positioning curve, create multiple auxiliary reference positioning curves parallel to the reference positioning curve; Use reference positioning curves and auxiliary reference positioning curves to determine position points on feature objects; Draws the edge of a feature object using location points on the feature object.

[0015] In a possible implementation of the present invention, using the position points on the feature object to draw the edge of the feature object includes: Moving the position point on the feature object to the feature object on the adjusted first projection model or the second model; Adjust the edge of the feature object so that all position points on the feature object are located on the edge of the feature object; Move the edge of the feature object onto the display screen or onto the image acquired by the image sensor on the display screen.

[0016] The beneficial effects of the present invention are: The simulation charging pile training device provided by the present invention can directly guide the trainees by means of field of view projection on a real charging pile, so that the trainees can quickly understand the equipment structure, fault types and solutions. This training method solves the problem of insufficient teaching staff while enabling practical operation, and can also realize real-time updating of training content, guiding the trainees to quickly complete their understanding of the charging pile. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a structural schematic diagram of a simulation charging pile training device provided by the present invention.

[0018] Figure 2 It is a distribution diagram of fault point guidance and inspection contents in a complete process provided by the present invention.

[0019] Figure 3 It is a schematic diagram of a feature object existing on an image provided by the present invention.

[0020] Figure 4 It is a schematic diagram of a process of creating a recognition grid using the edge of a positioning object provided by the present invention.

[0021] Figure 5 It is a schematic diagram provided by the present invention for determining the position of a feature object in an image according to the position of a positioning object.

[0022] Figure 6 It is a schematic diagram of an intersection provided by the present invention.

[0023] Figure 7 It is a schematic diagram of an intersection area provided by the present invention.

[0024] Figure 8 It is a schematic diagram of an analysis curve provided by the present invention.

[0025] Fig. 9 It is a schematic diagram provided by the present invention for transferring the analysis curve into the frequency domain and the time domain for processing.

[0026] Fig.10 It is a schematic diagram for obtaining multiple position points provided by the present invention.

[0027] Fig.11 It is a schematic diagram provided by the present invention for drawing the edge of a feature object using position points on the feature object.

[0028] In the picture, 1. Charging pile, 2. Mixed reality terminal, 3. Voice module. DETAILED DESCRIPTION

[0029] In order to more clearly understand the technical solution of the present invention, the relevant contents are first introduced.

[0030] The charging pile involved in the present invention refers to a charging pile that provides services for electric vehicles. The specific structure of the charging pile is as follows: Microcontroller Unit (MCU): It is the core processor of the charging pile motherboard and is responsible for controlling the operation of the entire charging pile system. It integrates CPU, memory (Flash and RAM), timer, timer, ADC / DAC and other functional modules, and can realize data acquisition, signal processing, communication control and other functions.

[0031] Power Management Integrated Circuit (PMIC): Responsible for power management and monitoring. It can provide multiple voltage outputs to ensure the power requirements of the system under various working conditions, and can also charge and protect the battery.

[0032] Communication interface chip: used to implement communication protocols between charging piles and vehicles, networks and other peripherals, such as CAN bus, RS232 / 485 serial port, Ethernet port, etc.

[0033] Power Switching IC (PSIC): A switch used to control the power supply, which can achieve fast switching and precise control.

[0034] Inverter: Converts AC power into DC power, or converts DC power into AC power, enabling energy transmission between the battery and the vehicle.

[0035] Charging control chip: used to control the current, voltage, power and other parameters during the charging process to ensure charging efficiency and safety. It can also dynamically adjust the charging speed and time, and provide protection functions such as overload protection, short circuit protection, over-temperature protection, etc.

[0036] Display Driver IC (DDI): used to control the display content and format of the display screen, allowing users to clearly understand the operating status and charging data of the charging pile.

[0037] It can be seen from the above description that the main structure of the charging pile involved in the present invention is a circuit. To understand and learn about the circuit, it is necessary to rely on the supporting manual and carry out practical operations.

[0038] Under the current conditions, teaching is divided into two parts: theoretical teaching and practical operation. Theoretical teaching includes basic theory and manual. Practical operation mainly involves actual operation on the structure of charging piles, the installation and connection relationship of various components, how to check and handle faults, etc.

[0039] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings.

[0040] The present invention discloses a simulation charging pile training device, please refer to Figure 1 In some examples, the simulation charging pile training device disclosed in the present invention includes three parts: a charging pile 1, a mixed reality terminal 2, and a voice module 3. The charging pile 1 refers to a real device purchased directly from the market, and the product model is not limited here. However, the charging pile 1 needs to be modeled, and the three-dimensional model obtained by modeling is used in the teaching process.

[0041] In some possible implementations, the voice module 3 is installed on the mixed reality terminal 2 , and the voice module 3 is directly controlled by the mixed reality terminal 2 .

[0042] In some possible implementations, the mixed reality terminal 2 uses MR glasses.

[0043] The mixed reality terminal 2 is provided for the training personnel to use, and its function is to project graphics on the product structure provided by the charging pile 1 according to the received instructions. The received instructions here refer to the instructions issued according to the teaching content, and the teaching content mainly refers to fault inspection and maintenance. Specifically, the fault inspection and maintenance is set as a specific case here. The specific case involves a complete process, which includes two aspects: fault point guidance and inspection content guidance, such as Figure 2 shown.

[0044] The fault point guidance and inspection contents all involve specific parts. Projecting a graphic on the product structure provided by the charging pile 1 is to guide the trained eyes to find the fault point guidance and inspection contents all involve specific parts.

[0045] The voice module 3 is used to issue voice prompts based on the received instructions. The instructions received here are also included in the teaching content. The teaching content contains specific steps, which will be given to the trainees through voice, and then the trainees will be guided to quickly locate by projecting graphics on the product structure provided by the charging pile 1.

[0046] In general, the simulation charging pile training device provided by the present invention guides the trainees to learn by combining projection of graphics and voice prompts on the product structure provided by the charging pile 1, and provides operation steps and checkpoints for a specific case.

[0047] In this way, the currently known faults can be stored in the form of teaching plans, and the training personnel can learn by using the simulation charging pile training device provided by the present invention. This method solves the problem of insufficient teaching personnel. When the teaching plan is updated, the training personnel can also directly carry out independent learning.

[0048] The specific implementation method of the mixed reality terminal 2 is introduced below.

[0049] The mixed reality terminal 2 projects the graphics on the product structure provided by the charging pile according to the received instruction in the following manner: S101, in response to a received command instruction to find a characteristic object, acquiring an image within a field of view; S102, searching for a characteristic object in the image, and when the characteristic object does not exist in the image, issuing a moving instruction according to the position of the characteristic object in the storage content until the characteristic object appears in the image; S103, extracting the edge of the feature object; S104, displaying the edge of the feature object on the display screen or using an image acquired by an image sensor on the display screen and loading the edge of the feature object on the image.

[0050] In step S101 to step S104, the command instruction of finding characteristic objects is first received. At this time, the mixed reality terminal 2 obtains the image within the field of view and searches for characteristic objects in the image. Figure 3 As shown, when there is no feature object in the image, a movement instruction is issued according to the position of the feature object in the storage content until the feature object appears in the image.

[0051] In some possible implementations, a neural recognition network is used to identify feature objects.

[0052] After the feature object is found, the edge of the feature object is extracted, and finally the edge of the feature object is displayed on the display screen or an image obtained by an image sensor is used on the display screen and the edge of the feature object is loaded on the image.

[0053] The above method can make the characteristic objects be displayed distinctively in the field of vision of the trainer, and the distinctive display can enable the trainer to locate them quickly.

[0054] The specific way to find feature objects in an image is: S201, acquiring positioning objects within a field of view, the positioning objects including a housing, a device housing, and a circuit board; S202, creating a recognition grid using the edge of the positioning object; S203, adjusting the first three-dimensional model in the feature library according to the image within the field of view and projecting the first three-dimensional model on a two-dimensional plane to obtain a first projection model; S204, moving the recognition grid to the first projection model and comparing them until the recognition grid coincides with the first projection model; S205, determining the position of the positioning object according to the coincidence result between the recognition grid and the first projection model; S206, determining the position of the feature object in the image according to the position of the positioning object; S207, searching at the location of the feature object to find the feature object.

[0055] In step S201 to step S207, the positioning objects within the field of view are first obtained. The positioning objects include the outer shell, the device shell and the circuit board. The positioning objects here refer to objects that are easy to identify, mainly referring to the outer shell of the charging pile, the outer shell of the internal device and the circuit board with a specific shape. The positions of the internal device and the circuit board are fixed, and the relative positions of the internal device and the circuit board are also fixed.

[0056] The identified positioning object will use the edge of the positioning object to create a recognition grid, such as Figure 4 As shown, the identification grid is Figure 4 The full dotted line figure in the lower right corner is then adjusted according to the image in the field of view, and the first three-dimensional model in the feature library is projected on a two-dimensional plane to obtain a first projection model.

[0057] The first three-dimensional model here refers to the model of the charging pile 1 .

[0058] Then the recognition grid is moved to the first projection model and compared until the recognition grid coincides with the first projection model. Here, it is necessary to simultaneously use the corresponding positioning object on the first three-dimensional model to create a recognition grid, and then make the two recognition grids coincide.

[0059] At this time, the position of the positioning object can be determined based on the coincidence result of the recognition grid and the first projection model. Then, the position of the characteristic object in the image is determined based on the position of the positioning object. Figure 5 As shown, finally a search is performed at the location of the feature object to find the feature object.

[0060] In the above method, the position of the positioning object refers to a range. Here, practical influencing factors such as errors are mainly considered. Therefore, it is also necessary to search at the position of the feature object to find the feature object.

[0061] The contents in step S201 to step S207 refer to searching for characteristic objects inside the charging pile. Of course, there is also a scenario of searching for characteristic objects on a circuit board. The specific implementation method of searching for characteristic objects on a circuit board is as follows: S301, extracting the complete edge of the image to obtain a complete edge contour; S302, using the complete edge profile to search for a matching second three-dimensional model in the three-dimensional models in the feature library; S303, projecting the second three-dimensional model onto a two-dimensional plane to obtain a second projection model; S304, adjusting the second projection model until the second projection model coincides with the complete edge contour; S305, selecting at least two position points on the complete edge contour; S306, creating a reference circle at the selected position point and obtaining an intersection point or an intersection area of ​​the reference circle; S307, searching in the intersection point or intersection area to find the feature object.

[0062] In steps S301 to S308, the complete edge of the image is first extracted to obtain a complete edge contour, and then the complete edge contour is used to find a matching second three-dimensional model in the three-dimensional model in the feature library. It should be additionally noted here that when a matching second three-dimensional model cannot be obtained through the complete edge contour or an error occurs in the matching of the second three-dimensional model, it is necessary to input information such as the circuit board code in order to accurately find a matching second three-dimensional model in the three-dimensional model in the feature library.

[0063] After obtaining the second 3D model, the second projection model is adjusted until the second projection model coincides with the complete edge contour. Then, at least two position points are selected on the complete edge contour. After the position points are determined, a reference circle is created at the selected position points and the intersection point of the reference circle is obtained ( Figure 6 shown) or the intersection area ( Figure 7 Then search in the intersection point or intersection area to find the feature object.

[0064] It should be noted that there are many components on the circuit board, and some components are difficult to identify due to their array arrangement or dense arrangement. If high-precision image processing is used, it will inevitably result in high hardware costs and slow processing speeds.

[0065] Therefore, the present invention adopts a processing method of first defining the range and then finding the feature object.

[0066] Search at the location of the feature object or at the intersection point or intersection area. The way to find the feature object is: S401, establishing a reference line at the location of the feature object or at the intersection point or intersection area; S402, establishing an analysis curve using pixel points on the reference line; S403, transferring the analysis curve into the frequency domain and the time domain for processing to obtain a first curve group having a starting point and an end point, wherein the number of the curves is a plurality; S404, processing the adjusted first projection model or the second model in the same manner to obtain a second curve group; S405, matching the first curve group and the second curve group to find a feature object; Wherein, when matching the first curve group and the second curve group, it also includes moving the reference line.

[0067] The content in steps S401 to S405 uses curve matching to determine the feature object. This is because the feature object on the three-dimensional model is obviously different from the actual feature object. This is because the two have obvious differences in material, texture, lighting and other conditions, and therefore it is difficult to determine them through simple comparison.

[0068] The solution used in the present invention is determined by position points. Specifically, a reference line is first established at the position of the feature object or at the intersection point or intersection area, and then the pixel points on the reference line are used to establish an analysis curve. Figure 8 As shown, the pixels have different pixel values.

[0069] In some possible implementations, the images at this time all need to be grayscale processed.

[0070] Then the analysis curve is transferred into the frequency domain and time domain for processing (Fourier transform and wavelet transform) to obtain the first curve group with starting point and cut-off point. Fig. 9 As shown, there are multiple curves, and the adjusted first projection model or the second model is processed in the same way to obtain a second curve group.

[0071] Then, the feature object can be found by matching the first curve group and the second curve group. This is because the relationship between the position points on the feature object is certain, and these position points can be represented by the first curve group and the second curve group having a starting point and an end point.

[0072] When matching the first curve group and the second curve group, if no result can be obtained, it is necessary to move the reference line and repeat the above process.

[0073] The way to match the first curve group and the second curve group is as follows: Determine a curve associated with the feature object in the second curve group according to the adjusted first projection model or the feature object on the second model, and record it as a positioning curve; Find a curve in the first curve group that is consistent with the starting point and the end point of the positioning curve, and record it as the reference positioning curve; The starting point and the end point of the reference positioning curve are used to define the range, and the feature object is determined within the defined range.

[0074] The specific method of extracting the edge of the feature object is: S501, after determining the reference positioning curve, creating a plurality of auxiliary reference positioning curves parallel to the reference positioning curve; S502, using the reference positioning curve and the auxiliary reference positioning curve to determine a position point on the feature object; S503: Draw the edge of the feature object using the position points on the feature object.

[0075] In step S501 to step S503, multiple location points are obtained. Fig.10 As shown, these position points belong to the same feature object. The position points can provide basic position references and can also participate in comparison work. This is because the present invention uses the feature objects on the first projection model or the second model to generate the edges of the feature objects.

[0076] The position point is used for comparison in the process of using the feature object on the first projection model or the second model to generate the edge of the feature object. After the edge of the feature object is generated, the edge of the generated feature object needs to be displayed. At this time, the position point provides a positioning function.

[0077] The steps to draw the edge of a feature object using the location points on the feature object are: S601, moving a position point on a feature object to the feature object on the adjusted first projection model or the second model; S602, adjusting the edge of the feature object so that all position points on the feature object are located on the edge of the feature object; S603: Move the edge of the feature object onto the display screen or onto the image acquired by the image sensor on the display screen.

[0078] See also Fig.11 In step S601 to step S603, the position points on the feature object will first be moved to the adjusted first projection model or the feature object on the second model. At this time, problems such as misalignment may occur. When the position points on the feature object cannot all be located on the edge of the feature object, the edge of the feature object needs to be adjusted. The adjustment at this time includes enlargement, reduction and rotation.

[0079] After the position points on the feature object are adjusted to be located on the edge of the feature object, the edge of the feature object is moved to the display screen or to an image acquired by an image sensor on the display screen.

[0080] The method of issuing a moving instruction according to the position of the feature object in the storage content until the feature object appears in the image is as follows; Acquire positioning objects within the field of view, including housings, device shells, and circuit boards; Create a recognition grid using the edges of the located objects; Adjusting a first three-dimensional model in the feature library according to the image within the field of view and projecting the first three-dimensional model onto a two-dimensional plane to obtain a first projection model; Moving the identification grid onto the first projection model and comparing them until the identification grid coincides with the first projection model; Determine the position of the positioning object according to the coincidence result between the recognition grid and the first projection model; Generate movement instructions based on the position of the located object.

[0081] The above content is partially similar to the content in step S201 to step S207. The similar content includes obtaining the identification grid and identifying the overlap between the grid and the first projection model. The difference is that the position of the positioning object is determined according to the overlap result between the identification grid and the first projection model, and then the movement instruction is generated according to the position of the positioning object.

[0082] The purpose of obtaining the identification grid and identifying the overlap between the grid and the first projection model is to determine the direction. After the direction is determined, a movement instruction can be generated according to the direction.

[0083] The embodiments of this specific implementation method are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A simulation charging pile training device, characterized in that: include: A charging pile (1), used to provide a product structure; A mixed reality terminal (2), used for projecting graphics on a product structure provided by the charging pile (1) according to a received instruction; A voice module (3), used for issuing voice prompts according to the received instructions; The mixed reality terminal (2) projects graphics on the product structure provided by the charging pile (1) according to the received instructions, including: In response to the received command instruction for finding a characteristic object, acquiring an image within a field of view; Finding a characteristic object in the image, and when the characteristic object does not exist in the image, issuing a moving instruction according to the position of the characteristic object in the storage content until the characteristic object appears in the image; Extract the edges of feature objects; The edge of the feature object is displayed on the display screen, or an image acquired by an image sensor is used on the display screen and the edge of the feature object is loaded on the image.

2. The simulation charging pile training device according to claim 1 is characterized in that: Finding characteristic objects in an image includes: Acquire positioning objects within the field of view, including housings, device shells, and circuit boards; Create a recognition grid using the edges of the located objects; Adjusting a first three-dimensional model in the feature library according to the image within the field of view and projecting the first three-dimensional model onto a two-dimensional plane to obtain a first projection model; Moving the identification grid onto the first projection model and comparing them until the identification grid coincides with the first projection model; Determine the position of the positioning object according to the coincidence result between the recognition grid and the first projection model; Determine the position of the characteristic object in the image according to the position of the positioning object; Search at the location of the feature object and find the feature object.

3. The simulation charging pile training device according to claim 1 is characterized in that: Finding characteristic objects in an image includes: Extract the complete edge of the image and obtain the complete edge contour; Using the complete edge profile, searching for a matching second three-dimensional model in the three-dimensional models in the feature library; Projecting the second three-dimensional model onto a two-dimensional plane to obtain a second projection model; adjusting the second projection model until the second projection model coincides with the complete edge contour; Select at least two position points on the complete edge contour; Create a reference circle at the selected position point and obtain the intersection point or intersection area of ​​the reference circle; Search in intersection points or intersection areas to find feature objects.

4. The simulation charging pile training device according to claim 2 or 3, characterized in that: Search at the location of the feature object or at the intersection point or intersection area. The feature objects found include: Establish reference lines at the locations of feature objects or at intersection points or intersection areas; Use the pixels on the reference line to create an analysis curve; The analysis curve is transferred into the frequency domain and the time domain for processing to obtain a first curve group having a starting point and a cut-off point, and the number of the curves is multiple; Process the adjusted first projection model or the second model in the same manner to obtain a second curve group; Match the first curve group and the second curve group to find the feature object; Wherein, when matching the first curve group and the second curve group, it also includes moving the reference line.

5. The simulation charging pile training device according to claim 4 is characterized in that: Matching the first curve group and the second curve group includes: Determine a curve associated with the feature object in the second curve group according to the adjusted first projection model or the feature object on the second model, and record it as a positioning curve; Find a curve in the first curve group that is consistent with the starting point and the end point of the positioning curve, and record it as the reference positioning curve; The starting point and the end point of the reference positioning curve are used to define the range, and the feature object is determined within the defined range.

6. The simulation charging pile training device according to claim 1, characterized in that: Issuing a moving instruction according to the position of the feature object in the storage content until the feature object appears in the image includes: Acquire positioning objects within the field of view, including housings, device shells, and circuit boards; Create a recognition grid using the edges of the located objects; Adjusting a first three-dimensional model in the feature library according to the image within the field of view and projecting the first three-dimensional model onto a two-dimensional plane to obtain a first projection model; Moving the identification grid onto the first projection model and comparing them until the identification grid coincides with the first projection model; Determine the position of the positioning object according to the coincidence result between the recognition grid and the first projection model; Generate movement instructions based on the position of the located object.

7. The simulation charging pile training device according to claim 5, characterized in that: Extracting the edges of feature objects includes: After determining the reference positioning curve, create multiple auxiliary reference positioning curves parallel to the reference positioning curve; Use reference positioning curves and auxiliary reference positioning curves to determine position points on feature objects; Draws the edge of a feature object using location points on the feature object.

8. The simulation charging pile training device according to claim 7, characterized in that: Using the position points on the feature object to draw the edge of the feature object includes: Moving the position point on the feature object to the feature object on the adjusted first projection model or the second model; Adjust the edge of the feature object so that all position points on the feature object are located on the edge of the feature object; Move the edge of the feature object onto the display screen or onto the image acquired by the image sensor on the display screen.

Images