Automatic alignment method and device for charging port of electric vehicle and medium

Abstract

The invention discloses an automatic alignment method and device for a charging port of an electric vehicle and a medium, and the method comprises the steps: collecting the overall contour image data of the vehicle to recognize the vehicle type, querying a vehicle type database to obtain the position parameter of the charging port, and calculating the coarse positioning coordinate data of the charging port based on the position parameter of the charging port; driving a telescopic mechanism to move the charging gun head to a coarse positioning position, acquiring stereoscopic image data of a charging port area through stereoscopic vision equipment, and identifying charging port characteristics based on the stereoscopic image data to calculate accurate three-dimensional coordinate data of the charging port; driving the telescopic mechanism to finely adjust the position of the charging gun head and controlling the telescopic mechanism to extend out of the charging gun head to move towards the charging port, and detecting contact pressure data between the charging gun head and the charging port through pressure sensing equipment; judging whether the charging gun head is successfully aligned with the charging port or not according to the contact pressure data and a preset pressure threshold value, if not, starting a manual fine adjustment mode, allowing a user to manually adjust the position of the charging gun head, and completing the gun insertion operation.

Description

Technical Field

[0001] This application relates to the field of electric vehicle technology, and in particular to an automatic alignment method, device and medium for electric vehicle charging ports. Background Technology

[0002] With the widespread application of electric vehicle charging infrastructure, automatic charging port technology has become a research hotspot due to its ability to reduce user workload. Existing automatic alignment solutions mostly rely on a single vision sensor for charging port positioning, such as using a monocular camera to directly identify charging port features and control the movement of a robotic arm. However, such methods reveal significant limitations in complex real-world scenarios: First, because the charging port positions vary considerably among different vehicle models, single-vision positioning lacks a coarse positioning stage for the overall vehicle outline, resulting in the telescopic mechanism's travel not being able to adaptively cover all vehicle models. This is especially true when the vehicle's parking position is offset, further amplifying the positioning error. Second, a single sensor is susceptible to changes in ambient light, partial obstruction, or interference from vehicle body dirt, leading to poor recognition stability and an actual alignment success rate that fails to meet the demands of high-frequency use.

[0003] Furthermore, existing technologies lack effective redundancy mechanisms when handling alignment failure scenarios. When a slight deviation occurs in automatic positioning, the system often directly reports an error and exits, forcing users to manually operate the heavy charging gun. This not only fails to reduce the burden but also increases operational complexity. While some solutions retain manual adjustment functions, mechanical interference or response delays exist when switching between automatic and manual modes, causing users to overcome significant resistance during adjustments, resulting in a poor user experience. In addition, existing methods have blind spots in safety monitoring during the insertion process, relying solely on position feedback while neglecting contact force monitoring, failing to prevent interface collisions or wear risks caused by alignment deviations in real time. Summary of the Invention

[0004] This application provides an automatic alignment method, device, and medium for electric vehicle charging ports to solve the aforementioned technical problems.

[0005] On one hand, embodiments of this application provide an automatic alignment method for an electric vehicle charging port, including: The image acquisition device acquires overall vehicle outline image data of the vehicle parking area, identifies the vehicle model based on the overall vehicle outline image data, queries the model database to obtain charging port location parameters, and calculates coarse positioning coordinate data of the charging port based on the charging port location parameters. The drive telescopic mechanism moves the charging gun head to the coarse positioning position corresponding to the coarse positioning coordinate data. The stereoscopic image data of the charging port area is obtained through the stereoscopic vision device, and the features of the charging port are identified based on the stereoscopic image data to calculate the precise three-dimensional coordinate data of the charging port. Drive the telescopic mechanism to fine-tune the position of the charging gun head, so that the charging gun head and the charging port are aligned based on the precise three-dimensional coordinate data, and control the telescopic mechanism to extend the charging gun head toward the charging port, and detect the contact pressure data between the charging gun head and the charging port through a pressure sensing device; Based on the contact pressure data and the preset pressure threshold, it is determined whether the charging gun head and the charging port are successfully aligned. If not, the manual fine-tuning mode is activated, allowing the user to manually adjust the position of the charging gun head to complete the insertion operation.

[0006] In one implementation of this application, the vehicle model is identified based on the overall vehicle outline image data, and a vehicle model database is queried to obtain charging port location parameters. Coarse positioning coordinate data of the charging port is then calculated based on these parameters. Specifically, this includes: The overall vehicle outline image data is preprocessed to generate optimized image data, and an edge detection algorithm is applied to process the optimized image data to extract vehicle outline feature data. The image preprocessing includes noise reduction and contrast enhancement operations, and the vehicle outline feature data includes front lines, rear lines, and wheel position features. The vehicle outline feature data is matched with pre-stored vehicle model templates in the vehicle model database. The vehicle model identifier is determined by similarity calculation. Based on the vehicle model identifier, the vehicle model database is queried to obtain the charging port location parameters. The charging port location parameters include distance data from the front or rear of the vehicle and ground clearance data. By combining the calibration parameters of the image acquisition device, the position parameters of the charging port are processed through a coordinate transformation algorithm to calculate the coarse positioning coordinate data of the charging port; the coarse positioning coordinate data includes the dimensions of lateral distance, ground height, and front-back distance.

[0007] In one implementation of this application, an edge detection algorithm is applied to process the optimized image data to extract vehicle contour feature data, specifically including: The optimized image data is converted to grayscale to generate grayscale image data. The Canny edge detection algorithm is then applied to the grayscale image data to perform Gaussian filtering, gradient calculation, and non-maximum suppression to extract preliminary edge data. The preliminary edge data is subjected to contour connection and filtering to generate stable vehicle contour feature data. The stable vehicle contour feature data is then processed by a geometric feature extraction algorithm to calculate key point coordinates and shape descriptors to enhance feature distinguishability.

[0008] In one implementation of this application, the feature recognition of the charging port based on the stereoscopic image data to calculate the precise three-dimensional coordinate data of the charging port specifically includes: The stereo image data is subjected to image enhancement processing to generate clear stereo image data, and a lightweight convolutional neural network model is applied to process the clear stereo image data to extract charging port feature data; the image enhancement processing includes brightness adjustment and occlusion removal operations, and the charging port feature data includes interface edges, positioning pin holes and brand logo features; Based on the charging port feature data, the corresponding point data of the charging port in the left and right views is determined by the feature matching algorithm, and the corresponding point data is processed by the triangulation algorithm. The accurate three-dimensional coordinate data of the charging port is calculated according to the parallax. The accurate three-dimensional coordinate data includes X, Y and Z direction coordinates. The precise three-dimensional coordinate data is processed by an error correction algorithm to ensure that the coordinate error is within a preset range and to generate the final precise three-dimensional coordinate data.

[0009] In one implementation of this application, a lightweight convolutional neural network model is applied to process the clear stereoscopic image data and extract charging port feature data, specifically including: The clear stereoscopic image data is normalized and scaled to generate standardized image data, and the standardized image data is processed by forward propagation of a lightweight convolutional neural network model to extract multi-level feature map data; the forward propagation process includes convolutional layer, pooling layer and activation layer operations; The multi-level feature map data is processed using an attention mechanism, focusing on the charging port area to generate enhanced feature map data. Based on the enhanced feature map data, charging port feature data is output through a fully connected layer and a Softmax classifier. The charging port feature data includes an interface edge probability map, a positioning pin hole location map, and brand identification label data.

[0010] In one implementation of this application, based on the contact pressure data and a preset pressure threshold, it is determined whether the charging gun head and the charging port are successfully aligned. If not, a manual fine-tuning mode is activated, allowing the user to manually adjust the position of the charging gun head to complete the insertion operation. Specifically, this includes: Based on the contact pressure data, the pressure distribution and unilateral pressure value are calculated to generate pressure state data, and the pressure state data is compared with a preset pressure threshold. If the pressure value on one side in the pressure status data exceeds the preset pressure threshold, a manual intervention signal is generated, and in response to the manual intervention signal, the system automatically switches to manual fine-tuning mode and releases the electromagnetic clutch to allow the user to adjust the charging gun head without resistance. In manual fine-tuning mode, users can input and adjust the position of the charging gun head until the contact pressure data is lower than the preset pressure threshold to complete the insertion operation; the adjustment of the charging gun head position includes lateral sliding, pitch swinging and rotation operations.

[0011] In one implementation of this application, it further includes: After coarse positioning is completed, fine positioning is triggered, and recognition success rate data is monitored in real time. If the recognition success rate is lower than the preset success threshold, the position of the telescopic mechanism is automatically adjusted, and stereo image data is re-acquired for iterative processing. During the precise positioning and alignment process, image data, coordinate data, and pressure data are processed through a real-time data fusion algorithm to generate comprehensive status data. Based on the comprehensive status data, the system automatically decides whether to insert the gun or perform manual fine-tuning, and outputs prompt information data through the human-computer interaction unit.

[0012] In one implementation of this application, it further includes: Alignment status data, charging data, and fault data are uploaded to the cloud platform in real time via a wireless communication module. The uploaded data is then stored and analyzed on the cloud platform to generate diagnostic reports and update instruction data. Based on the update instruction data, the newly added parameters from the vehicle model database are automatically downloaded and integrated to adapt to the new vehicle model; The system receives user input data via QR code scanning through a local interactive device, and initiates an automatic alignment process, along with voice prompts to indicate the operation progress.

[0013] On the other hand, embodiments of this application also provide an automatic alignment device for an electric vehicle charging port, the device comprising: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, which, when executed, enable the at least one processor to perform an automatic alignment method for an electric vehicle charging port as described above.

[0014] On the other hand, this application also provides a non-volatile computer storage medium storing computer-executable instructions, which, when executed, implement an automatic alignment method for an electric vehicle charging port as described above.

[0015] This application provides an automatic alignment method, device, and medium for an electric vehicle charging port, which has at least the following beneficial effects: By using overall vehicle contour image data for vehicle model recognition and coarse positioning, combined with stereo vision equipment to calculate the precise 3D coordinates of the charging port features, this collaborative mechanism of "coarse positioning + fine positioning" significantly improves alignment accuracy and success rate. In the coarse positioning stage, the vehicle model database adapts to the differences in charging port positions across different models, avoiding positioning errors caused by vehicle parking misalignment or model changes. In the fine positioning stage, stereo vision and feature recognition algorithms maintain low error even in complex environments. By monitoring contact pressure data in real time and comparing it with preset thresholds, the system can promptly identify alignment deviations and automatically switch to manual mode, preventing interface wear or damage caused by forced insertion. The manual fine-tuning mode allows users to quickly adjust the nozzle position without resistance, solving the pain points of manual intervention lag and high operational load in existing technologies, especially reducing the operational difficulty for elderly or female users. Through fully automated multi-stage coordinated control, strong adaptability to different environmental conditions is achieved without relying on high-cost robotic arm structures, significantly reducing hardware costs while ensuring safety and accuracy. The redundant design of pressure sensing and visual positioning ensures the safety boundaries of the insertion process, meeting the safety standards for electric vehicle charging facilities. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A flowchart illustrating an automatic alignment method for an electric vehicle charging port provided in an embodiment of this application; Figure 2 This is a schematic diagram of the internal structure of an automatic alignment device for an electric vehicle charging port provided in an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0019] Figure 1 This is a flowchart illustrating an automatic alignment method for an electric vehicle charging port provided in an embodiment of this application.

[0020] The analysis method involved in the embodiments of this application can be implemented by a terminal device or a server, and this application does not impose any special limitations on it. For ease of understanding and description, the following embodiments are all described in detail using a server as an example.

[0021] It should be noted that the server can be a single device or a system composed of multiple devices, i.e., a distributed server. This application does not make any specific limitations on this.

[0022] like Figure 1 As shown in the figure, an automatic alignment method for an electric vehicle charging port provided in this application includes: Step 101: Collect overall vehicle outline image data of the vehicle parking area using an image acquisition device, identify the vehicle model based on the overall vehicle outline image data, query the model database to obtain the charging port location parameters, and calculate the coarse positioning coordinate data of the charging port based on the charging port location parameters.

[0023] In this embodiment, the image acquisition device typically refers to a wide-angle camera fixedly mounted on the top or surrounding brackets of the charging pile. Its field of view is optimized to completely cover the vehicle parking area. It should be noted that when a vehicle enters the area, the camera automatically triggers the acquisition of image data containing the overall shape of the vehicle. This raw image data is transmitted to the system's central processor via wired or wireless means.

[0024] First, the system acquires raw vehicle outline image data from the image acquisition device. It should be noted that due to the complex and variable shooting environment, raw images often contain noise, uneven lighting, or slight distortion. Therefore, image preprocessing is the first crucial step. Specifically, preprocessing operations mainly include noise reduction and contrast enhancement. For example, noise reduction can use digital filtering algorithms to suppress random noise, while contrast enhancement stretches the image's grayscale range through methods such as histogram equalization, making the boundary between the vehicle and the background clearer. This ultimately generates optimized image data, laying a solid foundation for subsequent feature extraction.

[0025] Subsequently, edge detection algorithms are applied to process the optimized image data to extract key vehicle contour features. It should be noted that the purpose of edge detection is to accurately locate regions in the image where grayscale values ​​transition; these regions typically correspond to the contours of objects. The extracted vehicle contour feature data is a set, where crucial elements include, but are not limited to: front profile lines representing the shape of the vehicle's front boundary, rear profile lines defining the vehicle's rear boundary, and wheel position features serving as support points and important positional references for the vehicle. For example, front profile lines may include the hood contour and the lower edge of the windshield; rear profile lines may include the edge of the trunk lid and the contours of the taillights; and wheel position features mainly refer to the tire contact area with the ground or the curved edge of the wheel arch. These features collectively constitute the skeletal information of the vehicle's shape.

[0026] Next, the system performs vehicle model recognition and matching. Specifically, it compares the extracted vehicle contour feature data with pre-stored vehicle model templates in a pre-built vehicle model database stored locally or in the cloud. It should be noted that this database is a knowledge base containing standard contour data for various vehicle models. The core of the matching process lies in similarity calculation, which uses specific algorithms (such as calculating the Euclidean distance or cosine similarity of feature vectors) to quantify the degree of match between the currently extracted features and the features of each template. When a template with a similarity exceeding a predetermined threshold is found, the system determines a unique vehicle model identifier. Based on this identifier, querying the vehicle model database yields the charging port location parameters precisely bound to that vehicle model. These parameters are inherent attributes of the vehicle, typically including the baseline horizontal distance data of the charging port from the front or rear of the vehicle, and the ground clearance data of the charging port's center point relative to the horizontal ground. For example, for a certain SUV, its charging port location parameters might be defined as "a certain distance from the rear of the vehicle, with a certain ground clearance value."

[0027] Finally, the system needs to transform the charging port position parameters based on the vehicle itself into a coordinate system with the charging pile as a reference, thereby calculating the coarse positioning coordinate data of the charging port. Specifically, this transformation process requires combining the calibration parameters of the image acquisition device (such as a wide-angle camera). It should be noted that the calibration parameters are a set of parameters that have been precisely measured during system installation. They define the spatial position (three-dimensional coordinates) of the camera's optical center, the camera's orientation (pitch angle, yaw angle), and the lens's focal length, distortion coefficient, and other intrinsic properties. By processing the charging port position parameters and camera calibration parameters through coordinate transformation algorithms (such as rigid body transformation algorithms involving rotation matrices and translation vectors), the three-dimensional coarse positioning coordinate data of the charging port in the charging pile coordinate system can be calculated. It can be understood that this coarse positioning coordinate data must contain information in three dimensions: the lateral distance parallel to the charging pile mounting surface (usually in the left-right direction), the ground clearance perpendicular to the ground, and the front-back distance perpendicular to the front of the charging pile. This set of coordinates is the target position for the first stage of the telescopic mechanism's movement. This embodiment achieves accurate and automatic calculation of spatial coordinates that can be used for mechanical positioning from the original image through a series of interconnected image processing, feature matching, and coordinate transformation steps.

[0028] In this embodiment, to facilitate more efficient edge detection algorithm processing, the preprocessed optimized image data (usually a color image) is typically converted to grayscale first. For example, grayscale conversion can be achieved by calculating the weighted average of the RGB channels, converting the color image into grayscale image data containing only brightness information. This simplifies the data dimensions, highlights contour information, and improves the processing speed of subsequent algorithms.

[0029] Subsequently, the Canny edge detection algorithm is applied to process the generated grayscale image data. It should be noted that the Canny algorithm is a classic multi-stage edge detection algorithm widely used due to its excellent performance. First, Gaussian filtering is applied to the grayscale image data to smooth the image, suppress subtle noise that may interfere with edge detection, and preserve the main edge structures as much as possible. Next, the gradient intensity and direction of each pixel in the filtered image are calculated; regions with large gradient values ​​are potential edges. Then, non-maximum suppression is performed, a process that refines the edges by retaining only points with local maximum gradients in the gradient direction, thus slimming wide edges into fine lines of single-pixel width. After these steps, the algorithm outputs preliminary edge data composed of numerous discrete, potentially broken edge points.

[0030] Understandably, preliminary edge data often contains breaks, redundancy, or pseudo-edges generated by a small amount of noise. To obtain complete, stable features that represent the overall contour of the vehicle, the preliminary edge data needs further processing. Specifically, contour connection and filtering are performed first. For example, the contour connection algorithm connects breakpoints belonging to the same continuous edge based on criteria such as the gradient direction and spatial proximity of edge points, forming a complete contour chain. The filtering process then removes interference segments that are too short or clearly not part of the main contour of the vehicle, such as edges generated by ground gaps or leaf shadows, based on geometric properties such as the length, closure, and enclosed area of ​​the contour. Finally, it generates the main contour line that can reliably represent the shape of the vehicle, i.e., stable vehicle contour feature data.

[0031] Finally, to improve the accuracy and discriminative power of subsequent vehicle model matching, the system also processes these stable vehicle contour feature data through a geometric feature extraction algorithm. It should be noted that this step aims to extract more mathematically expressive features from the contour chain. Specifically, it calculates key point coordinates (such as corner points and curvature maxima) and shape descriptors (such as Hu moments and Fourier descriptors). For example, the center and radius of the circular contour at the wheel position can be extracted as key features; the corner coordinates of the front and rear of the vehicle are also important key points. These geometric features provide a more abstract and robust mathematical representation of the contours, greatly enhancing the discriminative power of contour features for different vehicle models, making the subsequent matching process with the vehicle model database template more accurate and reliable. This embodiment, through refined edge detection and post-processing, ensures high-quality extraction of vehicle contour feature data, providing a crucial guarantee for accurate vehicle model recognition.

[0032] Step 102: Drive the telescopic mechanism to move the charging gun head to the coarse positioning position corresponding to the coarse positioning coordinate data, acquire the stereo image data of the charging port area through the stereo vision device, and identify the features of the charging port based on the stereo image data to calculate the precise three-dimensional coordinate data of the charging port.

[0033] In this embodiment, the system first generates control commands based on coarse positioning coordinate data and drives the telescopic mechanism to move. For example, the telescopic mechanism typically consists of a stepper motor, a ball screw, and a multi-stage arm. After receiving the control command, the stepper motor rotates precisely, and the ball screw converts the rotational motion into linear motion of the arm, thereby smoothly and quickly moving the charging gun head installed at the end of the arm to the vicinity indicated by the coarse positioning coordinates.

[0034] Once the charging gun reaches the coarse positioning position, the binocular stereo vision device integrated near the gun head is activated. The stereo vision device (such as a binocular camera) acquires raw stereo image data. It should be noted that the raw image quality may fluctuate due to factors such as shooting distance, ambient lighting, reflections from the charging port surface, or even temporary deposits (such as a few water droplets or dust). Therefore, image enhancement processing is the first step. Specifically, enhancement processing mainly includes brightness adjustment and occlusion removal. For example, brightness adjustment can be achieved through an adaptive histogram equalization algorithm to make the overall image brightness more uniform and details clearer; while occlusion removal may be combined with image inpainting algorithms or by analyzing consecutive frames to attempt to reduce or eliminate interference from temporary occlusions (such as shadows cast by passing leaves on a single frame). After these processes, a significantly improved, clear stereo image is generated, creating conditions for high-precision feature recognition.

[0035] Subsequently, a lightweight convolutional neural network model is applied to process these clear 3D image data to extract essential charging port feature data. Understandably, this CNN model is specifically trained on a large amount of charging port image data; its lightweight nature is reflected in the optimized network structure, reducing the number of parameters and computational complexity while maintaining high accuracy to meet the real-time requirements of embedded devices. The extracted charging port feature data is a set of highly abstract and discriminative information. Key elements include the interface edges defining the shape of the charging port, the locating pin holes used for mechanical guidance and fixation, and brand identification features that often contain brand information. For example, the interface edge features describe the precise geometry of the charging port cover opening; the locating pin hole features precisely locate the pins used to guide insertion; and the brand identification features may include a specific logo or text, such as "DC FAST CHARGE," which helps to further verify the vehicle model or charging port type. It should be noted that this feature data is higher-level semantic information extracted from pixel-level images.

[0036] Next, based on the charging port feature data extracted from the left and right images respectively, a feature matching algorithm is needed to determine the precise location of the same feature point on the charging port in the left and right views, i.e., to generate corresponding point data. It should be noted that feature matching is crucial for binocular vision 3D reconstruction; its purpose is to find the corresponding point in the right image for a certain feature point (such as the center of a positioning pin hole) in the left image. The specific algorithm may be based on the similarity comparison of feature descriptors to complete the matching, such as calculating the distance between SIFT or ORB descriptors. Once high-confidence corresponding point data is obtained, triangulation algorithms can be used to calculate the 3D coordinates. Specifically, triangulation is an application of geometric principles. Based on the difference in pixel position of the same physical point in the views of the two cameras (i.e., parallax), the relative positional relationship between the two cameras (baseline distance), and internal parameters (such as focal length), it calculates the precise position coordinates of the physical point in actual 3D space by solving geometric triangles. The calculated precise 3D coordinate data includes X (horizontal), Y (vertical), and Z (depth) coordinate values ​​with a preset coordinate system (such as the optical center of the left camera as the origin).

[0037] Finally, to ensure the reliability of the coordinate data, the system also processes the initially calculated precise 3D coordinate data using an error correction algorithm. It is understandable that minor errors may be introduced during feature point matching and camera calibration. The error correction algorithm may be based on filtering techniques (such as Kalman filtering) or utilize multi-frame data fusion to smooth the trajectory and suppress jitter. It may also incorporate prior knowledge for plausibility verification, such as assuming the charging port is roughly located within a certain height range on the side of the vehicle. This ensures that the error in the final precise 3D coordinate data is controlled within a very small, pre-set allowable range. This verified coordinate data will be directly used to guide subsequent fine-tuning and alignment operations. This embodiment achieves sub-millimeter-level high-precision spatial positioning of the charging port through rigorous image enhancement, intelligent feature extraction, precise geometric calculation, and error correction.

[0038] In this embodiment, before inputting image data into the CNN model, data preprocessing is required, mainly normalization and scaling. It should be noted that normalization aims to scale the pixel values ​​of the image (e.g., the range of 0-255) to a fixed interval (e.g., 0-1 or -1 to 1), which helps improve the stability and convergence speed of model training. Scaling adjusts the image to a fixed size required by the model's input layer, such as 224x224 pixels. For example, for a candidate charging port image cropped from stereo image data, normalization is performed first, followed by scaling to a standard size using an interpolation algorithm, ultimately generating standardized image data with uniform specifications, facilitating batch computation by the model.

[0039] Subsequently, these standardized image data are processed through forward propagation of a lightweight convolutional neural network model. Specifically, forward propagation refers to the process of data flowing from the model's input layer to the output layer. During this process, the data is processed sequentially through a series of hierarchical structures within the model. Convolutional layers function by using multiple learnable convolutional kernels to perform sliding window calculations on the input image, extracting local features (such as edges and textures) and outputting feature maps. Pooling layers, typically following convolutional layers, are used to downsample the feature maps, reducing the amount of data and the number of parameters while maintaining the translation invariance of features; max pooling or average pooling are commonly used. Activation layers introduce non-linear transformations into the network, enabling the model to fit complex functions; commonly used activation functions include ReLU. Through these progressively layered processing steps, the original image is transformed into a series of feature maps containing information at different levels of abstraction—i.e., multi-level feature map data. Shallow feature maps may contain more details (such as edges), while deeper feature maps contain more abstract semantic information, such as part shapes.

[0040] To improve the targeting of feature extraction, the model applies an attention mechanism to process these multi-level feature map data. It's important to note that the core idea of ​​the attention mechanism is to teach the model to focus on the more important parts of the input data while ignoring less important ones. In this application, the goal is to get the network to focus on the actual charging port area in the image, rather than the surrounding sheet metal, license plate, or other distracting background elements. For example, the attention mechanism can be implemented by calculating the weights of different spatial locations or channels in the feature map, assigning higher weights to features related to the charging port, thereby generating enhanced feature map data focused on the key region.

[0041] Finally, based on the enhanced feature map data, the final required charging port feature data is output through a fully connected layer and a Softmax classifier. It can be understood that the role of the fully connected layer is to flatten and combine the two-dimensional feature maps extracted by the preceding convolutional and pooling layers into a high-level feature vector. Then, depending on the specific task, different output heads are used to generate the charging port feature data. For example, for interface edges, a segmentation head might output the probability of each pixel belonging to the edge, forming an interface edge probability map; for positioning pin holes, a regression head might output the coordinates of their center points, forming a positioning pin hole location map; for brand logos, a classification head might output the probability distribution of their respective categories, forming brand logo label data. This structured feature data provides accurate input for subsequent feature matching and 3D reconstruction. This embodiment achieves accurate and robust extraction of multi-dimensional features of the charging port through standardized data preprocessing, deep feature extraction, attention focusing, and task-specific output heads.

[0042] Step 103: Drive the telescopic mechanism to fine-tune the position of the charging gun head, so that the charging gun head and the charging port are aligned based on precise three-dimensional coordinate data, and control the telescopic mechanism to extend the charging gun head toward the charging port, and detect the contact pressure data between the charging gun head and the charging port through a pressure sensing device.

[0043] In this embodiment, the final precision alignment and contact detection are performed based on accurate positioning information. The system compares the precise three-dimensional coordinate data of the charging port with the real-time pose data of the charging gun head at the end of the telescopic mechanism, and calculates the minute deviation values ​​in the lateral, longitudinal, and elevation directions. Subsequently, the system generates high-precision fine-tuning command data, which includes the displacement amount to be compensated in each direction and the fine adjustment step size.

[0044] For example, the process of fine-tuning the telescopic mechanism is performed by a high-performance stepper motor. It should be noted that the motor adopts microstepping drive technology, which can achieve tiny angular displacements much smaller than its basic step angle. This displacement is then converted into linear displacements with nanometer-level precision through a ball screw pair, driving the charging gun head to perform extremely precise translational movements in three-dimensional space. Ultimately, this ensures that its center is precisely aligned with the center of the charging port, and the deviation is eliminated to a negligible range.

[0045] After alignment, the system controls the main drive components of the telescopic mechanism, causing the charging gun head to move smoothly and controllably along a straight trajectory toward the charging port for insertion. Understandably, during this crucial stage, the pressure sensing device integrated inside the charging gun head begins continuous operation. Specifically, this device typically includes multiple annular pressure sensors evenly distributed on the contact surface of the gun head. These sensors detect the contact pressure data generated between the gun head and the charging port at the moment of contact and during insertion in real time. For example, the contact pressure data includes, but is not limited to, the total pressure value, the distribution of pressure on the contact surface (whether it is uniform), and the rate of change of pressure over time. This real-time data is immediately fed back to the control system as a crucial basis for determining whether the insertion is normal and whether there is a risk of jamming or misalignment. This step, through extremely precise motion control and sensitive monitoring of contact force, ensures a smooth, safe, and reliable insertion process.

[0046] Step 104: Based on the contact pressure data and the preset pressure threshold, determine whether the charging gun head and the charging port are successfully aligned. If not, activate the manual fine-tuning mode to allow users to manually adjust the position of the charging gun head and complete the insertion operation.

[0047] In this embodiment, the system performs comprehensive analysis and processing on the contact pressure data continuously fed back by the pressure sensing device. Specifically, this processing includes calculating the real-time average pressure, monitoring whether the pressure distribution is uniform (e.g., whether there is a situation where the pressure on one side is significantly higher than in other areas), and determining whether the pressure value has stabilized within a preset reasonable time window.

[0048] It should be noted that the system has one or more preset pressure thresholds, which are determined based on extensive experiments and are used to define the mechanical conditions for safe and normal insertion. For example, the system compares real-time calculated pressure state data (such as maximum single-point pressure value, pressure distribution variance, etc.) with these preset pressure thresholds. If the comparison results show that the pressure data is within the allowable range of the thresholds and is uniformly distributed, the system determines that automatic alignment and insertion are successful, the process ends, and the charging process can start normally.

[0049] Understandably, if the judgment results reveal an anomaly, such as a pressure value on one side consistently exceeding the safety threshold, or severely uneven pressure distribution, it indicates the possible existence of minor deviations that have not been completely eliminated, and forced connection carries the risk of damaging the equipment. In this case, the system will automatically generate a manual intervention signal. Specifically, in response to this signal, the control system will first pause the automatic process and de-energize and release the electromagnetic clutch located between the telescopic mechanism and the charging gun head. For example, after the clutch is released, the mechanical connection between the charging gun head and the telescopic arm is disengaged, entering a floating state. The user can then apply a very small force to manually adjust the position of the charging gun head. It should be noted that the operating force in manual mode is much lower than the driving force in automatic mode.

[0050] It should be noted that in manual fine-tuning mode, users can fine-tune the charging gun head with multiple degrees of freedom, such as lateral sliding, pitching, or rotation, based on their actual feel and experience. During this process, pressure sensor data is still monitored in real time and may be fed back to the user visually or audibly. When the user manually adjusts the gun head to the correct position, and the pressure sensor detects that the contact pressure data has fallen back to the normal threshold range and is evenly distributed, the system can automatically, or with user confirmation, control the electromagnetic clutch to re-energize and lock, fixing the current position and finally completing the insertion operation. This step, by introducing a mechanism of automatic judgment and manual backup in coordination, greatly improves the system's robustness and final success rate in the face of complex abnormal situations.

[0051] In this embodiment, after the coarse positioning step is successfully completed and the telescopic mechanism has moved the charging gun head to a rough position, the system automatically triggers the fine positioning step. It should be noted that the system monitors a key indicator in real time during this stage: the recognition success rate. For example, this data can be calculated based on the confidence level of the stereo vision system's continuous recognition of the charging port features, the success rate of feature point matching, or the stability of the calculated 3D coordinates during the fine positioning stage. The system compares this recognition success rate data with a preset success threshold. This threshold is an empirical value used to determine the reliability of the current fine positioning attempt. If the monitored recognition success rate data is lower than the preset success threshold, it indicates that the fine positioning effect under the current viewpoint or conditions is poor; for example, the charging port is partially obscured or the lighting changes suddenly. In this case, the system will not blindly enter the subsequent potentially failed insertion process but will activate an automatic adjustment mechanism. Specifically, the system will control the telescopic mechanism to make minor adjustments to its position or orientation, for example, slightly retracting or slightly deflecting it in the Z-axis direction, and then re-acquire stereo image data of the charging port area at this new position and re-execute the fine positioning algorithm. This process constitutes an iterative processing loop, which attempts to obtain higher quality image data by actively changing the observation conditions until the recognition success rate is raised above the threshold, or after reaching the preset maximum number of iterations, it switches to manual mode for assistance, thereby significantly improving the system's fault tolerance in dealing with complex scenarios.

[0052] Furthermore, during precise positioning and subsequent fine-tuning alignment, the system processes multi-source heterogeneous sensor data using a real-time data fusion algorithm. It should be noted that this data primarily includes image data (or its derived features) provided by the vision system, precise three-dimensional coordinate data of the charging port generated by the coordinate calculation module, and contact pressure data detected by the pressure sensing device. For example, the data fusion algorithm (such as Kalman filtering or Bayesian estimation) can align these data of different properties and acquisition frequencies in time and perform comprehensive processing within a unified state space model. Specifically, it may leverage the high precision but susceptibility to transient interference of visual coordinates to complement the slightly lower precision of pressure data, which directly reflects the physical contact state. Through fusion processing, the system generates comprehensive state data that fully and accurately reflects the real-time progress of the alignment operation. This data not only includes an estimate of positional deviation but may also include an overall assessment of the contact state and the presence of a risk of jamming.

[0053] Based on the aforementioned comprehensive status data, the system executes automatic decisions. This decision-making logic is understood to be based on preset rules or a lightweight machine learning model. Specifically, the system analyzes key indicators in the comprehensive status data, such as whether the positional deviation is less than the final tolerance, whether the pressure data is uniform and within a safe range, and whether the overall trend of the process is stable. For example, if all indicators meet the preset "success conditions," the system automatically decides to continue the final insertion operation; if an anomaly is detected, such as persistently excessive pressure on one side or the positional deviation failing to converge, the system may decide to interrupt the automatic process and initiate manual fine-tuning. Regardless of the system's decision, clear prompts are output to the user through a human-machine interface (such as a touchscreen, status indicator, or voice module), such as "Alignment successful, charging started," or "deviation detected, please manually fine-tune," ensuring the user clearly understands the current status and makes necessary adjustments. This embodiment, by introducing real-time monitoring, iterative optimization, multi-sensor fusion, and intelligent decision-making, significantly improves the adaptability, reliability, and user experience of the entire automatic alignment system.

[0054] In this embodiment, the system establishes a continuous or on-demand connection with a remote cloud platform through a built-in wireless communication module (such as a 4G / 5G cellular module or a Wi-Fi module). It should be noted that during automatic alignment and charging, the system packages and uploads a series of key data to the cloud platform in real time or periodically. This data mainly includes alignment status data reflecting the progress of each stage of the alignment process (such as whether coarse positioning was successful, the number of fine positioning iterations, and the final alignment accuracy), charging data after charging begins (such as charging voltage, current, power level, and duration), and any abnormal or fault data encountered during operation (such as error codes for sensor failure, actuator jamming, etc.). Exemplarily, this data is encrypted and transmitted wirelessly to a cloud server for storage.

[0055] On the cloud platform, the massive amounts of uploaded data received are stored and subjected to in-depth analysis and processing. It's understood that the cloud platform possesses powerful computing and storage capabilities. Specifically, it statistically analyzes the alignment success rate and average time for different vehicle models, generates diagnostic reports on equipment health status to predict potential maintenance needs, and analyzes failure modes to pinpoint common technical issues. Based on these analyses, the cloud platform generates two important types of data: diagnostic reports for maintenance personnel, facilitating remote monitoring and equipment management; and update command data for the equipment itself. This update command data is crucial for the system's self-evolution. Specifically, based on the update command data, the system automatically downloads and integrates new parameters from the vehicle model database from the cloud platform. For example, when a new vehicle model appears on the market that the local database cannot recognize, maintenance personnel can enter the new model's outline template and charging port parameters in the cloud to generate an update command. Upon receiving the command, the system deployed on the charging pile automatically completes the data download and integration during idle periods, thus adapting to the new model without manual on-site upgrades, greatly improving the system's scalability and applicability.

[0056] Finally, at the user interaction level, the system provides convenient access points through local interactive devices. It should be noted that local interactive devices typically include barcode scanners, touchscreens, and speakers. Specifically, users can scan the vehicle's VIN code or a QR code near the charging port to input identification data for service activation. Upon receiving this user's input data, the system can immediately initiate the entire automatic alignment process, eliminating the need for users to make multiple selections on the screen. Simultaneously, throughout the alignment and charging process, the system uses voice synthesis technology to announce key operation progress data to the user in voice format, such as "Vehicle recognition in progress," "Starting precise positioning," "Please wait, connecting," and "Charging complete," providing clear and user-friendly non-visual feedback, especially suitable for scenarios with poor lighting or where users cannot easily focus on the screen. This embodiment, by constructing an end-to-cloud collaborative data pipeline and a user-friendly interactive interface, not only ensures the intelligent and efficient operation of individual devices but also enables remote operation and maintenance and continuous evolution of device clusters, optimizing the end-user experience.

[0057] The above are embodiments of the method proposed in this application. Based on the same inventive concept, embodiments of this application also provide an automatic alignment device for an electric vehicle charging port, the structure of which is as follows: Figure 2 As shown.

[0058] Figure 2 This is a schematic diagram of the internal structure of an automatic alignment device for an electric vehicle charging port, provided as an embodiment of this application. Figure 2 As shown, the device includes: At least one processor; And, a memory that is communicatively connected to at least one processor; The memory stores instructions that can be executed by at least one processor, and the instructions, when executed by at least one processor, enable at least one processor to: The image acquisition device collects the overall outline image data of the vehicle in the parking area, identifies the vehicle model based on the overall outline image data, queries the model database to obtain the charging port location parameters, and calculates the coarse positioning coordinate data of the charging port based on the charging port location parameters. The drive telescopic mechanism moves the charging gun head to the coarse positioning position corresponding to the coarse positioning coordinate data. The stereo vision device acquires stereo image data of the charging port area and identifies the features of the charging port based on the stereo image data to calculate the precise three-dimensional coordinate data of the charging port. The telescopic mechanism is driven to fine-tune the position of the charging gun head, aligning the charging gun head with the charging port based on precise three-dimensional coordinate data. The telescopic mechanism is also controlled to extend the charging gun head toward the charging port, and the contact pressure data between the charging gun head and the charging port is detected by a pressure sensing device. Based on the contact pressure data and the preset pressure threshold, it is determined whether the charging gun head and the charging port are successfully aligned. If not, the manual fine-tuning mode is activated, allowing users to manually adjust the position of the charging gun head to complete the insertion operation.

[0059] This application also provides a non-volatile computer storage medium storing computer-executable instructions, which, when executed, can: The image acquisition device collects the overall outline image data of the vehicle in the parking area, identifies the vehicle model based on the overall outline image data, queries the model database to obtain the charging port location parameters, and calculates the coarse positioning coordinate data of the charging port based on the charging port location parameters. The drive telescopic mechanism moves the charging gun head to the coarse positioning position corresponding to the coarse positioning coordinate data. The stereo vision device acquires stereo image data of the charging port area and identifies the features of the charging port based on the stereo image data to calculate the precise three-dimensional coordinate data of the charging port. The telescopic mechanism is driven to fine-tune the position of the charging gun head, aligning the charging gun head with the charging port based on precise three-dimensional coordinate data. The telescopic mechanism is also controlled to extend the charging gun head toward the charging port, and the contact pressure data between the charging gun head and the charging port is detected by a pressure sensing device. Based on the contact pressure data and the preset pressure threshold, it is determined whether the charging gun head and the charging port are successfully aligned. If not, the manual fine-tuning mode is activated, allowing users to manually adjust the position of the charging gun head to complete the insertion operation.

[0060] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0061] The devices and media provided in this application are one-to-one with the methods. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0062] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0063] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0064] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0065] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0066] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0067] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0068] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0069] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0070] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. An automatic alignment method for an electric vehicle charging port, characterized in that, The method includes: The image acquisition device acquires overall vehicle outline image data of the vehicle parking area, identifies the vehicle model based on the overall vehicle outline image data, queries the model database to obtain charging port location parameters, and calculates coarse positioning coordinate data of the charging port based on the charging port location parameters. The drive telescopic mechanism moves the charging gun head to the coarse positioning position corresponding to the coarse positioning coordinate data. The stereoscopic image data of the charging port area is obtained through the stereoscopic vision device, and the features of the charging port are identified based on the stereoscopic image data to calculate the precise three-dimensional coordinate data of the charging port. Drive the telescopic mechanism to fine-tune the position of the charging gun head, so that the charging gun head and the charging port are aligned based on the precise three-dimensional coordinate data, and control the telescopic mechanism to extend the charging gun head toward the charging port, and detect the contact pressure data between the charging gun head and the charging port through a pressure sensing device; Based on the contact pressure data and the preset pressure threshold, it is determined whether the charging gun head and the charging port are successfully aligned. If not, the manual fine-tuning mode is activated, allowing the user to manually adjust the position of the charging gun head to complete the insertion operation.

2. The automatic alignment method for an electric vehicle charging port according to claim 1, characterized in that, The vehicle model is identified based on the overall vehicle outline image data, and the charging port location parameters are obtained by querying the model database. Coarse positioning coordinate data of the charging port is then calculated based on these parameters, specifically including: The overall vehicle outline image data is preprocessed to generate optimized image data, and an edge detection algorithm is applied to process the optimized image data to extract vehicle outline feature data. The image preprocessing includes noise reduction and contrast enhancement operations, and the vehicle outline feature data includes front lines, rear lines, and wheel position features. The vehicle outline feature data is matched with pre-stored vehicle model templates in the vehicle model database. The vehicle model identifier is determined by similarity calculation. Based on the vehicle model identifier, the vehicle model database is queried to obtain the charging port location parameters. The charging port location parameters include distance data from the front or rear of the vehicle and ground clearance data. By combining the calibration parameters of the image acquisition device, the position parameters of the charging port are processed through a coordinate transformation algorithm to calculate the coarse positioning coordinate data of the charging port; the coarse positioning coordinate data includes the dimensions of lateral distance, ground height, and front-back distance.

3. The automatic alignment method for an electric vehicle charging port according to claim 2, characterized in that, The optimized image data is processed using an edge detection algorithm to extract vehicle contour feature data, specifically including: The optimized image data is converted to grayscale to generate grayscale image data. The Canny edge detection algorithm is then applied to the grayscale image data to perform Gaussian filtering, gradient calculation, and non-maximum suppression to extract preliminary edge data. The preliminary edge data is subjected to contour connection and filtering to generate stable vehicle contour feature data. The stable vehicle contour feature data is then processed by a geometric feature extraction algorithm to calculate key point coordinates and shape descriptors to enhance feature distinguishability.

4. The automatic alignment method for an electric vehicle charging port according to claim 1, characterized in that, Based on the stereoscopic image data, the features of the charging port are identified to calculate the precise three-dimensional coordinate data of the charging port, specifically including: The stereo image data is subjected to image enhancement processing to generate clear stereo image data, and a lightweight convolutional neural network model is applied to process the clear stereo image data to extract charging port feature data; the image enhancement processing includes brightness adjustment and occlusion removal operations, and the charging port feature data includes interface edges, positioning pin holes and brand logo features; Based on the charging port feature data, the corresponding point data of the charging port in the left and right views is determined by the feature matching algorithm, and the corresponding point data is processed by the triangulation algorithm. The accurate three-dimensional coordinate data of the charging port is calculated according to the parallax. The accurate three-dimensional coordinate data includes X, Y and Z direction coordinates. The precise three-dimensional coordinate data is processed by an error correction algorithm to ensure that the coordinate error is within a preset range and to generate the final precise three-dimensional coordinate data.

5. The automatic alignment method for an electric vehicle charging port according to claim 4, characterized in that, The lightweight convolutional neural network model is used to process the clear stereoscopic image data and extract the charging port feature data, specifically including: The clear stereoscopic image data is normalized and scaled to generate standardized image data, and the standardized image data is processed by forward propagation of a lightweight convolutional neural network model to extract multi-level feature map data; the forward propagation process includes convolutional layer, pooling layer and activation layer operations; The multi-level feature map data is processed using an attention mechanism, focusing on the charging port area to generate enhanced feature map data. Based on the enhanced feature map data, charging port feature data is output through a fully connected layer and a Softmax classifier. The charging port feature data includes an interface edge probability map, a positioning pin hole location map, and brand identification label data.

6. The automatic alignment method for an electric vehicle charging port according to claim 1, characterized in that, Based on the contact pressure data and the preset pressure threshold, it is determined whether the charging gun head and the charging port are successfully aligned. If not, the manual fine-tuning mode is activated, allowing the user to manually adjust the position of the charging gun head to complete the insertion operation. Specifically, this includes: Based on the contact pressure data, the pressure distribution and unilateral pressure value are calculated to generate pressure state data, and the pressure state data is compared with a preset pressure threshold. If the pressure value on one side in the pressure status data exceeds the preset pressure threshold, a manual intervention signal is generated, and in response to the manual intervention signal, the system automatically switches to manual fine-tuning mode and releases the electromagnetic clutch to allow the user to adjust the charging gun head without resistance. In manual fine-tuning mode, users can input and adjust the position of the charging gun head until the contact pressure data is lower than the preset pressure threshold to complete the insertion operation; the adjustment of the charging gun head position includes lateral sliding, pitch swinging and rotation operations.

7. The automatic alignment method for an electric vehicle charging port according to claim 1, characterized in that, The method further includes: After coarse positioning is completed, fine positioning is triggered, and recognition success rate data is monitored in real time. If the recognition success rate is lower than the preset success threshold, the position of the telescopic mechanism is automatically adjusted, and stereo image data is re-acquired for iterative processing. During the precise positioning and alignment process, image data, coordinate data, and pressure data are processed through a real-time data fusion algorithm to generate comprehensive status data. Based on the comprehensive status data, the system automatically decides whether to insert the gun or perform manual fine-tuning, and outputs prompt information data through the human-computer interaction unit.

8. The automatic alignment method for an electric vehicle charging port according to claim 1, characterized in that, The method further includes: Alignment status data, charging data, and fault data are uploaded to the cloud platform in real time via a wireless communication module. The uploaded data is then stored and analyzed on the cloud platform to generate diagnostic reports and update instruction data. Based on the update instruction data, the newly added parameters from the vehicle model database are automatically downloaded and integrated to adapt to the new vehicle model; The system receives user input data via QR code scanning through a local interactive device, and initiates an automatic alignment process, along with voice prompts to indicate the operation progress.

9. An automatic alignment device for an electric vehicle charging port, characterized in that, The device includes: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform an automatic alignment method for an electric vehicle charging port as described in any one of claims 1-8.

10. A non-volatile computer storage medium storing computer-executable instructions, characterized in that, When the computer-executable instructions are executed, an automatic alignment method for an electric vehicle charging port as described in any one of claims 1-8 is implemented.

Images