Charging pile use optimization method and system based on big data analysis

Abstract

The application is suitable for the field of charging pile use management, and provides a charging pile use optimization method and system based on big data analysis.The method comprises the following steps: monitoring a specified area in front of a charging pile through a perception array to obtain image data of a vehicle; analyzing the image data to obtain real-time pose data and vehicle size data of the vehicle, and combining a reference parking space model to calculate coordinate parameters of a dynamic virtual parking space line; sending the coordinate parameters to a laser projection device, so that two parallel guide light bands are projected on the ground in front of the charging pile and used to indicate a parking position of the vehicle; and when the vehicle is located in the two guide light bands and a charging gun is unlocked, sending a start instruction to the charging pile controller to activate the charging gun.The application uses flexible guide light bands instead of rigid parking space lines, and uses intelligent interaction instead of forced constraint, effectively solving the problem of pile space virtual consumption caused by non-standard parking.

Description

Technical Field

[0001] This invention relates to the field of charging pile usage management, specifically to a method and system for optimizing charging pile usage based on big data analysis. Background Technology

[0002] With the increasing popularity of new energy vehicles, supporting public charging stations have become an important part of current infrastructure. Consequently, the operation and management of charging stations have become particularly important.

[0003] Currently, car owners encounter numerous problems when using charging stations. For example, while parking spaces are generally clearly marked with corresponding charging station locations, car owners often fail to park their vehicles precisely within the designated area due to factors such as the location of the charging port and their operating habits. This improper parking behavior easily leads to the ineffective occupation of adjacent charging station spaces, resulting in the waste of charging station resources and leaving subsequent vehicles without available charging stations, severely reducing the utilization efficiency of charging stations and the user experience.

[0004] Existing solutions often rely on physical constraints or passive monitoring. For example, they involve installing fixed parking locks or bollards to enforce parking boundaries, or using video surveillance for post-parking reminders and billing. However, these methods often have significant limitations. First, the physical infrastructure lacks flexibility in construction, and even after the physical infrastructure is unlocked, it cannot effectively restrict drivers' more arbitrary parking behavior. Second, passive monitoring lacks real-time intervention capabilities, making it difficult to prevent resource waste. Therefore, this paper proposes a charging pile usage optimization method and system based on big data analysis, aiming to address the aforementioned problems. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a charging pile usage optimization method and system based on big data analysis to solve the problems existing in the above-mentioned background technology.

[0006] This invention is implemented as follows: a charging pile usage optimization method based on big data analysis, the method comprising the following steps:

[0007] The system monitors a designated area in front of the charging station using a sensing array, which includes a visual sensor and a millimeter-wave radar, to obtain image data of the vehicle.

[0008] The real-time pose data and vehicle size data of the vehicle are obtained by analyzing the image data, and the coordinate parameters of the dynamic virtual parking space line are calculated by combining the benchmark parking space model. The benchmark parking space model is generated based on historical parking big data.

[0009] The coordinate parameters are sent to the laser projection device, so that two parallel guide light strips are projected on the ground in front of the charging pile to indicate the parking position of the vehicle.

[0010] When the vehicle is parked, the vehicle's position is detected again by a vision sensor. When the vehicle is within the two guide light strips and the charging gun is unlocked, a start command is sent to the charging pile controller to activate the charging gun.

[0011] Specifically, the sensing array can be a combination of environmental sensing devices integrated on the charging pile itself, including a visual sensor and a millimeter-wave radar; the visual sensor can be a high-definition industrial camera; the millimeter-wave radar can be an automotive-grade millimeter-wave radar; the designated area can be a rectangular monitoring area in front of the charging pile; the image data can be high-definition image frames containing vehicle information collected by the visual sensor; the benchmark parking space model can be a standardized parking space parameter model trained by machine learning algorithms based on historical successful parking big data of the charging station; the coordinate parameters of the dynamic virtual parking space line can be the set of ground projection coordinates required for the laser projection device to project the guide light strip; the laser projection device can be a high-brightness laser projector installed on the top of the charging pile; the guide light strip can be two parallel high-brightness laser projection lines used to define the parking range of the vehicle; the charging pile controller can be a control unit built into the charging pile used to control the activation and locking of the charging gun; the charging gun can be a DC or AC charging gun matched with the charging pile.

[0012] As a further aspect of the present invention: the step of monitoring a designated area in front of a charging pile using a sensing array to obtain image data of the vehicle specifically includes:

[0013] Raw point cloud data is acquired using millimeter-wave radar and used to identify whether any objects matching vehicle characteristics have entered the designated area through point cloud analysis.

[0014] When a vehicle is detected, the visual sensor is activated to acquire high-definition image frames, and the vehicle feature region is extracted using a target detection model.

[0015] Using a pre-defined image classification and recognition model, the image of the vehicle's feature region is detected to determine whether the vehicle is a new energy vehicle.

[0016] When a new energy vehicle is detected, the image data is acquired.

[0017] Specifically, the raw point cloud data can be a set of three-dimensional point clouds collected by millimeter-wave radar that contains information on the spatial position and motion of objects; the target detection model can be a vehicle target detection model trained based on the YOLO algorithm; the vehicle feature region can be a local image region in the image that contains the vehicle body, front face, logo, and license plate; the image classification and recognition model can be a special recognition model for new energy vehicles trained based on convolutional neural networks; the new energy vehicle can be a pure electric vehicle or a plug-in hybrid vehicle.

[0018] As a further aspect of the present invention: the step of analyzing image data to obtain real-time vehicle pose data and vehicle size data, and calculating the coordinate parameters of the dynamic virtual parking space line in conjunction with a reference parking space model, specifically includes:

[0019] The image data is input into a vehicle recognition model trained with big data to obtain the vehicle's brand and model information;

[0020] Based on brand and model information, corresponding standardized vehicle size data is obtained by searching the vehicle model database;

[0021] The real-time pose data of the vehicle is calculated based on the relative geometric relationship between the vehicle and the charging pile in the image data.

[0022] The vehicle size data and real-time pose data are fitted with the reference parking space model to obtain the coordinate parameters of the dynamic virtual parking space line.

[0023] Specifically, the vehicle recognition model can be a ResNet convolutional neural network model trained on massive amounts of vehicle image data; brand and model information can be the vehicle's brand name, specific year, and configuration; the vehicle model database can be a structured database containing standardized parameters such as the brand and model of mainstream new energy vehicles on the market, vehicle width, vehicle length, wheelbase, and charging port location; vehicle size data can be the corresponding vehicle's body width, body length, wheelbase, and other size parameters; relative geometric relationships can be the pixel ratio and angular offset relationship between the vehicle and the charging pile in the image; real-time pose data can be the vehicle's lateral offset, longitudinal distance, and vehicle orientation angle relative to the charging pile; fitting can be the process of substituting the vehicle size and pose data into a benchmark parking space model and generating projected coordinates through coordinate transformation and geometric calculation.

[0024] As a further aspect of the present invention: the step of fitting the vehicle size data, real-time pose data, and the reference parking space model specifically includes:

[0025] Extract the vehicle's width data from the vehicle size data;

[0026] The preset small car width threshold is compared with the width data, and the small car width threshold is used to determine whether the width of the dynamic virtual parking space line should be reduced.

[0027] When the width of the vehicle is less than the width threshold of a small car, the standard spacing in the reference parking space model is scaled according to a preset spacing reduction coefficient to obtain the actual guide spacing that adapts to the current vehicle.

[0028] The coordinate parameters of the adjusted dynamic virtual parking space line are generated by combining real-time pose data and actual guide spacing.

[0029] Specifically, the vehicle width data can be the maximum width value of the vehicle body retrieved from the vehicle model database; the width threshold for small cars can be the width critical value obtained based on the statistical classification of mainstream vehicle models on the market; the spacing reduction coefficient can be a pre-set proportional coefficient used to reduce the spacing of the guide light strips; the standard spacing can be the standard vertical distance between two guide light strips set in the benchmark parking space model; and the actual guide spacing can be the vertical distance between two guide light strips after scaling to fit the current vehicle width.

[0030] As a further aspect of the present invention: the step of re-detecting the vehicle position using a visual sensor specifically includes:

[0031] The system continuously analyzes the deviation between the real-time pose data of the vehicle and the dynamic virtual parking lines, and generates a correction command data stream containing direction and distance deviations based on the analysis results.

[0032] The correction instruction data stream is converted into an auxiliary voice signal and sent to a voice playback device for broadcasting, in order to guide the vehicle into the area corresponding to the guide light strip.

[0033] Once the vehicle stops, the current position and pose data of the vehicle are obtained and the parking status is determined.

[0034] When it is determined that the vehicle is located within the area between the guide light strips, a confirmation voice signal is generated to indicate that the vehicle has been successfully parked and sent to the voice playback device.

[0035] Specifically, the correction instruction data stream can be structured data containing vehicle movement direction guidance and distance deviation values; the auxiliary voice signal can be an audio signal that converts the correction instruction into natural speech; the voice playback device can be a waterproof outdoor speaker installed on the charging pile itself; the determination of the parking status can be a compliance determination based on vehicle pose data and guide light strip coordinates; and the confirmation voice signal can be a natural voice audio signal that prompts the driver that the parking was successful.

[0036] As a further aspect of the present invention, the method further includes:

[0037] When a vehicle enters the guide light strip, the direction of the vehicle's entry is identified through image data acquired by the vision sensor.

[0038] When the rear of the vehicle is detected to be facing the charging pile, a first adjustment mechanism is executed on the coordinate parameters to shift the overall projection position of the two guide light strips a first preset distance toward the charging pile.

[0039] When the vehicle is detected to be facing the charging pile, a second adjustment mechanism is executed on the coordinate parameters to shift the overall projection position of the two guide light strips a second preset distance away from the charging pile.

[0040] The adjusted coordinate parameters are transmitted in real time to the laser projection device so that the driver can observe the guide light strip when parking.

[0041] Specifically, the direction in which the vehicle enters can be either a reverse parking direction with the rear of the vehicle facing the charging pile, or a straight parking direction with the front of the vehicle facing the charging pile; the first adjustment mechanism can be a coordinate transformation mechanism that performs an overall translation of the coordinate parameters of the dynamic virtual parking space line; the first preset distance can be a translation distance calculated based on the vehicle length; the second adjustment mechanism can be a coordinate transformation mechanism that performs a reverse overall translation of the coordinate parameters of the dynamic virtual parking space line; the second preset distance can be a translation distance calculated based on the vehicle length.

[0042] Another object of the present invention is to provide a charging pile usage optimization system based on big data analysis, the system comprising:

[0043] The area perception module is used to monitor a designated area in front of the charging pile through a perception array to obtain image data of the vehicle. The perception array includes a visual sensor and a millimeter-wave radar.

[0044] The pose analysis module is used to analyze image data to obtain real-time pose data and vehicle size data of the vehicle, and calculate the coordinate parameters of the dynamic virtual parking space line by combining the reference parking space model. The reference parking space model is generated based on historical parking big data.

[0045] The projection guidance module is used to send coordinate parameters to the laser projection device, so that two parallel guide light strips are projected on the ground in front of the charging pile to indicate the parking position of the vehicle.

[0046] The authorization confirmation module is used to re-detect the vehicle's position using a visual sensor when the vehicle is parked. When the vehicle is within the two guide light strips and the charging gun is unlocked, it sends a start command to the charging pile controller to activate the charging gun.

[0047] As a further aspect of the present invention: the region sensing module includes:

[0048] The radar screening unit is used to acquire raw point cloud data through millimeter-wave radar and to identify whether there are objects with vehicle characteristics entering the designated area through point cloud analysis.

[0049] The visual confirmation unit is used to activate the visual sensor to acquire high-definition image frames when a vehicle is identified, and to extract the vehicle feature region using a target detection model.

[0050] The vehicle model recognition unit is used to detect the image of the vehicle feature area using a preset image classification and recognition model, in order to determine whether the vehicle is a new energy vehicle.

[0051] The data acquisition unit is used to acquire the image data when the vehicle is detected to be a new energy vehicle.

[0052] As a further aspect of the present invention: the pose analysis module includes:

[0053] The model recognition unit is used to input the image data into a vehicle recognition model trained by big data to obtain the vehicle's brand and model information;

[0054] The size retrieval unit is used to retrieve standardized vehicle size data from the vehicle model database based on brand and model information.

[0055] The pose calculation unit is used to calculate the real-time pose data of the vehicle based on the relative geometric relationship between the vehicle and the charging pile in the image data.

[0056] The parameter fitting unit is used to fit the vehicle size data and real-time pose data with the reference parking space model to obtain the coordinate parameters of the dynamic virtual parking space line.

[0057] As a further aspect of the present invention: the system further includes a guidance adjustment module, which includes:

[0058] A direction recognition unit is used to identify the direction in which a vehicle enters the guide light strip by using image data acquired by the vision sensor.

[0059] The rear-end mode adjustment unit is used to perform a first adjustment mechanism on the coordinate parameters when the rear of the vehicle is detected to be facing the charging pile, so that the overall projection position of the two guide light strips is shifted a first preset distance toward the charging pile.

[0060] The vehicle front mode adjustment unit is used to perform a second adjustment mechanism on the coordinate parameters when the vehicle front is detected to be facing the charging pile, so as to shift the overall projection position of the two guide light strips away from the charging pile by a second preset distance.

[0061] The dynamic projection unit is used to transmit the adjusted coordinate parameters to the laser projection device in real time, so that the driver can observe the guide light strip when parking.

[0062] Compared with the prior art, the beneficial effects of the present invention are:

[0063] This invention uses a sensing array to detect parking behavior during vehicle charging in real time and generates a guide light strip based on the specific vehicle model. This guide light strip indicates the vehicle's parking position, and the charging gun is only activated after the vehicle is correctly parked. Firstly, the laser projection method of this invention reduces the sense of constraint that traditional parking lines impose on drivers, while also saving on infrastructure costs at charging stations. It transforms parking into a more interactive activity, significantly improving the user experience of charging stations. In summary, this invention frees charging stations from reliance on physical infrastructure, instead guiding and constraining behavior through dynamic and intelligent interaction. It balances standardization and user experience, using flexible guide light strips instead of rigid parking lines and intelligent interaction instead of forced constraints. This effectively solves the problem of wasted charging spaces due to improper parking, ensuring that all charging stations within the station can be used effectively and in a standardized manner. Attached Figure Description

[0064] Figure 1 This is a flowchart of a charging pile usage optimization method based on big data analysis.

[0065] Figure 2 This is a flowchart illustrating the process of acquiring vehicle image data in an optimization method for charging pile usage based on big data analytics.

[0066] Figure 3 This is a flowchart illustrating the coordinate parameters of the dynamic virtual parking space line obtained in a charging pile usage optimization method based on big data analysis.

[0067] Figure 4 This is a flowchart illustrating the process of fitting vehicle size data, real-time pose data, and the benchmark parking space model in a charging pile usage optimization method based on big data analysis.

[0068] Figure 5 This is a flowchart illustrating the process of re-detecting vehicle position using visual sensors in a charging pile usage optimization method based on big data analysis.

[0069] Figure 6 This is a flowchart illustrating the real-time transmission of adjusted coordinate parameters to a laser projection device in a charging pile usage optimization method based on big data analysis.

[0070] Figure 7 This is a schematic diagram of a charging pile usage optimization system based on big data analysis.

[0071] Figure 8 This is a schematic diagram of the area perception module in a charging pile usage optimization system based on big data analysis.

[0072] Figure 9 This is a schematic diagram of the pose analysis module in a charging pile usage optimization system based on big data analysis.

[0073] Figure 10 This is a schematic diagram of the guidance and adjustment module in a charging pile usage optimization system based on big data analysis. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0075] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0076] like Figure 1 As shown in the figure, this invention provides a method for optimizing the use of charging piles based on big data analysis. The method includes the following steps:

[0077] S100 monitors a designated area in front of a charging pile using a sensing array to obtain image data of the vehicle. The sensing array includes a visual sensor and a millimeter-wave radar.

[0078] S200: Based on the image data, the real-time pose data and vehicle size data of the vehicle are obtained by analysis, and the coordinate parameters of the dynamic virtual parking space line are calculated by combining the reference parking space model. The reference parking space model is generated based on historical parking big data.

[0079] S300 sends coordinate parameters to the laser projection device, so that two parallel guide light strips are projected on the ground in front of the charging pile to indicate the parking position of the vehicle.

[0080] When the vehicle is parked, the S400 uses a vision sensor to detect the vehicle's position again. When the vehicle is within the two guide light strips and the charging gun is unlocked, it sends a start command to the charging pile controller to activate the charging gun.

[0081] It should be noted that the data used in the benchmark parking space model can come from the historical successful parking data accumulated by the charging pile operation platform itself (including the precise location of the vehicle's final stop, vehicle model, timestamp, etc.). Then, machine learning algorithms (such as regression analysis and clustering algorithms) are used to statistically analyze the massive parking poses to find the optimal parking position range for different vehicle models. In addition, to ensure the visibility and guidance effect of the guide light strip in various complex environments, a color adaptation mechanism is also required. Before projecting the guide light strip through the laser projection device, the visual sensor will first collect the environmental background image of the ground in front of the charging pile and analyze the main color tone and brightness of the area in real time. Based on the analysis results, the color contrast model will automatically calculate and select a color value that has the most significant contrast with the current background, generate an instruction containing this specific color information, and then send the color instruction and parameter coordinates to the laser projection device simultaneously, so that it projects a high-contrast guide light strip with adaptive color adjustment, thereby significantly improving the driver's visual recognition efficiency under different lighting and ground conditions.

[0082] In this embodiment of the invention, a sensing array is used to detect parking behavior during vehicle charging in real time. A guide light strip is generated based on the specific vehicle model to indicate the vehicle's parking position. The charging gun of the charging station is only activated after the vehicle is correctly parked. Firstly, the laser projection method of this invention improves the sense of constraint that traditional parking lines impose on drivers, while also saving on infrastructure costs at charging stations. It transforms the parking behavior of car owners into a more interactive activity, greatly enhancing the user experience of charging stations. In summary, this invention allows charging stations to no longer rely on physical facilities, but instead guides and constrains behavior through dynamic and intelligent interaction, balancing standardization and user experience. The use of flexible guide light strips instead of rigid parking lines and intelligent interaction instead of mandatory constraints effectively solves the problem of wasted charging spaces due to improper parking, ensuring that all charging stations within the station can be used effectively and in a standardized manner.

[0083] like Figure 2 As shown, in a preferred embodiment of the present invention, the step of monitoring a designated area in front of the charging pile using a sensing array to obtain image data of the vehicle specifically includes:

[0084] S101, acquire raw point cloud data through millimeter-wave radar, and use point cloud analysis to identify whether there are objects with vehicle characteristics entering the designated area;

[0085] S102, when a vehicle is identified, the visual sensor is activated to acquire high-definition image frames, and the vehicle feature region is extracted using the target detection model;

[0086] S103 uses a preset image classification and recognition model to detect images of vehicle feature regions to determine whether the vehicle is a new energy vehicle.

[0087] S104, when the vehicle is detected to be a new energy vehicle, the image data acquisition is performed.

[0088] In this embodiment of the invention, the first step is to predict whether a vehicle will be charged. A millimeter-wave radar performs a preliminary scan and perception of the area in front of the charging station, acquiring point cloud data composed of numerous spatial points. This data reflects the position, speed, and basic size of objects. By analyzing the distribution characteristics of this point cloud, a reliable preliminary determination can be made as to whether a vehicle has entered the designated area. When the radar detects a vehicle entering, it triggers and activates a cooperating visual sensor (usually a high-definition camera) to capture a detailed image, obtaining a high-definition frame of the scene (i.e., a high-definition image frame). A trained target detection model is then used to accurately extract the feature region where the vehicle is located from the high-definition image. By identifying key visual features such as the vehicle's grille and license plate, it can be accurately determined whether the vehicle is a new energy vehicle. Only after confirming it is a new energy vehicle does subsequent image data acquisition proceed. If a gasoline-powered vehicle is detected, only recording is performed, and no guidance service is initiated.

[0089] like Figure 3 As shown, in a preferred embodiment of the present invention, the step of analyzing image data to obtain real-time vehicle pose data and vehicle size data, and calculating the coordinate parameters of the dynamic virtual parking space line in conjunction with a reference parking space model, specifically includes:

[0090] S201, Input the image data into a vehicle recognition model trained with big data to obtain the vehicle's brand and model information;

[0091] S202, based on brand and model information, obtains corresponding standardized vehicle size data by searching the vehicle model database;

[0092] S203, calculate the real-time pose data of the vehicle based on the relative geometric relationship between the vehicle and the charging pile in the image data;

[0093] S204, The vehicle size data and real-time pose data are fitted with the reference parking space model to obtain the coordinate parameters of the dynamic virtual parking space line.

[0094] In this embodiment of the invention, while vehicle dimensions can generally be obtained through visual inspection, there are often discrepancies. Therefore, this embodiment employs an alternative method to determine the specific dimensions of a vehicle. First, a vehicle recognition model is used to identify the vehicle in the image data. For example, features such as the vehicle logo, front grille, headlight design, and body lines are used to determine the specific brand and model of the vehicle. Then, a database query provides highly accurate dimensions of the vehicle, such as width, length, height, and wheelbase. Simultaneously, using geometric principles in computer vision (such as monocular vision ranging technology), the vehicle's current pose relative to the charging pile is calculated in real time by analyzing the pixel ratios and angles between a known-sized reference object and the vehicle in the image. This pose data includes the vehicle's position and orientation. Finally, the standardized vehicle dimensions obtained from the query, the real-time calculated vehicle pose, and a benchmark parking space model optimized based on a large amount of historical successful parking data are intelligently fitted. This benchmark parking space model defines core parameters such as the ideal parking space orientation and the distance range from the charging pile. By calculating and generating a set of coordinate parameters for controlling the laser projection of a dynamic virtual parking space line, it can be ensured that the guide light strip is the most reasonable parking space specifically customized for the current size and real-time position of the vehicle.

[0095] like Figure 4 As shown in the preferred embodiment of the present invention, the step of fitting the vehicle size data, real-time pose data, and the reference parking space model specifically includes:

[0096] S214, Extract the vehicle width data from the vehicle size data;

[0097] S224, compare the preset small car width threshold with the width data, the small car width threshold is used to determine whether the width of the dynamic virtual parking space line should be reduced;

[0098] S234, when the width of the vehicle is less than the width threshold of a small car, the standard spacing in the reference parking space model is scaled according to a preset spacing reduction coefficient to obtain the actual guide spacing that adapts to the current vehicle.

[0099] S244 generates the adjusted coordinate parameters of the dynamic virtual parking space line by combining real-time pose data and actual guide spacing.

[0100] In this embodiment of the invention, the width of vehicles is generally relatively uniform, so the variation range of the spacing of the generated guide light strips is small. However, there are special cases. Some existing new energy vehicles are smaller in size. If the guide light strips are not adjusted, it means that the space utilization rate of the charging station will decrease. Therefore, this embodiment first compares the obtained vehicle width with a preset threshold for small car width (e.g., 1.85 meters, which is based on the classification statistics of common vehicle models). If the current vehicle width is less than this threshold, the vehicle is identified as a small car. Then, a preset spacing reduction coefficient is used to scale the spacing of the guide light strips, thereby obtaining a narrower but more adaptable actual guide spacing. This method ensures that small cars can be parked in a more compact space, while large vehicles still use the standard spacing, thereby improving the overall parking capacity efficiency of the charging station. Finally, a fitting process is performed. It should be noted that the fitting is achieved through conventional coordinate transformation and geometric calculation techniques (e.g., converting the spacing parameters into absolute coordinates of the ground projection based on the vehicle's current position and orientation). The techniques used are relatively conventional, so they will not be described in detail.

[0101] like Figure 5 As shown in the preferred embodiment of the present invention, the step of re-detecting the vehicle position using a visual sensor specifically includes:

[0102] S401, continuously analyze the deviation between the real-time pose data of the vehicle and the dynamic virtual parking space line, and generate a correction instruction data stream containing direction and distance deviations based on the analysis results;

[0103] S402, the correction instruction data stream is converted into an auxiliary voice signal and sent to a voice playback device for broadcasting, in order to guide the vehicle into the area corresponding to the guide light strip;

[0104] S403: When the vehicle stops, the current position and pose data of the vehicle are obtained and the parking status is determined.

[0105] S404, when it is determined that the vehicle is located in the area between the guide light strips, a confirmation voice signal is generated to indicate that the vehicle has been successfully parked and sent to the voice playback device.

[0106] In this embodiment of the invention, during the parking process, the system continuously compares the real-time pose data of the vehicle calculated by the visual sensor with the coordinate parameters of the dynamic virtual parking space line, calculating the deviations in the lateral, longitudinal, and angular directions. Based on the calculation results, a structured correction command data stream is generated. This data stream includes operational guidance in the vehicle's movement direction, which is then converted into a natural and fluent auxiliary voice signal and broadcast through a voice playback device installed on the charging pile. This provides the driver with clear and intuitive auditory guidance, assisting them in smoothly driving the vehicle into the area defined by the ground guide light strip. After the vehicle comes to a complete stop, the vehicle's position is identified and determined again. As long as the vehicle is within the guide light strip, a voice signal indicating successful parking is generated. For example, the voice playback device can play the voice message "Parking successful, please connect the power supply gun." Of course, if the vehicle does not deviate from the guide light strip or the guide light strip appears on the vehicle body, it can be identified as improperly parked, and a voice message indicating unsuccessful parking can also be played.

[0107] like Figure 6 As shown in the preferred embodiment of the present invention, the charging pile usage optimization method based on big data analysis further includes:

[0108] S501, when the vehicle enters the guide light strip, the direction of the vehicle's entry is identified by the image data acquired by the vision sensor;

[0109] S502, when it is detected that the rear of the vehicle is facing the charging pile, the first adjustment mechanism is executed on the coordinate parameters to shift the overall projection position of the two guide light strips a first preset distance towards the charging pile.

[0110] S503, when it is detected that the front of the vehicle is facing the charging pile, the second adjustment mechanism is executed on the coordinate parameters to shift the overall projection position of the two guide light strips away from the charging pile by a second preset distance;

[0111] S504 transmits the adjusted coordinate parameters to the laser projection device in real time, so that the driver can observe the guide light strip when parking.

[0112] In this embodiment of the invention, although the spacing of the guide light strips is generally greater than the width of the vehicle, in reality, it is difficult for the driver to directly observe the guide light strips on the ground, just as the driver cannot directly observe the parking lines from the side windows. Of course, if the vehicle itself has a panoramic image, this problem can be solved, but not all vehicles have this function. Therefore, this embodiment can optimize the relative position of the guide light strips according to different parking methods (reversing into the parking space or straight in), so as to ensure that the driver can clearly observe the guide lines on the ground through the conventional field of vision such as the rearview mirror. When the vehicle enters, the visual sensor identifies the visual difference between the front of the vehicle (usually with unique headlights and grille features) and the rear of the vehicle (usually with taillights and trunk features) to determine whether the vehicle is driving in the direction of the charging pile (reversing into the parking space) or the front of the vehicle is driving in the direction of the charging pile (straight in). The guide light strips are adjusted in different ways for different situations. When the vehicle is reversing, the pre-calculated coordinate parameters of the dynamic virtual parking space lines are mathematically transformed. This causes the two parallel guide light strips to be simultaneously shifted a pre-set distance (calculated based on the vehicle length) towards the charging station when projected. This adjustment leverages the fact that drivers primarily observe the ground behind the vehicle through their rearview mirrors, aligning the light strips within the mirror's field of vision. Conversely, when the vehicle enters from the front, the guide light strips are shifted away from the charging station, allowing the driver to still observe them in the rearview mirror. This method ensures that the crucial guide light strips are always within optimal visibility, regardless of the parking method used, significantly reducing parking difficulty and improving universality and user experience.

[0113] like Figure 7 As shown, this embodiment of the invention also provides a charging pile usage optimization system based on big data analysis, the system comprising:

[0114] The area perception module 100 is used to monitor a designated area in front of the charging pile through a perception array to obtain image data of the vehicle. The perception array includes a visual sensor and a millimeter-wave radar.

[0115] The pose analysis module 200 is used to analyze image data to obtain real-time pose data and vehicle size data of the vehicle, and calculate the coordinate parameters of the dynamic virtual parking space line by combining the reference parking space model. The reference parking space model is generated based on historical parking big data.

[0116] The projection guidance module 300 is used to send coordinate parameters to the laser projection device so that two parallel guide light strips are projected on the ground in front of the charging pile to indicate the parking position of the vehicle.

[0117] The authorization module 400 is used to re-detect the vehicle's position using a vision sensor when the vehicle is parked. When the vehicle is within the two guide light strips and the charging gun is unlocked, it sends a start command to the charging pile controller to activate the charging gun.

[0118] like Figure 8 As shown, in a preferred embodiment of the present invention, the area sensing module 100 includes:

[0119] The radar screening unit 101 is used to acquire raw point cloud data through millimeter-wave radar and to identify whether there are objects with vehicle characteristics entering the designated area through point cloud analysis.

[0120] The visual confirmation unit 102 is used to activate the visual sensor to acquire high-definition image frames and extract vehicle feature regions using a target detection model when a vehicle is identified.

[0121] The vehicle model recognition unit 103 is used to detect images of vehicle feature areas using a preset image classification and recognition model, in order to determine whether the vehicle is a new energy vehicle.

[0122] The data acquisition unit 104 is used to acquire the image data when the vehicle is detected to be a new energy vehicle.

[0123] like Figure 9 As shown, in a preferred embodiment of the present invention, the pose analysis module 200 includes:

[0124] The model recognition unit 201 is used to input the image data into a vehicle recognition model trained by big data to obtain the vehicle's brand and model information;

[0125] The size retrieval unit 202 is used to retrieve corresponding standardized vehicle size data from the vehicle model database based on brand and model information.

[0126] The pose calculation unit 203 is used to calculate the real-time pose data of the vehicle based on the relative geometric relationship between the vehicle and the charging pile in the image data.

[0127] The parameter fitting unit 204 is used to fit the vehicle size data and real-time pose data with the reference parking space model to obtain the coordinate parameters of the dynamic virtual parking space line.

[0128] like Figure 10 As shown in a preferred embodiment of the present invention, the charging pile usage optimization system based on big data analysis further includes a guidance adjustment module 500, which includes:

[0129] The direction recognition unit 501 is used to identify the direction of the vehicle's entry into the guide light strip by using image data acquired by the vision sensor.

[0130] The rear-end mode adjustment unit 502 is used to perform a first adjustment mechanism on the coordinate parameters when the rear of the vehicle is detected to be facing the charging pile, so as to shift the overall projection position of the two guide light strips by a first preset distance in the direction closer to the charging pile.

[0131] The vehicle head mode adjustment unit 503 is used to perform a second adjustment mechanism on the coordinate parameters when the vehicle head is detected to be facing the charging pile, so as to shift the overall projection position of the two guide light strips away from the charging pile by a second preset distance.

[0132] The dynamic projection unit 504 is used to transmit the adjusted coordinate parameters to the laser projection device in real time, so that the driver can observe the guide light strip when parking.

[0133] The above description only details the preferred embodiments of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0134] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0135] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0136] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the disclosure in the specification and embodiments. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

Claims

1. A method for optimizing the use of charging piles based on big data analysis, characterized in that, The method includes the following steps: The system monitors a designated area in front of the charging station using a sensing array, which includes a visual sensor and a millimeter-wave radar, to obtain image data of the vehicle. The real-time pose data and vehicle size data of the vehicle are obtained by analyzing the image data, and the coordinate parameters of the dynamic virtual parking space line are calculated by combining the benchmark parking space model. The benchmark parking space model is generated based on historical parking big data. The coordinate parameters are sent to the laser projection device, so that two parallel guide light strips are projected on the ground in front of the charging pile to indicate the parking position of the vehicle. When the vehicle is parked, the vehicle's position is detected again by a vision sensor. When the vehicle is within the two guide light strips and the charging gun is unlocked, a start command is sent to the charging pile controller to activate the charging gun.

2. The charging pile usage optimization method based on big data analysis according to claim 1, characterized in that, The step of monitoring a designated area in front of the charging pile using a sensing array to obtain image data of the vehicle specifically includes: Raw point cloud data is acquired using millimeter-wave radar and used to identify whether any objects matching vehicle characteristics have entered the designated area through point cloud analysis. When a vehicle is detected, the visual sensor is activated to acquire high-definition image frames, and the vehicle feature region is extracted using a target detection model. Using a pre-defined image classification and recognition model, the image of the vehicle's feature region is detected to determine whether the vehicle is a new energy vehicle. When a new energy vehicle is detected, the image data is acquired.

3. The method for optimizing the use of charging piles based on big data analysis according to claim 1, characterized in that, The steps of analyzing image data to obtain real-time vehicle pose data and vehicle size data, and calculating the coordinate parameters of the dynamic virtual parking space line in conjunction with a reference parking space model, specifically include: The image data is input into a vehicle recognition model trained with big data to obtain the vehicle's brand and model information; Based on brand and model information, corresponding standardized vehicle size data is obtained by searching the vehicle model database; The real-time pose data of the vehicle is calculated based on the relative geometric relationship between the vehicle and the charging pile in the image data. The vehicle size data and real-time pose data are fitted with the reference parking space model to obtain the coordinate parameters of the dynamic virtual parking space line.

4. The charging pile usage optimization method based on big data analysis according to claim 3, characterized in that, The step of fitting the vehicle size data, real-time pose data, and the reference parking space model specifically includes: Extract the vehicle's width data from the vehicle size data; The preset small car width threshold is compared with the width data, and the small car width threshold is used to determine whether the width of the dynamic virtual parking space line should be reduced. When the width of the vehicle is less than the width threshold of a small car, the standard spacing in the reference parking space model is scaled according to a preset spacing reduction coefficient to obtain the actual guide spacing that adapts to the current vehicle. The coordinate parameters of the adjusted dynamic virtual parking space line are generated by combining real-time pose data and actual guide spacing.

5. The charging pile usage optimization method based on big data analysis according to claim 1, characterized in that, The step of re-detecting the vehicle position using a visual sensor specifically includes: The system continuously analyzes the deviation between the real-time pose data of the vehicle and the dynamic virtual parking lines, and generates a correction command data stream containing direction and distance deviations based on the analysis results. The correction instruction data stream is converted into an auxiliary voice signal and sent to a voice playback device for broadcasting, in order to guide the vehicle into the area corresponding to the guide light strip. Once the vehicle stops, the current position and pose data of the vehicle are obtained and the parking status is determined. When it is determined that the vehicle is located within the area between the guide light strips, a confirmation voice signal is generated to indicate that the vehicle has been successfully parked and sent to the voice playback device.

6. The method for optimizing the use of charging piles based on big data analysis according to claim 1, characterized in that, The method further includes: When a vehicle enters the guide light strip, the direction of the vehicle's entry is identified through image data acquired by the vision sensor. When the rear of the vehicle is detected to be facing the charging pile, a first adjustment mechanism is executed on the coordinate parameters to shift the overall projection position of the two guide light strips a first preset distance toward the charging pile. When the vehicle is detected to be facing the charging pile, a second adjustment mechanism is executed on the coordinate parameters to shift the overall projection position of the two guide light strips a second preset distance away from the charging pile. The adjusted coordinate parameters are transmitted in real time to the laser projection device so that the driver can observe the guide light strip when parking.

7. A charging pile usage optimization system based on big data analysis, characterized in that, The system includes: The area perception module is used to monitor a designated area in front of the charging pile through a perception array to obtain image data of the vehicle. The perception array includes a visual sensor and a millimeter-wave radar. The pose analysis module is used to analyze image data to obtain real-time pose data and vehicle size data of the vehicle, and calculate the coordinate parameters of the dynamic virtual parking space line by combining the reference parking space model. The reference parking space model is generated based on historical parking big data. The projection guidance module is used to send coordinate parameters to the laser projection device, so that two parallel guide light strips are projected on the ground in front of the charging pile to indicate the parking position of the vehicle. The authorization confirmation module is used to re-detect the vehicle's position using a visual sensor when the vehicle is parked. When the vehicle is within the two guide light strips and the charging gun is unlocked, it sends a start command to the charging pile controller to activate the charging gun.

8. The charging pile usage optimization system based on big data analysis according to claim 1, characterized in that, The region sensing module includes: The radar screening unit is used to acquire raw point cloud data through millimeter-wave radar and to identify whether there are objects with vehicle characteristics entering the designated area through point cloud analysis. The visual confirmation unit is used to activate the visual sensor to acquire high-definition image frames when a vehicle is identified, and to extract the vehicle feature region using a target detection model. The vehicle model recognition unit is used to detect the image of the vehicle feature area using a preset image classification and recognition model, in order to determine whether the vehicle is a new energy vehicle. The data acquisition unit is used to acquire the image data when the vehicle is detected to be a new energy vehicle.

9. The charging pile usage optimization system based on big data analysis according to claim 1, characterized in that, The pose analysis module includes: The model recognition unit is used to input the image data into a vehicle recognition model trained by big data to obtain the vehicle's brand and model information; The size retrieval unit is used to retrieve standardized vehicle size data from the vehicle model database based on brand and model information. The pose calculation unit is used to calculate the real-time pose data of the vehicle based on the relative geometric relationship between the vehicle and the charging pile in the image data. The parameter fitting unit is used to fit the vehicle size data and real-time pose data with the reference parking space model to obtain the coordinate parameters of the dynamic virtual parking space line.

10. The charging pile usage optimization system based on big data analysis according to claim 1, characterized in that, The system also includes a guide adjustment module, which includes: A direction recognition unit is used to identify the direction in which a vehicle enters the guide light strip by using image data acquired by the vision sensor. The rear-end mode adjustment unit is used to perform a first adjustment mechanism on the coordinate parameters when the rear of the vehicle is detected to be facing the charging pile, so that the overall projection position of the two guide light strips is shifted a first preset distance toward the charging pile. The vehicle front mode adjustment unit is used to perform a second adjustment mechanism on the coordinate parameters when the vehicle front is detected to be facing the charging pile, so as to shift the overall projection position of the two guide light strips away from the charging pile by a second preset distance. The dynamic projection unit is used to transmit the adjusted coordinate parameters to the laser projection device in real time, so that the driver can observe the guide light strip when parking.

Images