Battery replacement station vehicle in-place identification method and system, storage medium and battery replacement station

Abstract

The invention belongs to the technical field of visual target detection and positioning, and particularly relates to a battery swap station vehicle in-place identification method and system, a storage medium and a battery swap station. The method comprises the following steps: calibrating a battery replacement limit area and determining the standard length and width of the battery pack of the same specification according to the limit battery replacement position and the standard battery replacement position of the battery pack of the same specification in a battery replacement station; the method comprises the following steps: establishing a battery pack detection and positioning optimal model by using information of battery packs of the same specification, acquiring an image of a to-be-detected battery pack, obtaining positioning information by using the model, completing boundary detection and horizontal inclination angle detection of the to-be-detected battery pack, and determining the battery pack according to boundary detection and horizontal inclination angle detection results. Therefore, whether the current pose of the to-be-detected battery pack meets the battery replacement pose requirement or not is judged. According to the invention, the problem of low battery replacement efficiency caused by different battery packs in a battery replacement station is effectively solved, and compared with the existing multi-sensor joint detection parking guide mode, the cost is lower, the operation is convenient and rapid, and the method has significant advantages.

Description

Technical Field

[0001] The present invention belongs to the technical field of visual target detection and positioning, and specifically relates to a method and system for identifying the arrival of vehicles at a battery swap station, a storage medium, and a battery swap station. Background Art

[0002] At present, the parking guidance scheme in the battery swap station is mainly to install parking limit devices in the driving channel of the battery swap station, including speed bumps laid on the driving channel of the battery swap station or parking lines marked on the side wall of the driving channel of the battery swap station; the installation of speed bumps requires that the ground of the driving channel of the battery swap station must be hardened, and the speed bumps are installed protruding above the plane of the battery swap channel, which can be easily damaged by being stepped on, and some battery swap stations do not have the conditions to harden the road surface due to construction period and operational requirements; the parking line method marked on the side wall of the driving channel of the battery swap station cannot currently meet the parking guidance needs of vehicles of different models.

[0003] Moreover, when facing battery packs of different battery swapping vehicles, battery swapping efficiency is still low. Existing battery swapping stations use multi-sensor joint detection to guide parking to improve battery swapping efficiency, resulting in high construction costs of battery swapping stations. Therefore, a new method for identifying the arrival of battery swapping vehicles is urgently needed to solve the problem of low battery swapping efficiency caused by difficulty in guiding vehicle parking at existing battery swapping stations.

[0004] In view of this, the present invention is proposed. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and to provide a method and system for identifying the arrival of a vehicle at a battery swap station, a storage medium, and a battery swap station.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] On the one hand, the present invention provides a method for identifying a vehicle at a battery swap station, and the specific steps are as follows:

[0008] Step 1: Capture a first image of a vehicle with a battery pack of the same specification in a battery swap station at the same shooting distance and shooting angle;

[0009] Step 2: Based on the first image, obtain the coordinate information of the battery replacement limit boundary corresponding to the battery pack of the same specification, as well as the length k1 and width k2 of the battery pack of the same specification;

[0010] Step 3: Establish a target detection model, divide the first image into a training set and a test set after preprocessing, train the target detection model and optimize the parameters to obtain the optimal target detection model;

[0011] Step 4: Using the optimal target detection model to identify the battery pack to be detected and obtain the positioning information of the battery pack to be detected, the image of the battery pack to be detected is captured at the same shooting distance and shooting angle; the positioning information of the battery pack to be detected is the coordinates of the upper left corner and lower right corner pixels of the vertical rectangular frame used to determine the positioning contour of the battery pack;

[0012] Step 5: Use the positioning information of the battery pack to be tested in step 4, the coordinate information of the battery swap limit boundary in step 2, and the length k1 and width k2 of the battery pack to be tested to complete the posture calculation of the battery pack to be tested, and determine whether the posture of the battery pack to be tested meets the battery swap posture requirements;

[0013] Step 6: According to the above judgment result, a corresponding prompt message is issued to the driver.

[0014] Specifically, under the same shooting distance and shooting angle and the same shooting camera, the standard length and width of the battery pack of the same specification corresponding to the standard battery swapping position in the battery swap station are consistent with the length and width of the battery pack to be tested with a certain inclination angle. Furthermore, the battery pack of the same specification and the battery pack to be tested have the same size specifications, and are either installed on the same model of vehicle or on different models of vehicles.

[0015] Specifically, in step 1, the first image includes image A of a battery pack of the same specification of a vehicle with a battery pack of the same specification at the extreme position of battery replacement in the battery swap station at the same shooting distance and shooting angle, image B of a battery pack of the same specification of a vehicle with a battery pack of the same specification continuously entering and exiting the battery swap station, and image C of a battery pack of the same specification of a vehicle with a battery pack of the same specification at the standard parking position in the battery swap station.

[0016] Furthermore, the battery swap limit positions of vehicles with battery packs of the same specification in the battery swap station are: the front and rear lines perpendicular to the driving direction of vehicles with batteries of the same specification in the battery swap channel, specifically, the front and rear limit positions of the battery packs of the same specification corresponding to the front and rear limit positions of the vehicles parked in the driving direction of the battery swap channel;

[0017] Furthermore, the standard parking position of vehicles with battery packs of the same specifications in the battery swap station is as follows: the battery center point of the vehicle with battery packs of the same specifications is located directly below the shooting angle, and the posture of the battery swap vehicle has no angle with the horizontal driving channel in the battery swap station.

[0018] Specifically, the specific method of step 2 is as follows:

[0019] Step 2.1, determine the pixel position corresponding to the battery replacement limit position of the battery pack of the same specification in image A, and further determine the position of the boundary detection pixel point;

[0020] Step 2.2, determine the pixel position of the battery pack of the same specification in the C image, and further determine the length k1 and width k2 of the battery pack of the same specification in the C image.

[0021] Specifically, in step 3, the preprocessing includes: image annotation and image preprocessing; the image annotation includes: marking the outline of the battery pack of the same specification in the B image with a rectangular frame; the image preprocessing includes: enhancing the image of the battery pack of the same specification through resizing, multi-environment simulation and color change technology.

[0022] Specifically, in step 3, the target detection model is an improved YOLOv11 model;

[0023] The improved YOLOv11 model includes an improvement mechanism for the feature fusion layer structure of the convolution, backbone network, and Neck parts in the YOLOv11 basic model, or an improvement mechanism for the feature fusion layer structure of the fusion backbone network and Neck.

[0024] Specifically, the specific method of step 5 is:

[0025] Step 5.1, based on the positioning information of the battery pack to be detected, determine whether the battery swapping vehicle is in a stationary state that satisfies the battery pack to be detected replacement posture calculation;

[0026] Step 5.2: Under the vehicle state that satisfies the battery pack replacement posture calculation, determine whether the battery pack to be detected exceeds the battery replacement limit boundary by using the positioning information of the battery pack to be detected and the coordinate information of the battery replacement limit boundary;

[0027] Step 5.3, using the positioning information of the battery pack to be detected and the length k1 and width k2 of the battery pack to be detected, calculate the horizontal tilt angle θ0 of the battery pack to be detected;

[0028] Step 5.4: Based on the judgment result of step 5.1 and the ratio of the horizontal tilt angle θ0 of the battery pack to be tested to the preset horizontal critical angle θ, determine whether the battery pack to be tested meets the battery replacement posture requirements. When the battery pack to be tested does not exceed the battery replacement limit boundary and the horizontal tilt angle θ0<θ, the posture of the battery pack to be tested meets the battery replacement posture requirements.

[0029] Specifically, in step 5.3, the specific method for calculating the horizontal tilt angle θ0 of the battery pack to be tested is:

[0030] Step 5.3.1, based on the positioning information of the battery pack to be detected, determine the length L3 and width L4 of the rectangular frame of the battery pack positioning outline;

[0031] Step 5.3.2, based on the intersection of the vertex of the battery pack to be detected with the long side and the wide side of the rectangular frame of the battery pack positioning contour, take the intersection point as the segmentation point, assume that a distance intersecting with the long side of the rectangular frame of the battery pack positioning contour is k6, and another distance is k5, and the corresponding distance intersecting with the wide side of the rectangular frame of the battery pack positioning contour is k3, and another distance is k4;

[0032] Step 5.3.3, based on the fact that the structure of the battery pack to be tested and the positioning contour of the battery pack are both rectangular, k6 is calculated according to Formula 1:

[0033]

[0034] The horizontal tilt angle θ0 of the battery pack to be tested is determined by equation 2:

[0035]

[0036] On the other hand, the present invention provides a vehicle arrival recognition system at a battery swap station, which utilizes the vehicle arrival recognition method at a battery swap station as described above, and specifically includes a vehicle recognition device, a high-precision video acquisition device, an industrial computer with GPU computing power, and a display device;

[0037] The vehicle identification device is installed outside the entrance channel of the battery swap station, and is used to identify the battery swap vehicle and read the identity information, and send the read information to the industrial computer. The identity information includes the license plate number and the battery pack specification information of the battery swap vehicle, namely, the length, width and height;

[0038] The high-precision video acquisition device is installed at the top position just above the intersection of the driving channel in the battery swap station and the battery exit port in the battery swap station, and is used to collect images of the battery pack to be inspected of the battery swap vehicle in the battery swap channel and send the images to the industrial computer;

[0039] The industrial computer is installed inside the working compartment of the battery swap station, and is used to receive the identity information of the battery swap vehicle sent by the vehicle identification device, and to retrieve the coordinate information of the battery swap limit boundary of the battery pack to be detected, and at the same time receive the image of the battery pack to be detected of the battery swap vehicle in the battery swap channel captured by the high-precision video acquisition device, and input it into the target detection model after preprocessing, so as to complete the identification of the battery pack to be detected and the acquisition of the positioning information of the battery pack to be detected, and further judge whether the current posture of the battery pack to be detected meets the battery swap demand based on the coordinate information of the corresponding battery swap limit boundary, and send the judgment result to the display device;

[0040] The display device is installed at one side of the battery swap channel exit in the battery swap station, and is used to display the video collected by the high-precision video acquisition device in real time, and is used to show the driver whether the position of the battery pack to be tested of the current battery swap vehicle is within the set battery swap limit boundary position box;

[0041] When the industrial computer receives the battery pack information of the battery swap vehicle to be detected sent by the vehicle identification device, the industrial computer will retrieve the coordinate information data of the corresponding battery swap limit boundary and send it to the display device for display;

[0042] When the industrial computer detects that the battery pack posture of the battery swapping vehicle meets the battery swapping posture, the battery swapping limit boundary position box displayed on the display device will turn green;

[0043] When the industrial computer detects that the battery pack posture of the battery swapping vehicle does not meet the battery swapping posture, the battery swapping limit boundary position box displayed on the display device will turn red and flash.

[0044] Furthermore, if the current position of the battery pack to be tested is not within the battery swap limit boundary position box or exceeds the battery swap limit boundary position box, the industrial computer controls the display device to prompt the driver to re-park the vehicle to a position that meets the battery swap requirements; if the current horizontal inclination angle of the battery pack to be tested is greater than the maximum value of the preset angle, that is, θ0>θ, then the industrial computer controls the display device to prompt the driver to re-park the battery swap vehicle and straighten it to meet the horizontal posture of the battery pack to be tested that meets the battery swap requirements.

[0045] Specifically, the industrial computer also includes training the target detection model and controlling the high-precision video acquisition device to capture images. When the industrial computer receives the information of the battery pack to be inspected of the battery swapping vehicle sent by the vehicle identification device, it controls the high-precision video acquisition device to start collecting images of the battery pack to be inspected of the battery swapping vehicle in the battery swapping channel.

[0046] On the other hand, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the steps of the method for identifying the presence of a vehicle at a battery swap station as described above.

[0047] In still another aspect, the present invention provides a battery swap station, comprising the vehicle arrival identification system for the battery swap station as described above.

[0048] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0049] The present invention calibrates the battery swap limit area of ​​the battery packs of the same specifications in the image based on the limit battery swap position of the battery packs of the same specifications in the battery swap station, determines the standard length and width of the battery packs of the same specifications in the image based on the standard battery swap position of the battery packs of the same specifications in the battery swap station; completes the establishment of the optimal model for battery pack detection and positioning based on the image data of the battery packs of the same specifications; obtains the image information of the battery pack to be detected, uses the optimal model for battery pack detection and positioning to complete the acquisition of the positioning information of the battery pack to be detected, and further completes the boundary detection and horizontal tilt angle detection of the battery pack to be detected according to the corresponding battery swap limit area of ​​the battery pack to be detected, the positioning information of the battery pack to be detected, and the standard length and width of the battery pack to be detected. According to the boundary detection and horizontal tilt angle detection results, it is judged whether the current posture of the battery pack to be detected meets the battery swap posture requirements. The present invention overcomes the problem of low battery swap efficiency caused by different battery packs in the current battery swap station, and uses visual sensors to complete the identification and positioning of the battery pack to be detected, which is lower in cost than the prior art using a multi-sensor joint detection method for parking guidance, and is convenient and fast. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings are incorporated in and constitute a part of this specification and, together with the description, serve to explain the principles of the present invention.

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0052] Figure 1 A flow chart of the method for identifying the vehicle at a battery swap station based on vision provided by the present invention;

[0053] Figure 2 The network structure diagram of the improved YOLOv11 model provided by the present invention;

[0054] Figure 3 A schematic diagram of boundary detection of battery packs of the same specification provided by the present invention;

[0055] Figure 4 A top view of a battery pack of the same specification provided by the present invention;

[0056] Figure 5 A schematic diagram of calculating the horizontal tilt angle of a battery pack to be tested provided by the present invention;

[0057] Figure 6 This is a structural block diagram of the vision-based vehicle arrival recognition system for battery swap stations provided by the present invention. DETAILED DESCRIPTION

[0058] Here, exemplary embodiments will be described in detail, and examples thereof are shown in the accompanying drawings. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are only examples consistent with some aspects of the present invention as detailed in the appended claims.

[0059] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0060] On the one hand, the present invention provides a method for identifying a vehicle at a battery swap station, and the specific steps are as follows:

[0061] Step 1: Capture a first image of a vehicle with a battery pack of the same specification in a battery swap station at the same shooting distance and shooting angle;

[0062] Step 2: Based on the first image, obtain the coordinate information of the battery replacement limit boundary corresponding to the battery pack of the same specification, as well as the length k1 and width k2 of the battery pack of the same specification;

[0063] Step 3: Establish a target detection model, divide the first image into a training set and a test set after preprocessing, train the target detection model and optimize the parameters to obtain the optimal target detection model;

[0064] Step 4: Using the optimal target detection model to identify the battery pack to be detected and obtain the positioning information of the battery pack to be detected, the image of the battery pack to be detected is captured at the same shooting distance and shooting angle; the positioning information of the battery pack to be detected is the coordinates of the upper left corner and lower right corner pixels of the vertical rectangular frame used to determine the positioning contour of the battery pack;

[0065] Step 5: Use the positioning information of the battery pack to be tested in step 4, the coordinate information of the battery swap limit boundary in step 2, and the length k1 and width k2 of the battery pack to be tested to complete the posture calculation of the battery pack to be tested, and determine whether the posture of the battery pack to be tested meets the battery swap posture requirements;

[0066] Step 6: According to the above judgment result, a corresponding prompt message is issued to the driver.

[0067] Specifically, under the same shooting distance and shooting angle and the same shooting camera, the standard length, width and height of the battery pack of the same specification corresponding to the standard battery swapping position in the battery swapping station are consistent with the length, width and height of the battery pack to be tested with a certain inclination angle. Furthermore, the battery pack of the same specification and the battery pack to be tested have the same size specifications, and are either installed on the same model of battery swapping vehicles, or installed on different models of battery swapping vehicles.

[0068] Specifically, in step 1, the first image includes image A of a battery pack of the same specification of a vehicle at the extreme position of battery replacement in the battery swap station at the same shooting distance and shooting angle, image B of a battery pack of the same specification of a vehicle entering and exiting the battery swap station continuously, and image C of a battery pack of the same specification of a vehicle at the standard parking position in the battery swap station.

[0069] Furthermore, the battery swap limit positions of vehicles with battery packs of the same specification in the battery swap station are: the front and rear lines perpendicular to the driving direction of vehicles with batteries of the same specification in the battery swap channel, specifically, the front and rear limit positions of the battery packs of the same specification corresponding to the front and rear limit positions of the vehicles parked in the driving direction of the battery swap channel;

[0070] Furthermore, the standard parking position of vehicles with battery packs of the same specifications in the battery swap station is as follows: the battery center point of the vehicle with battery packs of the same specifications is located directly below the shooting angle, and the posture of the battery swap vehicle has no angle with the horizontal driving channel in the battery swap station.

[0071] Specifically, the specific method of step 2 is as follows:

[0072] Step 2.1, determine the pixel position corresponding to the battery replacement limit position of the battery pack of the same specification in image A, and further determine the position of the boundary detection pixel point;

[0073] Step 2.2, determine the pixel position of the battery pack of the same specification in the C image, and further determine the length k1 and width k2 of the battery pack to be detected in the C image.

[0074] Specifically, in step 3, the preprocessing includes: image annotation and image preprocessing; the image annotation includes: marking the outline of the battery pack of the same specification in the B image with a rectangular frame; the image preprocessing includes: enhancing the image of the battery pack of the same specification through resizing, multi-environment simulation and color change technology.

[0075] Specifically, in step 3, the target detection model is an improved YOLOv11 model;

[0076] The improved YOLOv11 model includes an improvement mechanism for the feature fusion layer structure of the convolution, backbone network, and Neck parts in the YOLOv11 basic model, or an improvement mechanism for the feature fusion layer structure of the fusion backbone network and Neck.

[0077] Specifically, the specific method of step 5 is:

[0078] Step 5.1, based on the positioning information of the battery pack to be detected, determine whether the battery swapping vehicle is in a stationary state that satisfies the battery pack to be detected replacement posture calculation;

[0079] Step 5.2: Under the vehicle state that satisfies the battery pack replacement posture calculation, determine whether the battery pack to be detected exceeds the battery replacement limit boundary by using the positioning information of the battery pack to be detected and the coordinate information of the battery replacement limit boundary;

[0080] Step 5.3, using the positioning information of the battery pack to be detected and the length k1 and width k2 of the battery pack to be detected, calculate the horizontal tilt angle θ0 of the battery pack to be detected;

[0081] Step 5.4: Based on the judgment result of step 5.1 and the ratio of the horizontal tilt angle θ0 of the battery pack to be tested to the preset horizontal critical angle θ, determine whether the battery pack to be tested meets the battery replacement posture requirements. When the battery pack to be tested does not exceed the battery replacement limit boundary and the horizontal tilt angle θ0<θ, the posture of the battery pack to be tested meets the battery replacement posture requirements.

[0082] Specifically, in step 5.3, the specific method for calculating the horizontal tilt angle θ0 of the battery pack to be tested is:

[0083] Step 5.3.1, based on the positioning information of the battery pack to be detected, determine the length L3 and width L4 of the rectangular frame of the battery pack positioning outline;

[0084] Step 5.3.2, based on the intersection of the vertex of the battery pack to be detected with the long side and the wide side of the rectangular frame of the battery pack positioning contour, take the intersection point as the segmentation point, assume that a distance intersecting with the long side of the rectangular frame of the battery pack positioning contour is k6, and another distance is k5, and the corresponding distance intersecting with the wide side of the rectangular frame of the battery pack positioning contour is k3, and another distance is k4;

[0085] Step 5.3.3, based on the fact that the structure of the battery pack to be tested and the positioning contour of the battery pack are both rectangular, k6 is calculated according to Formula 1:

[0086]

[0087] The horizontal tilt angle θ0 of the battery pack to be tested is determined by equation 2:

[0088]

[0089] On the other hand, the present invention provides a vision-based vehicle identification system for a battery swap station, which utilizes the vehicle identification method for a battery swap station as described above, and specifically includes a vehicle identification device, a high-precision video acquisition device, an industrial computer with GPU computing power, and a display device;

[0090] The vehicle identification device is installed outside the entrance channel of the battery swap station, and is used to identify the battery swap vehicle and read the identity information, and send the read information to the industrial computer. The identity information includes the license plate number and the battery pack specification information of the battery swap vehicle, namely, the length, width and height;

[0091] The high-precision video acquisition device is installed at the top position just above the intersection of the driving channel in the battery swap station and the battery exit port in the battery swap station, and is used to collect images of the battery pack to be inspected of the battery swap vehicle in the battery swap channel and send the images to the industrial computer;

[0092] The industrial computer is installed inside the working compartment of the battery swap station, and is used to receive the identity information of the battery swapping vehicle sent by the vehicle identification device, and to retrieve the coordinate information of the battery swapping limit boundary of the battery pack to be detected, and at the same time receive the image of the battery pack to be detected of the battery swapping vehicle in the battery swapping channel captured by the high-precision video acquisition device. The image is input into the target detection model after preprocessing, and the identification of the battery pack to be detected and the acquisition of the positioning information of the battery pack to be detected are completed, and further based on the coordinate information of the corresponding battery swapping limit boundary, it is judged whether the current posture of the battery pack to be detected meets the battery swapping demand, and the judgment result is sent to the display device;

[0093] The display device is installed at one side of the battery swap channel exit in the battery swap station, and is used to display the video collected by the high-precision video acquisition device in real time, and is used to show the driver whether the position of the battery pack to be tested of the current battery swap vehicle is within the set battery swap limit boundary position box;

[0094] When the industrial computer receives the battery pack information to be detected sent by the vehicle identification device, the industrial computer will retrieve the coordinate information data corresponding to the battery replacement limit boundary and send it to the display device for display;

[0095] When the industrial computer detects that the battery pack posture of the battery swapping vehicle meets the battery swapping posture, the battery swapping limit boundary position box displayed on the display device will turn green;

[0096] When the industrial computer detects that the battery pack posture of the battery swapping vehicle does not meet the battery swapping posture, the battery swapping limit boundary position box displayed on the display device will turn red and flash.

[0097] Furthermore, if the current position of the battery pack to be tested is not within the battery swap limit boundary position box or exceeds the battery swap limit boundary position box, the industrial computer controls the display device to prompt the driver to re-park the vehicle to a position that meets the battery swap requirements; if the current horizontal inclination angle of the battery pack to be tested is greater than the maximum value of the preset angle, that is, θ0>θ, then the industrial computer controls the display device to prompt the driver to re-park the battery swap vehicle and straighten it to meet the horizontal posture of the battery pack to be tested that meets the battery swap requirements.

[0098] Specifically, the industrial computer also includes training the target detection model and controlling the high-precision video acquisition device to capture images. When the industrial computer receives the battery pack information to be detected of the battery swapping vehicle sent by the vehicle identification device, it controls the high-precision video acquisition device to start collecting images of the battery pack to be detected of the battery swapping vehicle in the battery swapping channel.

[0099] Example

[0100] This embodiment provides a method for identifying a vehicle at a battery swap station, and the specific steps are as follows:

[0101] Step 1: Use a high-precision video acquisition device installed at the intersection of the driving channel and the battery swapping bay in the battery swapping station, which is located at the vertical top of the center of the driving channel in the battery swapping station, to take the A image of the battery pack of the same specification of the vehicle at the battery swapping limit position in the battery swapping station, the B image of the battery pack of the same specification of the vehicle entering and exiting the battery swapping station, and the C image of the battery pack of the same specification of the vehicle at the standard parking position in the battery swapping station. Determine the pixel position corresponding to the battery swapping limit position of the battery pack of the same specification in the A image, and further determine the position of the boundary detection pixel point;

[0102] In image A, mark the contours of battery packs of the same specification and find the two pixel extreme points in the contour boundary that are consistent with the driving direction of the battery swap vehicle, such as Figure 3 As shown, Figure 3 It is a top view of the boundary detection of battery packs of the same specifications; in the figure, the battery swapping vehicle enters from the left and exits from the right in the battery swapping station; the outline of the battery pack of the same specifications photographed at this viewing angle is represented by a dotted rectangular frame, wherein the pixel positions of the left and right bottom corners of the battery pack of the same specifications in the image are the left bottom corner (x3, y3) and the right bottom corner (x4, y4), respectively, L1 and L2 are respectively the boundary lines determined by the pixel point positions corresponding to the leftmost and rightmost battery swapping limit positions of the battery pack of the same specifications photographed at this viewing angle in the driving direction of the battery swapping vehicle in the channel of the battery swapping station, wherein the pixel point coordinates of the bottom ends of L1 and L2 are (x1, y1) and (x2, y2), respectively;

[0103] Step 2, determine the pixel position of the battery pack with the same specification in the C image, and further determine the length k1 and width k2 of the battery pack with the same specification in the C image;

[0104] The standard battery swap position for battery packs of the same specifications in the battery swap station is the standard parking position for battery swap vehicles with battery packs of the same specifications in the battery swap station, specifically: the battery center point of the battery swap vehicle with battery packs of the same specifications is directly below the shooting angle of the high-precision video acquisition device, and the posture of the battery swap vehicle has no angle with the horizontal driving channel in the battery swap station; Figure 4As shown, the left figure shows a top view contour image of a battery pack of the same specification at a standard battery swapping position in a battery swapping station obtained from the viewing angle of a high-precision video acquisition device. The standard length k1 and width k2 of a battery pack to be tested at this viewing angle can be determined from the pixel positions of the four vertex corners of the contour image;

[0105] Step 3, establish a target detection model, perform image annotation and image preprocessing on the B image, first annotate the outline of the battery pack of the same specification in the B image with a rectangular frame, then enhance the image of the battery pack of the same specification through size adjustment, multi-environment simulation and color change technology, and then divide the image preprocessed into a training set and a test set, train the target detection model and optimize the parameters to obtain the optimal target detection model;

[0106] The target detection model is an improved YOLOv11 model, and the battery pack detection and positioning optimal model uses the improved YOLOv11 model; wherein, the improved YOLOv11 model mainly selects the improved mechanism for optimizing the network structures such as the convolution, backbone (feature extraction module in the backbone-YOLOv11 model), and Neck (feature integration module in the YOLOv11 model) parts of the YOLOv11 basic model, the feature fusion layer, the detection head, and the loss function, or integrates the improved mechanism of the network structure of the feature fusion layer in the backbone and the neck structure; preferably, the improved mechanism of the network structure of the feature fusion layer in the backbone and the neck is integrated;

[0107] The specific improvement strategies are as follows: Figure 2 As shown in the figure, the backbone network in the improved YOLOv11 model introduces the Universal InvertedBottleneck (UIB, inverted bottleneck search block) structure of MobileNetV4 to the backbone network in the YOLOv11 model, replacing the original attention mechanism module (C2PSA). Although the UIB structure is simple, it successfully unifies several popular architectures, including Inverted Bottleneck (IB, inverted bottleneck), ConvNext (convolutional network improved based on ResNet50 model), and Feed Forward Network (FFN, feedforward network), and introduces extra depth convolution (Extra Depthwise) technology, so that the UIB block allows flexible trade-offs between space / channel mixing, receptive field adjustment and computational utilization, so that the network can be adaptively adjusted according to the optimization goal;

[0108] Another improvement is to replace the neck structure of the YOLOv11 model with a BiFPN (bidirectional feature pyramid network) structure. Figure 2 As shown in the figure, BiFPN introduces a bidirectional connection, which allows information to propagate bidirectionally between different resolution levels, which helps to better fuse low-level and high-level features, promotes the contextual propagation of features, and improves the accuracy of object detection and segmentation. This improvement method can enable the feature map to be detected to fuse more feature information without increasing too much computational effort;

[0109] Before completing the training of the optimal model for battery pack detection and positioning, the image data of battery packs of the same specification based on the shooting angle are preprocessed. The image data of battery packs of the same specification can be selected to include battery packs of the same specification under different lighting conditions (such as at night and under strong light), different weather conditions (such as rain and snow) and battery swapping vehicle states (battery swapping vehicles are tilted, that is, battery packs of the same specification are tilted) and images of battery packs of different colors; the preprocessing includes image annotation and image preprocessing; the image annotation is specifically: rectangular frame annotation of the outline of battery packs of the same specification in the continuous images of battery swapping vehicles entering and leaving the battery swapping station; further the annotated images are adjusted by adjusting the image size, enhancing the contrast, etc., to improve the accuracy of the battery pack detection and positioning model algorithm; then the annotated image data is used to train the improved YOLOv11 model, and the optimal model is trained by setting parameters such as the learning rate and the number of iterations;

[0110] Step 4: Using the optimal target detection model to identify the battery pack to be detected and obtain the positioning information of the battery pack to be detected, the image of the battery pack to be detected is captured at the same shooting distance and shooting angle; the positioning information of the battery pack to be detected is the coordinates of the upper left corner and lower right corner pixels of the vertical rectangular frame used to determine the positioning contour of the battery pack;

[0111] The image of the battery pack to be inspected is resized and contrast enhanced, and then input into the optimal model for battery pack detection and positioning to complete the positioning of the battery pack to be inspected, wherein the positioning information is specifically to locate the outline of the battery pack to be inspected in the image of the battery pack to be inspected with a vertical rectangular frame and output the coordinates of the upper left corner and lower right corner pixels of the vertical rectangular frame; Figure 5 As shown in the right figure, the dotted frame is a vertical rectangular frame diagram of the battery pack positioning contour under the posture of the battery to be detected obtained through the battery pack detection and positioning optimal model;

[0112] Step 5: Use the positioning information of the battery pack to be tested in step 4, the coordinate information of the battery swap limit boundary in step 2, and the length k1 and width k2 of the battery pack to be tested to complete the posture calculation of the battery pack to be tested, and determine whether the posture of the battery pack to be tested meets the battery swap posture requirements. The specific method is as follows:

[0113] Step 5.1: According to the positioning information of the battery pack to be detected, determine whether the battery swapping vehicle is in a stationary state that meets the calculation requirements of the battery swapping position and attitude of the battery pack to be detected;

[0114] If it is satisfied, continue to execute the following steps; if not, continue to judge the state of the battery swapping vehicle until it reaches a stationary state that meets the calculation requirements of the battery swapping position and attitude of the battery pack to be detected;

[0115] Judge the state of the battery swapping vehicle as follows: Determine whether the current vehicle is in a stationary state or a driving state by judging the change of the pixel point coordinates at the lower right corner of the vertical rectangular frame of the battery pack positioning profile of the battery pack to be detected at an interval time. If the change amount of the pixel point coordinates corresponding to the lower right corner of the vertical rectangular frame of the battery pack to be detected is less than the set limit value Lc, the vehicle is in a stationary state; otherwise, the vehicle is in a driving state. It can also be judged by the vehicle speed information uploaded by the current vehicle;

[0116] Step 5.2: In the vehicle state that meets the calculation requirements of the battery swapping position and attitude of the battery pack to be detected, use the positioning information of the battery pack to be detected and the coordinate information of the battery swapping limit boundary to determine whether the battery pack to be detected exceeds the battery swapping limit boundary;

[0117] Specifically, as Figure 3 shown, in the vehicle state that meets the calculation requirements of the battery swapping position and attitude of the battery pack to be detected, for the vertical rectangular block diagram of the battery pack positioning profile corresponding to the battery pack to be detected, further obtain the pixel point coordinates at the lower left corner and the lower right corner of the vertical rectangular block diagram. Compare the x3 coordinate in the lower left corner pixel point and the x4 coordinate in the lower right corner pixel point with the x1 in the pixel point coordinates at the lowest end of the left limit position boundary line L1 of the same specification battery pack and the x2 in the pixel point coordinates at the lowest end of the right limit position boundary line L2 obtained in step S1. If x3>x1 and x4<x2, the battery pack to be detected is within the battery swapping limit boundary, and vice versa;

[0118] Step 5.3: Use the positioning information of the battery pack to be detected and the length k1 and width k2 of the battery pack to be detected to calculate the horizontal tilt angle θ0 of the battery pack to be detected. The specific method is as follows:

[0119] As Figure 5 shown, Figure 5In order to obtain a top view of the vertical rectangular frame diagram of the battery pack positioning outline of the battery pack to be detected at a certain tilt angle under this viewing angle, the dotted frame is the vertical rectangular frame of the battery pack positioning outline, and the solid frame is the actual boundary frame of the battery pack to be detected; under the same shooting distance and shooting angle and the same shooting camera, the standard length, width, and height of the battery pack of the same specification corresponding to the standard battery swapping position in the battery swapping station are consistent with the length, width, and height of the battery pack to be detected with a certain tilt angle, so that the image under this viewing angle can be obtained. Figure 5 As shown, the length k1 and width k2 of the battery pack to be detected with a certain tilt angle; wherein θ1 is the angle between the wide side of the battery pack to be detected and the vertical rectangular frame of the battery pack positioning contour, and θ0 is the angle between the long side of the right bottom corner of the battery pack to be detected and the vertical rectangular frame of the battery pack positioning contour. Since the battery pack to be detected is actually a rectangular frame, the battery pack positioning contour of the battery pack detection and positioning optimal model is also a vertical rectangular frame, so (θ1+θ0)=90°, and the pixel coordinates of the upper left corner and lower right corner of the vertical rectangular frame of the battery pack positioning contour of the battery pack to be detected with a certain tilt angle are known, the length L3 and width L4 of the vertical rectangular frame of the battery pack positioning contour can be obtained, as shown in Figure 5 As shown, where L3=k5+k6, L4=k3+k4; according to the Pythagorean theorem, it is shown by the following formula:

[0120]

[0121] If L3, L4, k1, and k2 are known, k3, k4, k5, and k6 can be obtained; and the horizontal tilt angle of the battery pack to be tested θ0 = acrsin(k6 / k2) can be further obtained;

[0122] Step 6: According to the boundary detection and tilt angle detection results, determine whether the current posture of the battery pack to be detected meets the battery swap posture requirements; wherein, the critical angle θ is set to 3°. If the battery pack to be detected is within the battery swap limit boundary as determined by step 5, and the horizontal tilt angle θ0 of the battery pack to be detected is <θ, then the battery swap posture requirements are met;

[0123] If the horizontal tilt angle θ0 of the battery pack to be tested is ≥θ or the battery pack to be tested is not within the battery swap limit boundary, the posture of the battery pack to be tested does not meet the battery swap posture requirements;

[0124] Step 6: According to the above judgment result, a corresponding prompt message is issued to the driver.

[0125] like Figure 6 As shown, this embodiment also provides a vehicle arrival recognition system for a battery swap station, and this system adopts the vehicle arrival recognition method for a battery swap station as described above;

[0126] Specifically include: vehicle identification device, high-precision video acquisition device, industrial computer with GPU computing power and display device;

[0127] The vehicle identification device is installed outside the entrance channel of the battery swap station, and is used to interact with the battery swap vehicle, to obtain the vehicle information of the battery swap vehicle, and to send the obtained information to the industrial computer; wherein the battery swap vehicle information mainly includes the vehicle VIN code, the specification information of the battery pack of the battery swap vehicle, that is, the length, width, height and other information. In this embodiment, the vehicle identification device specifically uses an RFID reader / writer. When a vehicle with an electronic tag approaches the RFID reader / writer, the RFID reader / writer reads the specific information in the electronic tag, and the electronic tag contains the vehicle information.

[0128] The high-precision video acquisition device is installed at the top position just above the intersection of the driving channel and the battery exit port in the battery swap station. It is used to capture the image of the battery pack to be inspected of the battery swap vehicle in the battery swap channel and send the image to the industrial computer.

[0129] At the same time, the high-precision video acquisition device is also used to calibrate the image acquisition work of the battery replacement limit area of ​​the battery pack to be tested in the image and to calibrate the standard length and width corresponding to the standard battery replacement position of the battery pack to be tested for image acquisition work.

[0130] The industrial computer is installed inside the working compartment of the battery swap station. The program in the industrial computer receives the battery pack information to be detected of the battery swap vehicle sent by the RFID reader, and retrieves the battery swap limit boundary position of the corresponding battery pack according to the acquired battery pack information to be detected. At the same time, the high-precision video acquisition device is controlled to acquire the video information in the battery swap channel in real time, and the image of the battery pack to be detected is captured according to the video information in the battery swap channel received from the high-precision video acquisition device. At the same time, the video information in the battery swap channel acquired by the high-precision video acquisition device in real time is synchronously displayed on the display device, and the retrieved battery swap limit boundary position of the corresponding battery pack is also displayed on the display device. Then, the image of the battery pack to be detected is preprocessed, and then input into the optimal model for battery pack detection and positioning to complete the acquisition of the positioning information of the battery pack to be detected, and further determine the vehicle state that satisfies the calculation of the posture of the battery pack to be detected through the positioning information of the battery pack to be detected. Under the condition of starting to calculate the posture of the battery pack to be detected, that is, the vehicle is stationary, the horizontal tilt angle θ0 of the battery pack to be detected is calculated through the positioning information of the battery pack to be detected and the standard length and width of the corresponding battery pack; and further retrieve the battery replacement limit boundary position of the corresponding battery pack, and the preset maximum horizontal tilt angle θ. Determine whether the battery pack to be detected is within the battery replacement limit boundary, whether the horizontal tilt angle θ0 of the battery pack to be detected is greater than the preset maximum horizontal tilt angle θ, and send the judgment result to the display device.

[0131] The display device is installed on the exit of the battery swap channel in the battery swap station, and displays the video collected by the high-precision video acquisition device in real time, so as to show the driver whether the position of the battery pack to be detected of the current battery swap vehicle is within the set battery swap limit boundary position frame. When the industrial computer receives the battery pack information to be detected of the battery swap vehicle sent by the vehicle identification device, it retrieves the corresponding battery swap limit boundary position frame data and sends it to the display device for display, and the initial battery swap limit boundary position frame is red. When the industrial computer detects that the posture of the battery pack to be detected meets the corresponding battery pack swap posture, the battery swap limit boundary position frame on the display device changes from red to green and displays TRUE; if the industrial computer detects that the posture of the battery pack to be detected does not meet the corresponding battery pack swap posture, the battery swap limit boundary position frame changes from a red border to a red dotted frame, flashes, and displays FALSE at the same time;

[0132] If the current position of the battery pack to be tested is not within or exceeds the battery swap limit boundary position box, the industrial computer controls the display device to prompt the driver to re-park the vehicle at a position that meets the battery swap requirements; if the current horizontal inclination angle of the battery pack to be tested is greater than the maximum value of the preset angle, that is, θ0>θ, the industrial computer controls the display device to prompt the driver to re-park the battery swap vehicle and straighten it to meet the horizontal posture of the battery pack to be tested that meets the battery swap requirements.

[0133] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention.

[0134] It should be understood that the present invention is not limited to what has been described above and that various modifications and changes may be made without departing from its scope. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for identifying a vehicle at a battery swap station, characterized in that: The specific steps are as follows: Step 1: Capture a first image of a vehicle with a battery pack of the same specification in a battery swap station at the same shooting distance and shooting angle; Step 2: Based on the first image, obtain the coordinate information of the battery replacement limit boundary corresponding to the battery pack of the same specification, as well as the length k1 and width k2 of the battery pack of the same specification; Step 3: Establish a target detection model, divide the first image into a training set and a test set after preprocessing, train the target detection model and optimize the parameters to obtain the optimal target detection model; Step 4: Using the optimal target detection model to identify the battery pack to be detected and obtain the positioning information of the battery pack to be detected, the images of the battery pack to be detected are captured at the same shooting distance and shooting angle; Step 5: Use the positioning information of the battery pack to be tested in step 4, the coordinate information of the battery swap limit boundary in step 2, and the length k1 and width k2 of the battery pack to be tested to complete the posture calculation of the battery pack to be tested, and determine whether the posture of the battery pack to be tested meets the battery swap posture requirements; Step 6: According to the above judgment result, a corresponding prompt message is issued to the driver.

2. The method for identifying vehicle arrival at a battery swap station according to claim 1, characterized in that: In step 1, the first image includes image A of a battery pack of the same specification of a vehicle at the limit position of battery replacement in a battery swap station at the same shooting distance and shooting angle, image B of a battery pack of the same specification of a vehicle entering and exiting the battery swap station continuously, and image C of a battery pack of the same specification of a vehicle at the standard parking position in the battery swap station.

3. The method for identifying vehicle arrival at a battery swap station according to claim 2, characterized in that: The specific method of step 2 is as follows: Step 2.1, determine the pixel position corresponding to the battery replacement limit position of the battery pack of the same specification in image A, and further determine the position of the boundary detection pixel point; Step 2.2, determine the pixel position of the battery pack of the same specification in the C image, and further determine the length k1 and width k2 of the battery pack of the same specification in the C image.

4. The method for identifying vehicle arrival at a battery swap station according to claim 2, characterized in that: In step 3, the preprocessing includes: image annotation and image preprocessing; The image annotation includes: annotating the outlines of battery packs of the same specification in the B image with rectangular frames; the image preprocessing includes: enhancing the images of battery packs of the same specification that have been annotated by resizing, multi-environment simulation, and color change technology.

5. The method for identifying vehicle arrival at a battery swap station according to claim 1, characterized in that: In step 3, the target detection model is an improved YOLOv11 model; The improved YOLOv11 model includes an improvement mechanism for the feature fusion layer structure of the convolution, backbone network, and Neck parts in the YOLOv11 basic model, or an improvement mechanism for the feature fusion layer structure of the fusion backbone network and Neck.

6. The method for identifying a vehicle at a battery swap station according to claim 1, characterized in that: Step 5: Step 5.1, based on the positioning information of the battery pack to be detected, determine whether the battery swapping vehicle is in a stationary state that satisfies the battery pack to be detected replacement posture calculation; Step 5.2: Under the vehicle state that satisfies the battery pack replacement posture calculation, determine whether the battery pack to be detected exceeds the battery replacement limit boundary by using the positioning information of the battery pack to be detected and the coordinate information of the battery replacement limit boundary; Step 5.3, using the positioning information of the battery pack to be detected and the length k1 and width k2 of the battery pack to be detected, calculate the horizontal tilt angle θ0 of the battery pack to be detected; Step 5.4: Based on the judgment result of step 5.1 and the ratio of the horizontal tilt angle θ0 of the battery pack to be tested to the preset horizontal critical angle θ, determine whether the battery pack to be tested meets the battery replacement posture requirements. When the battery pack to be tested does not exceed the battery replacement limit boundary and the horizontal tilt angle θ0<θ, the posture of the battery pack to be tested meets the battery replacement posture requirements.

7. The method for identifying a vehicle at a battery swap station according to claim 6, characterized in that: In step 5.3, the specific method for calculating the horizontal tilt angle θ0 of the battery pack to be tested is: Step 5.3.1, based on the positioning information of the battery pack to be detected, determine the length L3 and width L4 of the rectangular frame of the battery pack positioning outline; Step 5.3.2, based on the intersection of the vertex of the battery pack to be detected with the long side and the wide side of the rectangular frame of the battery pack positioning contour, take the intersection point as the segmentation point, assume that a distance intersecting with the long side of the rectangular frame of the battery pack positioning contour is k6, and another distance is k5, and the corresponding distance intersecting with the wide side of the rectangular frame of the battery pack positioning contour is k3, and another distance is k4; Step 5.3.3, based on the fact that the structure of the battery pack to be tested and the positioning contour of the battery pack are both rectangular, k6 is calculated according to Formula 1: The horizontal tilt angle θ0 of the battery pack to be tested is determined by equation 2:

8. A vehicle arrival recognition system for a battery swap station, characterized in that: The method for identifying a vehicle at a battery swap station as claimed in any one of claims 1 to 7 specifically comprises a vehicle identification device, a high-precision video acquisition device, an industrial computer with GPU computing power, and a display device; The vehicle identification device is installed outside the entrance channel of the battery swap station, and is used to identify the battery swap vehicle and read its identity information, and send the read information to the industrial computer. The identity information includes the license plate number and the battery pack specification information of the battery swap vehicle; The high-precision video acquisition device is installed at the top position just above the intersection of the driving channel in the battery swap station and the battery exit port in the battery swap station, and is used to collect images of the battery pack to be inspected of the battery swap vehicle in the battery swap channel and send the images to the industrial computer; The industrial computer is installed inside the working compartment of the battery swap station, and is used to receive the identity information of the battery swap vehicle sent by the vehicle identification device, and to retrieve the coordinate information of the battery swap limit boundary of the battery pack to be detected, and at the same time receive the image of the battery pack to be detected of the battery swap vehicle in the battery swap channel captured by the high-precision video acquisition device, and input it into the target detection model after preprocessing, so as to complete the identification of the battery pack to be detected and the acquisition of the positioning information of the battery pack to be detected, and further judge whether the current posture of the battery pack to be detected meets the battery swap demand based on the coordinate information of the corresponding battery swap limit boundary, and send the judgment result to the display device; The display device is installed at one side of the battery swap channel exit in the battery swap station, and is used to display the video collected by the high-precision video acquisition device in real time, and is used to show the driver whether the position of the battery pack to be tested of the current battery swap vehicle is within the set battery swap limit boundary position box; When the industrial computer receives the battery pack information of the battery swap vehicle to be detected sent by the vehicle identification device, the industrial computer will retrieve the coordinate information data of the corresponding battery swap limit boundary and send it to the display device for display; When the industrial computer detects that the battery pack posture of the battery swapping vehicle meets the battery swapping posture, the battery swapping limit boundary position box displayed on the display device changes from red to green; When the industrial computer detects that the battery pack posture of the battery swapping vehicle does not meet the battery swapping posture, the battery swapping limit boundary position box displayed on the display device will turn into a red dotted box and flash.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the steps of the method for identifying the vehicle at the battery swap station as described in any one of claims 1-7 are implemented.

10. A battery swap station, characterized in that: It includes the vehicle arrival identification system at the battery swap station as described in claim 8.

Images