Battery module angle self-adaptive adjusting mechanism and method based on dual-power lifting

Abstract

The invention provides a battery module angle self-adaptive adjusting mechanism and method based on dual-power lifting, and the mechanism comprises a power execution unit which is used for driving the vertical movement of lifting platforms at two sides; the lifting platform is connected with the two power execution units; according to the flexible link structure, when the lifting platforms on the two sides have height difference, the cargo carrying platform rotates through a hinge shaft and is matched with rolling of a cam groove and a supporting roller to achieve inclination adjustment; the visual camera is installed on the lower side of the center of the cargo carrying platform and used for measuring angle information of the cluster entering storage location in real time; and the control module is used for receiving angle information measured by the visual camera, calculating a lifting height difference required by the power execution units on the two sides, controlling the power execution units to realize height adjustment, and stopping adjustment until the angle measured by the visual camera is 0 degree. The angle of the battery module can be automatically and accurately adjusted in the battery cluster entering process, so that the battery module can be accurately butted with the storage location of the energy storage cabinet, and the stability and the reliability of battery module installation are ensured.

Description

Technical Field

[0001] This invention relates to the field of battery module installation technology, specifically to a battery module angle adaptive adjustment mechanism and method based on dual-power lifting, which is suitable for precise adjustment of the battery module angle during battery clustering. Background Technology

[0002] Traditional angle adjustment mechanisms have significant drawbacks in the battery module clustering process. Firstly, they require an additional adjustment platform and a separate power unit, complicating the overall structure and increasing manufacturing costs. Secondly, manual adjustment is inefficient, taking at least 3 minutes per adjustment, severely impacting the battery module clustering cycle time and reducing the overall production efficiency.

[0003] Furthermore, existing adjustment methods are insufficient in terms of precision, resulting in significant angle deviations that fail to meet the accuracy requirements for battery module installation angles as stipulated in relevant standards (such as GB / T 36276-2018). Therefore, developing a battery module angle adjustment solution that is simple in structure, highly efficient, and highly accurate is of significant practical importance. Summary of the Invention

[0004] To address the various shortcomings of existing technologies, this invention provides a battery module angle adaptive adjustment mechanism and method based on dual-power lifting, which is mainly used to automatically and accurately adjust the angle of the battery module during the battery clustering process, so that it can accurately align with the energy storage cabinet location and ensure the stability and reliability of battery module installation.

[0005] The present invention achieves the above objectives through the following technical solutions: A battery module angle adaptive adjustment mechanism based on dual-power lifting, comprising: The power actuators are arranged symmetrically on the left and right sides. Each power actuator includes at least one servo motor and a sprocket and chain transmission mechanism to drive the vertical movement of the lifting platforms on both sides. A lifting platform connecting two power actuators, wherein the left lifting platform is connected to the left power actuator and the right lifting platform is connected to the right power actuator; The flexible link structure includes a hinge shaft, a cam groove structure, and support rollers. When there is a height difference between the two lifting platforms, the cargo platform can achieve tilt adjustment by rotating the hinge shaft and cooperating with the cam groove structure and the rolling of the support rollers. A vision camera installed on the lower side of the center of the cargo platform is used to measure the angle information of the inbound storage location in real time; The main control module receives the angle information measured by the vision camera, calculates the required lifting height difference between the two power actuators, and controls the power actuators to adjust the height. The adjustment stops when the angle measured by the vision camera reaches 0° through a closed-loop control process.

[0006] The battery module angle adaptive adjustment mechanism based on dual-power lifting provided by the present invention further includes a lifting guide rail assembly, which includes: The lifting linear guide rails are respectively set on both sides of the lifting platform. The lifting linear guide rails are fixedly connected to the frame and serve as the vertical guide structure of the lifting platform to prevent the platform from deviating or tilting during movement and to ensure that the lifting platform always remains vertical. Each side of the lifting linear guide rail is connected to the corresponding lifting platform through two front and rear lifting linear guide rails. The four guide rail sliders form a rectangular layout to ensure the stable movement of the lifting platform in the vertical direction.

[0007] According to the present invention, a battery module angle adaptive adjustment mechanism based on dual-power lifting is provided, wherein the flexible link structure is specifically as follows: The left lifting platform is connected to the hinge shaft on the left side of the cargo platform via a bearing, forming a pivot point for rotation. The surface of the right lifting platform is provided with a cam groove structure, and the support roller on the right side of the cargo platform is embedded in the cam groove structure; When there is a height difference between the two lifting platforms, the cargo platform rotates around the hinge axis, and the support rollers roll along the trajectory of the cam groove structure, so as to achieve a smooth transition of the cargo platform from a horizontal state to an inclined state.

[0008] According to the present invention, a battery module angle adaptive adjustment mechanism based on dual-power lifting provides that when the height of the right lifting platform is higher than that of the left lifting platform, the cargo platform tilts through the following structure: With the mounting bearing of the left lifting platform as the center of rotation, the support roller on the right side of the cargo platform moves upward in an arc along the cam groove structure trajectory of the right lifting platform. The curved surface design of the cam groove structure enables the support roller to generate horizontal displacement synchronously during the rolling process, forming a compound motion mechanism of rolling and sliding, so that the right edge of the cargo platform tilts upward while smoothly shifting to the left. During the tilting process, the angle θ between the cargo platform and the horizontal plane is monitored in real time by a vision camera. When θ=0°, the main control module stops the adjustment to ensure that the cargo platform reaches a horizontal state.

[0009] A method for adaptive adjustment of battery module angle based on dual-power lifting, the method employing the aforementioned adaptive adjustment mechanism for battery module angle based on dual-power lifting, includes the following steps: The AGV carrying the battery module moves to the designated docking position of the energy storage cabinet to complete the spatial position calibration; After receiving the target storage location information, the AGV calculates the required lifting height of the two lifting platforms. The main control system drives the servo motors on both sides to operate synchronously, and controls the two lifting platforms to rise vertically to the calculated height through the sprocket and chain transmission mechanism. The vision camera installed on the lower side of the center of the cargo platform makes the first measurement of the angle θ between the storage location plane and the horizontal plane. The θ value includes positive and negative direction indicators. The vision camera feeds back the θ value to the main control system in real time. The main control system calculates the required height difference ΔH between the two platforms based on the θ value and the distance between the two power actuators. The ΔH value includes positive and negative direction indicators. The main control system controls the operation of the right-side servo motor based on the ΔH value: when ΔH is positive, it drives the right-side lifting platform to rise; when ΔH is negative, it drives the right-side lifting platform to fall. During the adjustment process, the cargo platform tilts through the flexible connection between the hinge shaft and the cam groove structure; the support rollers roll along the trajectory of the cam groove structure and generate horizontal displacement, so that the cargo platform smoothly transitions to the target angle; The vision camera performs a second measurement on the adjusted storage location angle to obtain a new θ value; If the new angle θ value is not equal to 0°, then ΔH is recalculated and adjustment is performed, forming a closed-loop control process of measurement-calculation-adjustment, until the adjustment is terminated when θ = 0°; When the visual camera detects that the angle deviation of the storage location is less than the preset threshold, the main control system determines that the adjustment is complete and issues a command to allow the battery module to enter the cluster, thus completing the entire adjustment process.

[0010] According to the present invention, a battery module angle adaptive adjustment method based on dual-power lifting is provided. The main control system calculates the required height difference ΔH between the two lifting platforms using the following formula: ΔH=(α×L×tanθ)+β ΔH is the height difference between the two lifting platforms; L is the distance between the power execution units on the left and right sides; θ is the angle between the storage location plane and the horizontal plane measured by the vision camera for the first time; α is the environmental correction factor; when calculating the height difference, a dynamic compensation term β is added. The main control system obtains the lifting speed v and load weight m in real time through the speed sensor and weight sensor installed on the lifting platform, and calculates the dynamic compensation term β in combination with the preset mechanical characteristic parameters. Based on the calculated ΔH value, the main control system generates adjustment commands for the right-side servo motor through a closed-loop control algorithm, thereby realizing the differential height adjustment of the two lifting platforms and calculating the required height difference between the two platforms.

[0011] According to the present invention, a battery module angle adaptive adjustment method based on dual-power lifting is provided. After the relevant movement of the storage location is completed, the main control system sends a measurement command to the vision camera, triggering the vision camera to perform image acquisition and measurement operations on the storage location. The visual camera transmits the acquired storage location image information to the main control system. The main control system uses a preset visual algorithm to analyze and process the image, and calculates the angle θ between the current storage location plane and the horizontal plane. Based on the calculated included angle θ and the geometric dimensions of the storage location, the main control system calculates the height difference ΔH between the left and right lifting platforms; further, based on the height difference ΔH and the transmission parameters of the servo motor of the right lifting platform, it calculates the number of rotations N required by the servo motor of the right lifting platform. The main control system sends the calculated number of revolutions N to the servo system. The servo system controls the right lifting platform motor to rotate according to the required number of revolutions based on the command, and feeds back the motor rotation information in real time through the encoder to realize servo closed-loop control and ensure that the right lifting platform is accurately adjusted to the target height. After one adjustment is completed, the main control system notifies the vision camera to take measurements again, repeating the angle calculation, height difference and number of circles calculation, and servo adjustment steps until the storage position angle measured by the vision camera is 0°, realizing closed-loop adjustment of the entire system and ensuring that the storage position reaches a horizontal state.

[0012] According to the present invention, a battery module angle adaptive adjustment method based on dual-power lifting is provided. After the main control system acquires the original image of the storage location captured by the vision camera, the image preprocessing operation is first performed. The median filtering algorithm is used to remove noise interference in the image. The median filter is used to sort the neighboring pixel values ​​of each pixel in the image and take the median value as the new pixel value of that point. Then, histogram equalization technology is used to enhance the contrast of the denoised image. On the preprocessed image, the main control system uses an accelerated segmented feature transformation algorithm to extract feature points of the storage location plane; at the same time, it generates a unique descriptor for each feature point, which at least contains gradient information of the surrounding area of ​​the feature point; then, it matches the extracted feature points with feature points of the pre-stored standard horizontal storage location plane template image, and uses the nearest neighbor ratio method for matching and filtering, that is, it calculates the Euclidean distance between feature point descriptors and sets a distance threshold, and only retains matching point pairs with a distance less than the threshold. For a successfully matched pair of feature points, the main control system uses a random sampling consensus algorithm to estimate the geometric transformation model. Based on the optimal geometric transformation model, the rotation angle of the current storage location plane relative to the standard horizontal storage location plane is calculated, that is, the angle θ between the current storage location plane and the horizontal plane.

[0013] According to the present invention, a battery module angle adaptive adjustment method based on dual-power lifting is provided, which uses an accelerated segmented feature transformation algorithm to extract feature points of the storage location plane, including the following steps: The main control system first converts the original image of the storage location plane captured by the vision camera into a grayscale image, and then constructs an integral image by traversing each pixel of the grayscale image; For any point in the integral image ( x , y ), whose value is from (0,0) to ( x , y The sum of the gray values ​​of all pixels within the rectangular area formed by ) is, i.e. ,in I ( i , j ) represents the original image at point ( i , j The grayscale value at () Based on the constructed integral image, the main control system constructs the Hessian matrix in different scale spaces of the storage location planar image, for each pixel in the image. X =( x , y ), in scale σ Hessian matrix under H ( X , σ ) is defined as:

[0014] in, L xx ( X , σ ) is the second-order partial derivative of Gauss. With images I Convolution at point X The value at that location.

[0015] According to the battery module angle adaptive adjustment method based on dual-power lifting provided by the present invention, the determinant of the Hessian matrix is ​​calculated and expressed as the following formula:

[0016] By comparing the determinant value of each pixel at different scales with the determinant values ​​of its surrounding neighboring pixels, if the determinant value of that pixel is a local maximum, it is marked as a candidate feature point. The non-maximum suppression method is used to filter in scale space and image space, retaining the feature points with the strongest response, thereby detecting feature points in the storage location planar image; Centered on the detected feature point, within a radius of 6s Within a circular neighborhood, calculate the position of all pixels. x and y The Haar wavelet response of the direction is obtained by sliding and accumulating these response values ​​in a 60° fan-shaped window centered on the feature point. The direction of the fan-shaped window with the largest accumulated value is the main direction of the feature point. A 20s×20s square region is selected around the feature point and rotated to the main direction of the feature point. Then, the region is divided into 4×4 sub-regions. In each sub-region, the cumulative values ​​of the Haar wavelet responses in the x and y directions and the cumulative values ​​of the absolute values ​​of the responses are calculated to form a 4-dimensional vector. Combine the vectors of all sub-regions to form a 4×4×4 = 64-dimensional vector, which is the descriptor of the feature point.

[0017] Therefore, compared with the prior art, the battery module angle adaptive adjustment mechanism and method based on dual-power lifting proposed in this invention significantly improves the accuracy, efficiency and safety of warehouse management, and has the following beneficial effects: 1. Traditional angle adjustment mechanisms require an adjustment platform and a separate power unit, resulting in a complex structure. This invention utilizes dual power sources on both sides of the lifting mechanism to directly generate the tilt angle through the height difference, eliminating the need for an adjustment platform and a separate power unit. This significantly simplifies the mechanical structure, reducing the number of parts, lowering manufacturing and assembly difficulty, and making the entire adjustment mechanism more compact and space-saving. Due to the simplified structure, the procurement, processing, and assembly costs of parts are significantly reduced. Compared to traditional technologies, this invention reduces costs, which can significantly lower production costs and improve the product's market competitiveness for large-scale production and application.

[0018] 2. In traditional methods, manual intervention for adjustment is inefficient, with each adjustment taking ≥3 minutes, affecting the battery module clustering cycle. This invention, however, adjusts the angle simultaneously with the lifting, eliminating the need for additional adjustment time. For example, after the AGV moves the battery to the docking position of the energy storage cabinet, the AGV system calculates the required height, and the two lifting platforms begin to rise. Simultaneously, based on the angle information measured by the vision camera, the height difference between the two lifting platforms is adjusted, completing the angle adjustment. This significantly shortens the adjustment time and improves the efficiency of battery module clustering.

[0019] 3. The core workflow of this invention is a continuously looping process. The vision camera measures the angle of the storage location in real time. If the angle does not meet the requirements, the main control system will immediately calculate the new height difference and adjust the lifting platform until the angle measured by the camera is 0. This rapid loop adjustment mechanism can ensure that the angle of the battery module is quickly and accurately adjusted to the appropriate position, further improving the adjustment efficiency.

[0020] 4. This invention uses a visual camera to measure the angle of the cluster entry position in real time, enabling precise acquisition of angle information. Based on this angle information and combined with the height difference generation mechanism, the main control system accurately calculates the height difference between the two lifting platforms, thereby achieving high-precision angle adjustment with an angle deviation ≤0.5°, fully meeting the requirements of GB / T 36276-2018 standard.

[0021] 5. The dual-power source independent control and flexible structure of this invention enable the adjustment mechanism to adapt to the angle requirements of different storage locations. Regardless of whether the storage location angle is positive or negative, or how the angle changes, the tilt of the loading platform can be achieved by adjusting the height difference between the two lifting platforms, thus accurately aligning with storage locations at different angles. The power actuator can be selected from various forms such as chain, electric cylinder, hydraulic cylinder, or motor screw, offering strong compatibility. In practical applications, a suitable power actuator can be selected based on different working environments, load requirements, and operating costs, improving the flexibility and applicability of the adjustment mechanism.

[0022] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of an embodiment of a battery module angle adaptive adjustment mechanism based on dual-power lifting according to the present invention.

[0024] Figure 2 This is a schematic diagram of the flexible link structure in an embodiment of a battery module angle adaptive adjustment mechanism based on dual-power lifting according to the present invention.

[0025] Figure 3 This is a flowchart of an embodiment of a battery module angle adaptive adjustment method based on dual-power lifting according to the present invention.

[0026] Figure 4 This is a schematic diagram illustrating the principle of calculating the lifting value of the control platform for the height difference in an embodiment of a battery module angle adaptive adjustment method based on dual-power lifting according to the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0028] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0029] An embodiment of a battery module angle adaptive adjustment mechanism based on dual-power lifting. See Figure 1 and Figure 2 This embodiment provides a battery module angle adaptive adjustment mechanism based on dual-power lifting, including: The power execution units are arranged symmetrically on the left and right sides. Each power execution unit includes at least one servo motor and a sprocket and chain transmission mechanism 5, which are used to drive the vertical movement of the lifting platforms on both sides. A lifting platform connecting two power actuators, wherein the left lifting platform 3 is connected to the left power actuator 1, and the right lifting platform 4 is connected to the right power actuator 2; The flexible link structure includes a hinge shaft 11, a cam groove structure 12, a support roller 13, and a connecting seat 14. When there is a height difference between the two lifting platforms, the cargo platform 6 can achieve tilt adjustment by rotating the hinge shaft 11 and cooperating with the rolling of the cam groove structure 12 and the support roller 13. A vision camera 7 installed on the lower side of the center of the cargo platform 6 is used to measure the angle information of the inbound storage location in real time; The main control module receives the angle information measured by the vision camera 7, calculates the required lifting height difference between the two power actuators, and controls the power actuators to adjust the height. The adjustment stops when the angle measured by the vision camera 7 is 0° through a closed-loop control process.

[0030] In this embodiment, a lifting guide rail assembly is also included, which includes: The lifting linear guide rails 8 are respectively set on both sides of the lifting platform. The lifting linear guide rails 8 are fixedly connected to the frame and serve as the vertical guide structure of the lifting platform to prevent the platform from deviating or tilting during the movement and to ensure that the lifting platform always remains vertical. Each side of the lifting linear guide rail is connected to the corresponding lifting platform through two front and rear lifting linear guide rails 8. The four guide rail sliders form a rectangular layout to ensure the stable movement of the lifting platform in the vertical direction.

[0031] In this embodiment, the flexible link structure is specifically as follows: The left lifting platform is connected to the hinge shaft 11 on the left side of the cargo platform 6 by installing bearings, forming a rotation fulcrum; A cam groove structure 12 is provided on the surface of the right lifting platform, and the support roller 13 on the right side of the cargo platform 6 is embedded in the cam groove structure 12; When there is a height difference between the two lifting platforms, the cargo platform 6 rotates around the hinge shaft 11, and the support roller 13 rolls along the trajectory of the cam groove structure 12, so as to achieve a smooth transition of the cargo platform 6 from a horizontal state to an inclined state.

[0032] When the height of the right-side lifting platform 4 is higher than that of the left-side lifting platform 3, the cargo platform 6 tilts through the following structure: With the mounting bearing of the left lifting platform as the center of rotation, the support roller 13 on the right side of the cargo platform 6 moves in an arc-shaped upward motion along the trajectory of the cam groove structure 12 of the right lifting platform. The curved surface design of the cam groove structure 12 enables the support roller 13 to generate horizontal displacement synchronously during the rolling process, forming a composite motion mechanism of rolling and sliding, so as to achieve the upward tilting of the right edge of the cargo platform 6 while smoothly shifting to the left. During the tilting process, the angle θ between the cargo platform 6 and the horizontal plane is monitored in real time by the vision camera 7. When θ=0°, the main control module stops the adjustment to ensure that the cargo platform reaches a horizontal state.

[0033] Furthermore, the power execution unit includes, but is not limited to, any one or more combinations of chain drive mechanism, electric cylinder drive mechanism, hydraulic cylinder push mechanism or motor screw drive mechanism.

[0034] When a chain drive mechanism is used, the two motors on both sides drive the left and right chains respectively through the sprocket structure to realize the independent lifting and lowering movement of the left and right lifting platforms; when an electric cylinder, hydraulic cylinder or motor screw is used, it is directly or indirectly connected to the lifting platform through the corresponding drive element to realize the vertical lifting and lowering of the platform.

[0035] Furthermore, the flexible link structure is a movable flexible platform based on a combination of fulcrums and cams, specifically including: The left lifting platform 3 and the cargo platform 6 are connected by a bearing fulcrum to form a fixed rotation point; the right lifting platform 4 is provided with a cam groove structure 12, and the right support roller 13 of the cargo platform 6 is embedded in the cam groove; when there is a height difference between the two lifting platforms, the cargo platform 6 moves in an arc along the cam groove with the left bearing fulcrum as the center, thereby realizing the tilt adjustment of the cargo platform 6.

[0036] The power actuator and the flexible link structure work together. The power actuator independently controls the lifting height of the left and right lifting platforms, generating a height difference ΔH. The flexible link structure converts this height difference into a tilt angle θ of the cargo platform 6, satisfying the geometric relationship of ΔH. Through a rolling contact design between the fulcrum and the cam, the flexible link structure achieves low-friction, high-stability movement of the cargo platform 6 during tilting, avoiding mechanical jamming or excessive wear. Combining the angle information measured in real time by the vision camera 7, the main control system dynamically calculates the required height difference ΔH and adjusts the position of the lifting platforms through the power actuator. The flexible link structure responds to changes in the height difference, adjusting the tilt angle of the cargo platform in real time, forming a closed-loop control process of "measurement-calculation-execution-re-inspection".

[0037] An embodiment of a battery module angle adaptive adjustment method based on dual-power lifting like Figure 3 As shown, this embodiment provides a battery module angle adaptive adjustment method based on dual-power lifting. This method employs the aforementioned battery module angle adaptive adjustment mechanism based on dual-power lifting and includes the following steps: Step S1: The clustered AGV carries the battery module to the designated docking position of the energy storage cabinet to complete the spatial position calibration. In step S2, after the AGV receives the target storage location information, it calculates the required lifting height of the two lifting platforms. The main control system drives the servo motors on both sides to operate synchronously, and controls the two lifting platforms to rise vertically to the calculated height through the sprocket and chain transmission mechanism 5. In step S3, the vision camera 7 installed on the lower side of the center of the cargo platform 6 performs the first measurement of the angle θ between the storage location plane and the horizontal plane. The θ value includes positive and negative direction indicators. The vision camera 7 feeds back the θ value to the main control system in real time. Step S4: The main control system calculates the required height difference ΔH between the two platforms based on the θ value and the distance between the two power actuators. The ΔH value includes positive and negative direction indicators. Step S5: The main control system controls the operation of the right-side servo motor according to the ΔH value: when ΔH is positive, it drives the right-side lifting platform 4 to rise; when ΔH is negative, it drives the right-side lifting platform 4 to fall. During the adjustment process, the cargo platform 6 tilts through the flexible connection between the hinge shaft 11 and the cam groove structure 12; the support roller 13 rolls along the trajectory of the cam groove structure 12 and generates horizontal displacement, so that the cargo platform 6 smoothly transitions to the target angle; Step S6: The vision camera 7 performs a second measurement on the adjusted storage location angle to obtain a new θ value; If the new angle θ value is not equal to 0°, then ΔH is recalculated and adjustment is performed, forming a closed-loop control process of measurement-calculation-adjustment, until the adjustment is terminated when θ = 0°; In step S7, when the visual camera 7 detects that the deviation of the storage position angle is less than the preset threshold, the main control system determines that the adjustment is complete and issues a command to allow the battery module to enter the cluster, thus completing the entire adjustment process.

[0038] In this embodiment, the main control system calculates the required height difference ΔH between the two lifting platforms using the following formula: ΔH=(α×L×tanθ)+β ΔH represents the height difference between the two lifting platforms. A positive value indicates that the right platform needs to be higher than the left platform, while a negative value indicates that the right platform needs to be lower than the left platform. L represents the distance between the power execution units on the left and right sides, which is a fixed design parameter. θ represents the angle between the storage location plane and the horizontal plane measured for the first time by the vision camera 7, with a value range of [-90°, 90°]. Positive and negative values ​​represent the clockwise or counterclockwise tilt direction of the storage location plane relative to the horizontal plane, respectively. α represents the environmental correction factor. By introducing the environmental correction factor α, the influence of factors such as temperature, humidity, and air disturbance on the measurement accuracy of the vision camera and the stability of the storage location structure is comprehensively considered. Specifically, during the system installation and commissioning phase, multiple storage location angle measurements and actual height difference measurements are conducted under different environmental conditions to establish a database of environmental parameters and measurement errors. During actual operation, the main control system acquires the current environmental parameters such as temperature and humidity in real time, and calculates the corresponding environmental correction factor α by querying the database or using a preset mathematical model. A dynamic compensation term β is added when calculating the height difference. The dynamic compensation term β is related to the lifting speed of the storage location, the load weight, and the mechanical characteristics of the lifting platform. The main control system obtains the lifting speed v and load weight m in real time through speed sensors and weight sensors installed on the lifting platform. Combined with the preset mechanical characteristic parameters, it calculates the dynamic compensation term β using a preset algorithm. Based on the calculated ΔH value, the main control system generates adjustment commands for the right-side servo motor through a closed-loop control algorithm, thereby realizing the differential height adjustment of the two lifting platforms and calculating the required height difference between the two platforms.

[0039] like Figure 4 As shown, after the relevant movements of the storage location are completed, the main control system sends a measurement command to the vision camera 7, triggering the vision camera 7 to perform image acquisition and measurement operations on the storage location. The visual camera 7 transmits the acquired storage location image information to the main control system. The main control system uses a preset visual algorithm to analyze and process the image, and calculates the angle θ between the current storage location plane and the horizontal plane. Based on the calculated included angle θ and the geometric dimensions of the storage location, the main control system calculates the height difference ΔH between the left and right lifting platforms 4; further, based on the height difference ΔH and the transmission parameters of the servo motor of the right lifting platform 4, it calculates the number of rotations N required by the servo motor of the right lifting platform 4. The main control system sends the calculated number of revolutions N to the servo system. The servo system controls the servo motor of the right lifting platform 4 to rotate according to the required number of revolutions, and feeds back the motor rotation information in real time through the encoder to realize servo closed-loop control and ensure that the right lifting platform 4 is accurately adjusted to the target height. After one adjustment is completed, the main control system notifies the vision camera 7 to take measurements again, repeating the angle calculation, height difference and number of circles calculation, and servo adjustment steps until the storage position angle measured by the vision camera 7 is 0°, realizing closed-loop adjustment of the entire system and ensuring that the storage position reaches a horizontal state.

[0040] After the main control system acquires the original image of the storage location captured by the vision camera 7, it first performs image preprocessing. The median filtering algorithm is used to remove noise interference from the image. The median filter sorts the neighboring pixel values ​​of each pixel in the image and takes the median value as the new pixel value of that point, effectively eliminating impulse noise and salt-and-pepper noise. Then, histogram equalization technology is used to enhance the contrast of the denoised image. By redistributing the gray values ​​of the image pixels, the gray distribution of the image is made more uniform, thereby highlighting the feature information of the storage location plane. On the preprocessed image, the main control system uses the Accelerated Segmented Feature Transform (SURF) algorithm to extract feature points from the storage location plane. The SURF algorithm detects feature points by constructing a Hessian matrix and calculating its determinant, utilizing integral images to accelerate the calculation process and improve the efficiency of feature point detection. Simultaneously, it generates a unique descriptor for each feature point, containing gradient information from the surrounding region. Then, the extracted feature points are matched with feature points from a pre-stored standard horizontal storage location plane template image. The nearest neighbor ratio method is used for matching and filtering; that is, the Euclidean distance between feature point descriptors is calculated, and a distance threshold is set. Only matching point pairs with a distance less than the threshold are retained, ensuring the accuracy and reliability of the matching.

[0041] For successfully matched feature point pairs, the main control system uses the Random Sample Consensus (RANSAC) algorithm to estimate the geometric transformation model. The RANSAC algorithm fits a geometric transformation model, such as an affine transformation model, by randomly selecting a certain number of matching point pairs. Then, it calculates the error of other matching point pairs based on this model. Through multiple iterations, it selects the model with the most inliers (matching point pairs with errors less than a set threshold) as the optimal model. Based on the optimal geometric transformation model, especially its rotation parameters, the main control system accurately calculates the rotation angle of the current storage location plane relative to the standard horizontal storage location plane, i.e., the angle θ between the current storage location plane and the horizontal plane. This calculation method fully considers interference factors such as noise and occlusion in the image, and through the organic combination of feature extraction, matching, and model estimation, it achieves high-precision calculation of the angle between the storage location plane and the horizontal plane.

[0042] In this embodiment, the accelerated segmented feature transform (SURF) algorithm is used to extract feature points on the storage location plane, including the following steps: The main control system first converts the original image of the storage location plane captured by vision camera 7 into a grayscale image to reduce computation. Then, it constructs an integral image by traversing each pixel of the grayscale image. For any point in the integral image ( x , y ), whose value is from (0,0) to ( x , y The sum of the gray values ​​of all pixels within the rectangular area formed by ) is, i.e. ,in I ( i , j ) represents the original image at point ( i , j The grayscale value at () is used. By utilizing the integral image, the pixel sum of any rectangular region in subsequent calculations can be quickly obtained through four table lookup operations, greatly accelerating the subsequent calculation of the Hessian matrix.

[0043] Based on the constructed integral image, the main control system constructs the Hessian matrix in different scale spaces of the storage location planar image. For each pixel in the image... X =( x , y ), in scale σ Hessian matrix under H ( X , σ ) is defined as:

[0044] in, L xx ( X , σ ) is the second-order partial derivative of Gauss. With images I Convolution at point X The value at that location, L xy ( X , σ )and L yy ( X , σ Similar to the definition. To improve computational efficiency, an integral image and a pre-computed box filter are used to approximate the convolution operation between the Gaussian second-order partial derivative and the image.

[0045] The determinant of the Hessian matrix is ​​calculated using the following formula:

[0046] By comparing the determinant value of each pixel at different scales with the determinant values ​​of its neighboring pixels, if the determinant value of that pixel is a local maximum, it is marked as a candidate feature point. Then, a non-maximum suppression method is used to filter in both scale space and image space, retaining the feature points with the strongest response, thereby accurately detecting feature points in the storage location planar image.

[0047] Centered on the detected feature point, within a radius of 6 s ( s Within a circular neighborhood (at the scale of the feature point), calculate the value of all pixels. x and y The Haar wavelet response of the direction is obtained by sliding and accumulating these response values ​​in a 60° fan-shaped window centered on the feature point. The direction of the fan with the largest accumulated value is the main direction of the feature point.

[0048] A 20s×20s square region is selected around the feature point and rotated to align with the principal direction of the feature point to ensure rotation invariance of the descriptor. This region is then divided into 4×4 sub-regions. Within each sub-region, the cumulative values ​​of the Haar wavelet responses in the x and y directions, as well as the cumulative values ​​of their absolute values, are calculated, forming a 4-dimensional vector. The vectors from all sub-regions are combined to form a 4×4×4 = 64-dimensional vector, which is the descriptor for the feature point. This descriptor contains gradient information from the region surrounding the feature point, exhibiting uniqueness and robustness, and accurately describes the characteristics of the feature point.

[0049] This embodiment innovatively achieves efficient and accurate extraction of feature points of the storage location plane by accelerating calculation with integral images, detecting accurate Hessian matrix feature points, and generating descriptors with rotation invariance. This provides a reliable feature basis for the subsequent accurate calculation of the angle between the storage location plane and the horizontal plane, as well as the closed-loop adjustment of the storage location angle.

[0050] In practical applications, the angle of the inbound storage location is measured in real time by the vision camera 7 installed on the lower side of the center of the cargo platform 6. When the measured angle is positive, it is determined that the left side of the inbound storage location is higher (when the right side of the AGV docks, it is directly facing the left side of the storage location).

[0051] Given the distance between the power units is L, according to the formula ΔH = (α × L × tanθ) + β (where ΔH is the required height difference and θ is the positive angle measured by the vision camera 7), when the distance between the power units is L = 900mm and the angle measured by the vision camera 7 is θ = 2°, the required height ΔH = 900tan(2°) = 900 * 0.0349 = 31.41mm is calculated. Setting the reduction ratio to n, and with the sprocket mounted on the output shaft of the reduction gearbox, the distance S that the platform moves when the sprocket rotates once is determined based on its size. When the reduction ratio n = 30 and the platform moves a distance S = 76.35mm when the sprocket rotates once, the number of rotations required by the motor is calculated as C = ΔH / Sn = 31.41 / 76.3530 = 12.342 rotations.

[0052] The calculated number of motor rotations is sent to the servo system, which controls the motor to rotate the corresponding number of times, thereby raising the right-side lifting platform 4 to the required height. After the motor has finished rotating, the vision camera 7 measures the cluster storage position angle again. If the angle is 0°, the adjustment is complete; if the angle is not 0°, the above steps are repeated until the angle is accurately adjusted.

[0053] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0054] The above embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of protection of the present invention. Any non-substantial changes and substitutions made by those skilled in the art based on the present invention shall fall within the scope of protection claimed by the present invention.

Claims

1. A battery module angle adaptive adjustment mechanism based on dual-power lifting, characterized in that, include: The power actuators are arranged symmetrically on the left and right sides. Each power actuator includes at least one servo motor and a sprocket and chain transmission mechanism to drive the vertical movement of the lifting platforms on both sides. A lifting platform connecting two power actuators, wherein the left lifting platform is connected to the left power actuator and the right lifting platform is connected to the right power actuator; The flexible link structure includes a hinge shaft, a cam groove structure, and support rollers. When there is a height difference between the two lifting platforms, the cargo platform can achieve tilt adjustment by rotating the hinge shaft and cooperating with the cam groove structure and the rolling of the support rollers. A vision camera installed on the lower side of the center of the cargo platform is used to measure the angle information of the inbound storage location in real time; The main control module receives the angle information measured by the vision camera, calculates the required lifting height difference between the two power actuators, and controls the power actuators to adjust the height. The adjustment stops when the angle measured by the vision camera reaches 0° through a closed-loop control process.

2. The mechanism according to claim 1, characterized in that, It also includes a lifting guide rail assembly, which comprises: The lifting linear guide rails are respectively set on both sides of the lifting platform. The lifting linear guide rails are fixedly connected to the frame and serve as the vertical guide structure of the lifting platform to prevent the platform from deviating or tilting during movement and to ensure that the lifting platform always remains vertical. Each side of the lifting linear guide rail is connected to the corresponding lifting platform through two front and rear lifting linear guide rails. The four guide rail sliders form a rectangular layout to ensure the stable movement of the lifting platform in the vertical direction.

3. The mechanism according to claim 1, characterized in that, The flexible link structure is specifically as follows: The left lifting platform is connected to the hinge shaft on the left side of the cargo platform via a bearing, forming a pivot point for rotation. The surface of the right lifting platform is provided with a cam groove structure, and the support roller on the right side of the cargo platform is embedded in the cam groove structure; When there is a height difference between the two lifting platforms, the cargo platform rotates around the hinge axis, and the support rollers roll along the trajectory of the cam groove structure, so as to achieve a smooth transition of the cargo platform from a horizontal state to an inclined state.

4. The mechanism according to claim 1, characterized in that: When the height of the right-side lifting platform is higher than that of the left-side lifting platform, the cargo platform tilts using the following structure: With the mounting bearing of the left lifting platform as the center of rotation, the support roller on the right side of the cargo platform moves upward in an arc along the cam groove structure trajectory of the right lifting platform. The curved surface design of the cam groove structure enables the support roller to generate horizontal displacement synchronously during the rolling process, forming a compound motion mechanism of rolling and sliding, so that the right edge of the cargo platform tilts upward while smoothly shifting to the left. During the tilting process, the angle θ between the cargo platform and the horizontal plane is monitored in real time by a vision camera. When θ=0°, the main control module stops the adjustment to ensure that the cargo platform reaches a horizontal state.

5. A method for adaptive adjustment of battery module angle based on dual-power lifting, characterized in that, The method employs the battery module angle adaptive adjustment mechanism based on dual-power lifting as described in any one of claims 1 to 4, and includes the following steps: The AGV carrying the battery module moves to the designated docking position of the energy storage cabinet to complete the spatial position calibration; After receiving the target storage location information, the AGV calculates the required lifting height of the two lifting platforms. The main control system drives the servo motors on both sides to operate synchronously, and controls the two lifting platforms to rise vertically to the calculated height through the sprocket and chain transmission mechanism. The vision camera installed on the lower side of the center of the cargo platform makes the first measurement of the angle θ between the storage location plane and the horizontal plane. The θ value includes positive and negative direction indicators. The vision camera feeds back the θ value to the main control system in real time. The main control system calculates the required height difference ΔH between the two platforms based on the θ value and the distance between the two power actuators. The ΔH value includes positive and negative direction indicators. The main control system controls the operation of the right-side servo motor based on the ΔH value: when ΔH is positive, it drives the right-side lifting platform to rise; when ΔH is negative, it drives the right-side lifting platform to fall. During the adjustment process, the cargo platform tilts through the flexible connection between the hinge shaft and the cam groove structure; the support rollers roll along the trajectory of the cam groove structure and generate horizontal displacement, so that the cargo platform smoothly transitions to the target angle; The vision camera performs a second measurement on the adjusted storage location angle to obtain a new θ value; If the new angle θ value is not equal to 0°, then ΔH is recalculated and adjustment is performed, forming a closed-loop control process of measurement-calculation-adjustment, until the adjustment is terminated when θ = 0°; When the visual camera detects that the angle deviation of the storage location is less than the preset threshold, the main control system determines that the adjustment is complete and issues a command to allow the battery module to enter the cluster, thus completing the entire adjustment process.

6. The method according to claim 5, characterized in that, The main control system calculates the required height difference ΔH between the two lifting platforms using the following formula: ΔH=(α×L×tanθ)+β ΔH is the height difference between the two lifting platforms; L is the distance between the power execution units on the left and right sides; θ is the angle between the storage location plane and the horizontal plane measured by the vision camera for the first time; α is the environmental correction factor; when calculating the height difference, a dynamic compensation term β is added. The main control system obtains the lifting speed v and load weight m in real time through the speed sensor and weight sensor installed on the lifting platform, and calculates the dynamic compensation term β in combination with the preset mechanical characteristic parameters. Based on the calculated ΔH value, the main control system generates adjustment commands for the right-side servo motor through a closed-loop control algorithm, thereby realizing the differential height adjustment of the two lifting platforms and calculating the required height difference between the two platforms.

7. The method according to claim 5, characterized in that: After the relevant movements of the storage location are completed, the main control system sends a measurement command to the vision camera, triggering the vision camera to perform image acquisition and measurement operations on the storage location; The visual camera transmits the acquired storage location image information to the main control system. The main control system uses a preset visual algorithm to analyze and process the image, and calculates the angle θ between the current storage location plane and the horizontal plane. Based on the calculated included angle θ and the geometric dimensions of the storage location, the main control system calculates the height difference ΔH between the left and right lifting platforms; further, based on the height difference ΔH and the transmission parameters of the servo motor of the right lifting platform, it calculates the number of rotations N required by the servo motor of the right lifting platform. The main control system sends the calculated number of revolutions N to the servo system. The servo system controls the right lifting platform motor to rotate according to the required number of revolutions based on the command, and feeds back the motor rotation information in real time through the encoder to realize servo closed-loop control and ensure that the right lifting platform is accurately adjusted to the target height. After one adjustment is completed, the main control system notifies the vision camera to take measurements again, repeating the angle calculation, height difference and number of circles calculation, and servo adjustment steps until the storage position angle measured by the vision camera is 0°, realizing closed-loop adjustment of the entire system and ensuring that the storage position reaches a horizontal state.

8. The method according to claim 7, characterized in that: After the main control system acquires the original images of the storage location captured by the vision camera, it first performs image preprocessing. The median filtering algorithm is used to remove noise interference from the image. The median filter sorts the neighboring pixel values ​​of each pixel in the image and takes the median value as the new pixel value of that point. Then, histogram equalization technology is used to enhance the contrast of the denoised image. On the preprocessed image, the main control system uses an accelerated segmented feature transformation algorithm to extract feature points of the storage location plane; at the same time, it generates a unique descriptor for each feature point, which at least contains gradient information of the surrounding area of ​​the feature point; then, it matches the extracted feature points with feature points of the pre-stored standard horizontal storage location plane template image, and uses the nearest neighbor ratio method for matching and filtering, that is, it calculates the Euclidean distance between feature point descriptors and sets a distance threshold, and only retains matching point pairs with a distance less than the threshold. For a successfully matched pair of feature points, the main control system uses a random sampling consensus algorithm to estimate the geometric transformation model. Based on the optimal geometric transformation model, the rotation angle of the current storage location plane relative to the standard horizontal storage location plane is calculated, that is, the angle θ between the current storage location plane and the horizontal plane.

9. The method according to claim 8, characterized in that, The accelerated segmented feature transformation algorithm is used to extract feature points of the storage location plane, including the following steps: The main control system first converts the original image of the storage location plane captured by the vision camera into a grayscale image, and then constructs an integral image by traversing each pixel of the grayscale image; For any point in the integral image ( x , y ), whose value is from (0,0) to ( x , y The sum of the gray values ​​of all pixels within the rectangular area formed by ) is, i.e. ,in I ( i , j ) represents the original image at point ( i , j The grayscale value at () Based on the constructed integral image, the main control system constructs the Hessian matrix in different scale spaces of the storage location planar image, for each pixel in the image. X =( x , y ), in scale σ Hessian matrix under H ( X , σ ) is defined as: in, L xx ( X , σ ) is the second-order partial derivative of Gauss. With images I Convolution at point X The value at that location.

10. The method according to claim 9, characterized in that: The determinant of the Hessian matrix is ​​calculated using the following formula: By comparing the determinant value of each pixel at different scales with the determinant values ​​of its surrounding neighboring pixels, if the determinant value of that pixel is a local maximum, it is marked as a candidate feature point. The non-maximum suppression method is used to filter in scale space and image space, retaining the feature points with the strongest response, thereby detecting feature points in the storage location planar image; Centered on the detected feature point, within a radius of 6 s Within a circular neighborhood, calculate the position of all pixels. x and y The Haar wavelet response of the direction is obtained by sliding and accumulating these response values ​​in a 60° fan-shaped window centered on the feature point. The direction of the fan-shaped window with the largest accumulated value is the main direction of the feature point. A 20s×20s square region is selected around the feature point and rotated to the main direction of the feature point. Then, the region is divided into 4×4 sub-regions. In each sub-region, the cumulative values ​​of the Haar wavelet responses in the x and y directions and the cumulative values ​​of the absolute values ​​of the responses are calculated to form a 4-dimensional vector. Combine the vectors of all sub-regions to form a 4×4×4 = 64-dimensional vector, which is the descriptor of the feature point.

Images