Tail gate control method and system, medium and program product

Abstract

The invention discloses a tail gate control method and system, a medium and a program product. The method comprises the following steps: acquiring an environment image in a tail gate motion envelope through a camera; inputting the environment image into a pre-trained travel prediction model, and predicting the distance between a tail gate and an obstacle; judging whether the tail gate meets an opening condition or not according to the predicted distance; if the opening condition is met, the tail door is controlled to be opened; wherein in the opening process of the tail gate, images are collected in real time and input into the travel prediction model, and the distance between the tail gate and an obstacle is continuously predicted; and adjusting the opening stroke of the tail gate according to the continuously predicted distance. The possible collision risk of the tail gate can be pre-judged in advance, personnel or vehicle damage is avoided, and the after-sales maintenance cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control, and in particular to a tailgate control method, system, medium and program product. Background Art

[0002] At present, the anti-collision function of the electric tailgate of the vehicle mainly relies on the Hall motor. During the movement of the tailgate, if it encounters an obstacle, the movement speed will decrease and the motor speed will decrease accordingly. The Hall sensor will send the information of the pulse width change to the tailgate controller. When the difference between the pulse width and the reference value exceeds the set threshold, the tailgate controller will trigger the anti-collision program and execute the reverse action of the tailgate to avoid further damage.

[0003] However, the above anti-collision solution requires that the obstacle be identified and reversed only after the collision occurs through speed changes, which cannot prevent the occurrence of collisions and is likely to cause minor damage to the vehicle or obstacles. In addition, the solution lacks the ability to accurately identify the type and location of obstacles, and the anti-collision performance is limited. More importantly, this technology fails to provide intelligent tailgate opening travel prediction, making it difficult to meet the high requirements of modern users for vehicle intelligence and safety. Summary of the invention

[0004] In view of the above technical problems, the present invention provides a tailgate control method, system, medium and program product, which can predict possible collision risks in advance, avoid damage to people or vehicles, and reduce after-sales maintenance costs.

[0005] A first aspect of the present invention provides a tailgate method, comprising: The environment image within the tailgate motion envelope is collected by a camera; Inputting the environment image into a pre-trained travel prediction model to predict the distance between the tailgate and the obstacle; Determine whether the tailgate meets the opening condition based on the predicted distance; if the opening condition is met, control the tailgate to open; In the process of opening the tailgate, images are collected in real time and input into the travel prediction model to continuously predict the distance between the tailgate and the obstacle; The tailgate opening distance is adjusted based on the continuously predicted distance.

[0006] In a possible implementation, the travel prediction model is trained by the following steps: Collect image data at different tailgate opening angles; Annotating the image data, wherein the annotation content includes a tailgate angle and an obstacle position; Use the labeled image data to train the preset neural network model.

[0007] In a possible implementation, during the tailgate opening process, real-time image acquisition and input into the travel prediction model to continuously predict the distance between the tailgate and the obstacle include: The environment video within the tailgate motion envelope is collected by the camera; Decomposing the environment video into multiple frame images according to a preset frame rate; Each frame of image is input into the travel prediction model to obtain the distance between the tailgate and the obstacle corresponding to each frame of image.

[0008] In a possible implementation, adjusting the opening stroke of the tailgate according to the continuously predicted distance includes: When the predicted distance is less than the preset safety distance, the tailgate opening angle is reduced or the tailgate opening is stopped.

[0009] In a possible implementation, the method further includes: The maximum permissible opening angle of the tailgate is determined based on the predicted distance.

[0010] In a possible implementation, determining the maximum allowable opening angle of the tailgate according to the predicted distance includes: Comparing the predicted distance with a preset safety distance threshold, and calculating a maximum allowable opening angle according to the comparison result; When the predicted distance is greater than the safety distance threshold, the maximum allowable opening angle is set to a preset maximum angle; When the predicted distance is less than or equal to the safety distance threshold, the maximum allowable opening angle is calculated according to the ratio of the predicted distance to the safety distance threshold.

[0011] In a possible implementation, the method further includes: When the predicted distance is less than the preset warning distance, a warning signal is issued.

[0012] A second aspect of the present invention provides a tailgate control system, comprising: An image acquisition module, at least used to acquire an environmental image within the tailgate motion envelope through a camera; a travel prediction module, at least used to input the environment image into a pre-trained travel prediction model to predict the distance between the tailgate and the obstacle; A tailgate control module, at least used to determine whether the tailgate meets the opening condition according to the predicted distance; if the opening condition is met, control the tailgate to open; The travel prediction module is also used to collect images in real time and input them into the travel prediction model during the tailgate opening process, so as to continuously predict the distance between the tailgate and the obstacle; The tailgate control module is further configured to adjust the opening travel of the tailgate according to the continuously predicted distance.

[0013] According to a third aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a computer, the method according to the first aspect of the embodiment of the present invention is executed.

[0014] A fourth aspect of the present invention provides a computer program product, which includes a computer program. When the computer program is executed by a computer, the method according to the first aspect of the embodiment of the present invention is executed.

[0015] The present invention uses a camera to capture environmental images within the tailgate motion envelope, and uses a pre-trained travel prediction model to predict the distance between the tailgate and obstacles in real time, thereby realizing intelligent tailgate opening control and anti-collision functions. Compared with traditional sensor-based solutions, this method can predict the position of obstacles in advance before the tailgate opens, avoid collisions, and thus improve safety. At the same time, through real-time image acquisition and prediction, the tailgate opening stroke can be dynamically adjusted, and the tailgate opening control can be further optimized, thereby improving ease of use and the intelligence level of the vehicle. In addition, this method does not require the installation of additional hardware equipment, effectively reducing the cost of the entire vehicle, reducing after-sales maintenance requirements, and improving customer experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The figure is a flowchart of an embodiment of a tailgate control method of the present invention.

[0017] Figure 2 A schematic diagram of an application scenario of the tailgate control method of the present invention.

[0018] Figure 3 FIG. 4 is a flowchart of another embodiment of a tailgate control method of the present invention.

[0019] Figure 4 FIG. 4 is a flow chart of an embodiment of a tailgate control method of the present invention.

[0020] Figure 5 FIG. 1 is a schematic structural diagram of an embodiment of a tailgate control system of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0022] It should be understood that the terms "first", "second", and "third" etc. in the claims, specifications, and drawings of the present disclosure are used to distinguish different objects rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of the present disclosure indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections. It should also be understood that the terms used in this disclosure specification are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure.

[0023] Reference Figure 1 , an embodiment of the present invention provides a tailgate control method, comprising the following steps.

[0024] S1, collects the environmental image within the tailgate motion envelope through the camera.

[0025] Specifically, refer to Figure 2 In step S1, a camera installed on the tailgate of the vehicle is used to collect an environmental image within the tailgate motion envelope. The camera can work at different tailgate opening angles and continuously or intermittently capture images throughout the entire process from fully closed to fully opened.

[0026] In some embodiments, in order to ensure the comprehensiveness and applicability of the data, image acquisition can be performed under different weather conditions (such as sunny days, rainy days, foggy days, nights, etc.) and different scenes (such as garages, parking lots, outdoors, etc.).

[0027] In addition, in some embodiments, in order to improve image quality and model training effects, the captured images can be enhanced, such as adding noise, adjusting lighting conditions, etc., so as to generate richer and more diverse data sets to support subsequent model training and application.

[0028] S2, inputting the environment image into a pre-trained travel prediction model to predict the distance between the tailgate and the obstacle.

[0029] Specifically, in step S2, the environment image within the tailgate motion envelope captured by the camera is input into a pre-trained travel prediction model.

[0030] In some embodiments, the trip prediction model can be trained based on a variety of visual algorithms (such as image recognition, image segmentation, etc.), and the model type used can be a deep learning framework such as ResNet, VGG or Faster R-CNN.

[0031] During the training process, an image dataset with annotated tailgate angles and obstacle locations was used, and training and validation sets were generated under various weather conditions and scenarios to improve the generalization ability of the model.

[0032] The image input into the travel prediction model can be a single frame image or a sequence of frames extracted from a video, ensuring that the model can fully perceive the environmental information during the movement of the tailgate, thereby accurately predicting the distance between the tailgate and the obstacle, providing a basis for subsequent control.

[0033] S3, judging whether the tailgate meets the opening condition according to the predicted distance; if the opening condition is met, controlling the tailgate to open.

[0034] Step S3 determines whether the tailgate meets the opening condition based on the distance between the tailgate and the obstacle output by the travel prediction model. Specifically, the controller compares the predicted distance with the set safety threshold: if the distance is greater than the threshold, it is determined that the opening condition is met and the tailgate can continue to operate as planned; if the distance is less than or equal to the threshold, it is determined that the opening condition is not met, and an alarm prompt (such as sound, text or multiple forms) is used to notify the driver to adjust the vehicle position or prevent the tailgate from opening.

[0035] In some embodiments, the process can dynamically adjust the safety threshold to adapt to different scene requirements (such as parking lots or outdoor spaces) to ensure that the tailgate action is both safe and efficient.

[0036] S4, during the tailgate opening process, real-time image collection and input into the travel prediction model to continuously predict the distance between the tailgate and the obstacle.

[0037] Specifically, during the tailgate opening process, the camera is used to collect environmental images within the tailgate motion envelope in real time, and these images are input into a pre-trained travel prediction model to continuously predict the distance between the tailgate and the obstacle.

[0038] Images acquired in real time can be extracted from the video stream at a fixed frame rate, ensuring that key environmental changes during the tailgate movement are captured.

[0039] In order to adapt to different environmental conditions (such as weather and light changes), real-time images can be preprocessed, such as denoising or contrast enhancement, to improve the accuracy of the prediction model.

[0040] The travel prediction model will dynamically adjust the predicted results of the distance between the tailgate and the obstacle based on the image data of consecutive frames, thereby providing real-time judgment basis for the next action of the tailgate and ensuring the safe and smooth opening of the tailgate.

[0041] S5, adjusting the opening stroke of the tailgate according to the continuously predicted distance.

[0042] Step S5 dynamically adjusts the opening stroke of the tailgate according to the continuously predicted distance between the tailgate and the obstacle. Specifically, the distance information of the stroke prediction model is continuously received during the opening process: if the predicted distance is always within the safety range, the tailgate will continue to open at the preset opening degree and speed; if it is detected that the distance between the tailgate and the obstacle is gradually decreasing and approaching the safety threshold, the control module will adjust the opening angle of the tailgate, limit its maximum opening degree, and only partially open it to avoid collision; if the distance is less than the set safety threshold, the tailgate will immediately stop opening, and even perform a reverse operation to ensure safety.

[0043] In addition, the opening speed of the tailgate can also be adjusted according to real-time distance information to achieve a smoother movement process.

[0044] Through these measures, the tailgate can intelligently adjust its opening stroke according to environmental changes, avoid collisions with obstacles, and improve safety and user experience.

[0045] Further, as an embodiment of the present invention, the travel prediction model is trained by the following steps: Collect image data at different tailgate opening angles; Annotating the image data, wherein the annotation content includes a tailgate angle and an obstacle position; Use the labeled image data to train the preset neural network model.

[0046] The training of the trip prediction model first completes data preparation by collecting image data at different tailgate opening angles. Specifically, the camera's field of view image can be collected at fixed angles (such as every 3°) during the process of the tailgate from fully closed to fully open to ensure that all key states within the tailgate motion envelope are covered. Data collection needs to be carried out in a variety of scenarios (such as garages, parking lots, outdoors, etc.) and weather conditions (such as sunny days, rainy days, nights, etc.) to ensure data diversity and model adaptability. In addition, data enhancement can be performed by adding noise to the original image, changing brightness or contrast, etc. to further enrich the training data.

[0047] Next, the collected image data is annotated, including the tailgate angle and obstacle position. The tailgate angle can be calibrated according to the current state of the tailgate, while the obstacle position is marked on the image manually or automatically. The accuracy and consistency of the data must be ensured during the annotation process to ensure the training effect of the model. After the annotation is completed, the annotated image data is divided into a training set and a validation set according to a certain ratio (such as 8:2) for model training and evaluation.

[0048] Finally, the annotated image data is used to train a preset neural network model, such as ResNet, VGG, or FasterR-CNN. The model training process includes setting hyperparameters (such as learning rate, batch size, number of iterations, etc.), inputting training data, performing multiple iterations, and updating model parameters. After training, the generalization performance of the model is evaluated using the validation set to ensure that the model can accurately predict the tailgate angle and its distance from the obstacle. After the model is verified, it can be deployed to the vehicle control system to achieve real-time tailgate travel prediction.

[0049] Further, as an embodiment of the present invention, refer to Figure 3 , step S4, during the tailgate opening process, real-time image collection and input into the travel prediction model, and continuous prediction of the distance between the tailgate and the obstacle, including the following steps.

[0050] S41, collects the environment video within the tailgate motion envelope through the camera.

[0051] Specifically, a camera installed on the tailgate collects environmental video within the tailgate motion envelope in real time to capture dynamic changes in the environment during the tailgate opening process.

[0052] The camera can be a high-definition wide-angle camera to ensure that the entire field of view of the tailgate motion envelope is covered. During the video acquisition process, the camera needs to continuously record the environmental changes during the tailgate motion at a fixed frame rate (e.g., 30 frames per second).

[0053] To adapt to different usage environments, video acquisition can be carried out in a variety of weather conditions (such as sunny days, rainy days, snowy days, foggy days, and nights) and scenes (such as garages, narrow parking spaces, and outdoor spaces) to ensure the diversity and applicability of the collected data.

[0054] In addition, the video quality can be optimized by calibrating the camera's focal length, aperture, and exposure parameters in real time to ensure the accuracy of subsequent image decomposition and model input.

[0055] At the same time, in order to reduce the system burden and ensure real-time performance, the video captured by the camera can be compressed to reduce the amount of data while retaining key information.

[0056] S42, decomposing the environment video into multiple frame images according to a preset frame rate.

[0057] Specifically, the environmental video captured by the camera is decomposed into multiple frames of images according to a preset frame rate so as to be subsequently input into the travel prediction model for analysis.

[0058] The frame rate setting can be adjusted according to the speed and real-time requirements of the tailgate movement. For example, the common frame rate is 15 to 30 frames per second to ensure sufficient time resolution to capture environmental changes.

[0059] During the video decomposition process, a time interval-based frame extraction algorithm can be used to evenly extract key frames from the video stream, or the frame rate can be dynamically adjusted according to the important stages of the tailgate movement (such as acceleration and deceleration) to optimize the computational efficiency.

[0060] To further improve the image quality, the decomposed frame images can be preprocessed, including denoising, color correction, contrast enhancement and other operations, to ensure that the trip prediction model can accurately identify the relationship between the tailgate and the obstacle.

[0061] In addition, during the processing, the frame images can be timestamped to facilitate the subsequent analysis of the tailgate movement trajectory and dynamic changes. Through these measures, the process of decomposing the video into frame images can efficiently and accurately support the tailgate travel prediction function.

[0062] S43, inputting each frame of image into the travel prediction model to obtain the distance between the tailgate and the obstacle corresponding to each frame of image.

[0063] Specifically, each frame of the image decomposed in step S42 is input into a pre-trained travel prediction model to predict the distance between the tailgate and the obstacle in real time.

[0064] After receiving the frame image, the trip prediction model uses a deep learning algorithm (such as ResNet, VGG or Faster R-CNN) to process the image, extract key features such as the tailgate position, obstacle shape and relative position, and calculate the distance between the tailgate and the obstacle based on the model's training parameters.

[0065] To ensure real-time performance, the trip prediction model can be deployed on a high-performance computing unit on board the vehicle, or the results can be quickly fed back to the vehicle controller after being processed in the cloud.

[0066] In addition, to improve prediction accuracy, the travel prediction model can adjust and correct the distance by combining the dynamic features of continuous frame images (such as the movement trajectory of objects).

[0067] During this process, the model output results can also be dynamically optimized according to the obstacle type (such as stationary objects, moving objects) to support smarter tailgate control and collision avoidance functions.

[0068] This continuous prediction process ensures that the tailgate can avoid obstacles in real time during the opening process, achieving precise and safe dynamic control.

[0069] Further, as an embodiment of the present invention, step S5, adjusting the opening stroke of the tailgate according to the continuously predicted distance, includes: when the predicted distance is less than a preset safety distance, reducing the opening angle of the tailgate or stopping the opening of the tailgate.

[0070] Specifically, by continuously predicting the distance between the tailgate and the obstacle, the tailgate opening stroke is adjusted in real time to ensure safety. When the predicted distance is less than the preset safety distance, the tailgate controller will dynamically reduce the tailgate opening angle or directly stop the tailgate opening action based on the output of the stroke prediction model.

[0071] This adjustment can be achieved by reducing the opening speed of the electric tailgate or limiting the maximum opening angle of the tailgate to avoid further approaching obstacles.

[0072] In addition, the action strategy can be flexibly adjusted according to different environments and obstacle types (such as stationary or moving obstacles). For example, when the obstacle moves, the opening action is paused and the safe distance is recalculated. This method can not only prevent the tailgate from colliding with obstacles, but also reduce user intervention through intelligent control, thereby improving the safety and user experience of the vehicle.

[0073] Furthermore, as an implementation manner of the present invention, the tailgate control method described in the embodiment of the present invention further includes: determining a maximum allowable opening angle of the tailgate according to the predicted distance.

[0074] Specifically, determining the maximum allowable opening angle of the tailgate according to the predicted distance includes: Comparing the predicted distance with a preset safety distance threshold, and calculating a maximum allowable opening angle according to the comparison result; When the predicted distance is greater than the safety distance threshold, the maximum allowable opening angle is set to a preset maximum angle; When the predicted distance is less than or equal to the safety distance threshold, the maximum allowable opening angle is calculated according to the ratio of the predicted distance to the safety distance threshold.

[0075] Specifically, the predicted distance between the tailgate and the obstacle is compared with the preset safety distance threshold. If the predicted distance is greater than the safety distance threshold, the tailgate can be opened safely, and the maximum allowable opening angle is directly set to the preset maximum angle; if the predicted distance is less than or equal to the safety distance threshold, the maximum allowable opening angle is dynamically adjusted by calculating the ratio of the predicted distance to the safety distance threshold to limit the opening of the tailgate and avoid contact with obstacles. This method, combined with real-time distance prediction, can not only make full use of the safe space to achieve the maximum opening of the tailgate, but also effectively prevent the tailgate from colliding with obstacles, thereby improving the flexibility and intelligence of the control strategy.

[0076] For example, when the tailgate of the vehicle is in motion, the travel prediction model detects that the predicted distance between the tailgate and the obstacle is 50 centimeters, and the preset safety distance threshold is 40 centimeters, then the predicted distance is judged to be greater than the safety distance threshold, so the maximum allowable opening angle of the tailgate is directly set to the preset maximum angle (such as 70 degrees). If in another case, the predicted distance is 30 centimeters, which is less than the safety distance threshold, the maximum allowable opening angle will be adjusted to 75% of the preset maximum angle (such as 70 degrees × 0.75 = 52.5 degrees) by calculating the ratio of the predicted distance to the safety distance threshold (30 / 40=0.75). This dynamic adjustment process ensures the safe opening of the tailgate under different distance conditions, while maximizing the available opening of the tailgate and improving the user experience.

[0077] Furthermore, as an implementation manner of the present invention, the tailgate control method described in the embodiment of the present invention further includes: when the predicted distance is less than a preset warning distance, issuing a warning signal.

[0078] Specifically, when the predicted distance between the tailgate and the obstacle is less than the preset warning distance, a warning signal is issued in a variety of ways to remind the user to intervene or adjust the vehicle status in time. The implementation methods of the warning signal include but are not limited to: displaying warning information on the vehicle display screen (such as "tailgate opening is restricted, adjust the vehicle position"), issuing a sound warning through a buzzer (continuous or intermittent audio can be used, and the volume and frequency are adjusted according to the degree of danger), and prompting the driver through the dashboard indicator light or vibration feedback device. The triggering conditions and forms of the warning signal can be flexibly adjusted according to the vehicle configuration and usage scenarios. For example, more obvious sound prompts are used in parking lots, and visual prompts are given priority in quiet places. In addition, these warning signals can be linked with the tailgate motion control. When the warning signal is triggered, the tailgate action is automatically paused or stopped to further ensure the safety of the vehicle and the environment.

[0079] Reference Figure 4 The implementation process of a specific embodiment of a tailgate control method of the present invention includes: First, environmental data images are collected: the tailgate camera collects environmental images within the tailgate motion envelope, providing real-time visual input data for analyzing obstacles and surrounding environment within the tailgate motion path; Next, start the trip prediction: input the collected environmental image into the trip prediction model to predict the distance between the tailgate and the obstacle. If the trip does not meet the preset opening conditions, go directly to step 3 to trigger the trip not meeting the alarm prompt (such as sound, text or light prompt) to remind the user to adjust the vehicle position or tailgate operation;

[0080] Next, the tailgate opening control and condition judgment are performed: if the tailgate meets the opening conditions, the tailgate control operation continues. Otherwise, a prohibition opening warning prompt is issued, and step 8 is entered at the same time, and the tailgate stops moving;

[0081] Next, the image information of the tailgate movement process is input: during the tailgate opening process, the camera collects images in real time and inputs them into the travel prediction model to continuously predict the distance between the tailgate and the obstacle; Then, the tailgate's motion state and travel are determined: the tailgate's motion state and allowed travel are determined dynamically based on the real-time image and the data output by the prediction model; If environmental conditions permit and there are no obstacles, the tailgate will be fully opened according to the set maximum opening angle. If it is detected that the distance between the tailgate and the obstacle is not enough to fully open, the tailgate opening angle will be adjusted proportionally according to the predicted distance to avoid collision risks. When the tailgate does not meet the opening conditions, the user will be notified through sound, text or indicator light; if the conditions are not met, the tailgate action will be suspended to ensure safety.

[0082] Through the above implementation, the camera is used to collect environmental images in real time and combined with the travel prediction model to predict the location of obstacles in advance, ensuring that the tailgate is fully evaluated for safety before opening; at the same time, the opening angle can be continuously predicted and adjusted during the tailgate opening process to avoid collision risks. Through dynamic alarm prompts and a mechanism for proportional adjustment of the opening angle, this method not only ensures the safety of the vehicle and surrounding personnel, but also improves the intelligence level of tailgate opening, reduces the repair cost of collision damage, and provides users with a more convenient and safer use experience.

[0083] The embodiment of the present invention also discloses a tailgate control system.

[0084] Reference Figure 5 A tailgate control system includes an image acquisition module 1, a travel prediction module 2 and a tailgate control module 3.

[0085] The image acquisition module 1 includes a high-definition camera installed on the tailgate, which is used to collect environmental images and / or environmental videos within the tailgate movement envelope in real time. The camera needs to have a wide-angle or panoramic field of view to cover all areas within the possible movement range of the tailgate and can adapt to various weather and light conditions (such as sunny days, nights, rainy days, etc.).

[0086] The image acquisition module 1 continuously acquires images at a preset acquisition frequency (e.g., 30 frames per second) to ensure that dynamic changes in the environment are captured. At the same time, in order to improve image quality and meet the needs of the trip prediction model, the image acquisition module 1 can include basic image preprocessing functions such as denoising, contrast enhancement, and brightness adjustment. The acquired image data will be transmitted to the trip prediction module 2 in the form of a digital signal for subsequent distance prediction and dynamic adjustment. The image acquisition module 1 can also support extended functions, such as dynamically adjusting image acquisition parameters (such as frame rate or resolution) according to the movement state of the tailgate, further improving acquisition efficiency and data quality.

[0087] The travel prediction module 2 is based on a travel prediction model run by the on-board computing unit, which receives the environmental image data from the image acquisition module 1 and analyzes it through a pre-trained deep learning model (such as ResNet, VGG or Faster R-CNN) to predict the distance between the tailgate and the obstacle.

[0088] The trip prediction module 2 has a built-in trip prediction algorithm, which can accurately identify the position and relative distance of obstacles within the tailgate motion envelope through steps such as image feature extraction, target detection and distance estimation.

[0089] To ensure real-time performance, the travel prediction module 2 can quickly process multiple frames of images per second and dynamically update the distance information between the tailgate and the obstacle. During the tailgate opening process, the travel prediction module 2 continuously receives the real-time acquired images, performs continuous predictions, and passes the analysis results to the tailgate control module 3 so that the tailgate opening travel can be adjusted in real time. The implementation of the travel prediction module 2 can accelerate the model reasoning process through embedded hardware (such as an on-board GPU) and support online adjustment of algorithm parameters to adapt to different environments and vehicle configuration requirements.

[0090] The tailgate control module 3 is used to receive the predicted distance data from the travel prediction module 2, and dynamically determine the opening condition and execution strategy of the tailgate based on this data. When the predicted distance between the tailgate and the obstacle is greater than the preset safety threshold, the tailgate control module 3 will send a command to start the tailgate motor and open the tailgate according to the preset maximum opening; when the predicted distance is less than or close to the safety threshold, the tailgate control module 3 will adjust the opening travel of the tailgate according to the real-time data provided by the travel prediction module, limit the maximum angle of the tailgate, or suspend the opening action to avoid collision.

[0091] In addition, the tailgate control module 3 can also trigger a variety of warning prompts (such as sound, light or vehicle display information) to remind the user that the opening conditions are not met. During the tailgate opening process, the tailgate control module 3 continuously receives real-time predicted distance data and dynamically adjusts the speed and angle of the tailgate movement to ensure safety and intelligent control under different environments and obstacles. The hardware of the tailgate control module 3 may include a drive unit that directly communicates with the vehicle motor and is connected to the vehicle bus to achieve linkage control with other systems.

[0092] The embodiment of the present invention also discloses a readable storage medium.

[0093] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the tailgate control method described in any one of the above embodiments.

[0094] The embodiment of the present invention also discloses a computer program product.

[0095] A computer program product comprises a computer program, wherein when the computer program is executed by a processor, the steps of the tailgate control method described in any one of the above embodiments are implemented.

[0096] It can be understood that computer-readable storage media may include: any entity or device capable of carrying a computer program, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), and a software distribution medium, etc. A computer program includes a computer program code. The computer program code may be in source code form, object code form, an executable file, or some intermediate form, etc. A computer-readable storage medium may include: any entity or device capable of carrying a computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), and a software distribution medium, etc.

[0097] In certain embodiments of the present invention, the electronic device may include a controller or a processor, and the controller is a single-chip microcomputer chip that integrates a processor, a memory, a communication module, etc. The processor may refer to a processor included in the controller. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.

[0098] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present invention belong.

[0099] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0100] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A tailgate control method, characterized in that: include: The environment image within the tailgate motion envelope is collected by a camera; Inputting the environment image into a pre-trained travel prediction model to predict the distance between the tailgate and the obstacle; Determine whether the tailgate meets the opening condition based on the predicted distance; if the opening condition is met, control the tailgate to open; In the process of opening the tailgate, images are collected in real time and input into the travel prediction model to continuously predict the distance between the tailgate and the obstacle; The tailgate opening distance is adjusted based on the continuously predicted distance.

2. The tailgate control method according to claim 1, characterized in that: The travel prediction model is trained by the following steps: Collect image data at different tailgate opening angles; Annotating the image data, wherein the annotation content includes a tailgate angle and an obstacle position; Use the labeled image data to train the preset neural network model.

3. The tailgate control method according to claim 1, characterized in that: During the tailgate opening process, real-time image acquisition and input into the travel prediction model are performed to continuously predict the distance between the tailgate and the obstacle, including: The environment video within the tailgate motion envelope is collected by the camera; Decomposing the environment video into multiple frame images according to a preset frame rate; Each frame of image is input into the travel prediction model to obtain the distance between the tailgate and the obstacle corresponding to each frame of image.

4. The tailgate control method according to claim 1, characterized in that: The adjusting the opening stroke of the tailgate according to the continuously predicted distance includes: When the predicted distance is less than the preset safety distance, the tailgate opening angle is reduced or the tailgate opening is stopped.

5. The tailgate control method according to claim 1, characterized in that: Also includes: The maximum permissible opening angle of the tailgate is determined based on the predicted distance.

6. The tailgate control method according to claim 5, characterized in that: Determining the maximum allowable opening angle of the tailgate according to the predicted distance includes: Comparing the predicted distance with a preset safety distance threshold, and calculating a maximum allowable opening angle according to the comparison result; When the predicted distance is greater than the safety distance threshold, the maximum allowable opening angle is set to a preset maximum angle; When the predicted distance is less than or equal to the safety distance threshold, the maximum allowable opening angle is calculated according to the ratio of the predicted distance to the safety distance threshold.

7. The tailgate control method according to claim 1, characterized in that: Also includes: When the predicted distance is less than the preset warning distance, a warning signal is issued.

8. A tailgate control system, characterized in that: include: An image acquisition module, at least used to acquire an environmental image within the tailgate motion envelope through a camera; a travel prediction module, at least used to input the environment image into a pre-trained travel prediction model to predict the distance between the tailgate and the obstacle; A tailgate control module, at least used to determine whether the tailgate meets an opening condition based on the predicted distance; if the opening condition is met, control the tailgate to open; The travel prediction module is also used to collect images in real time and input them into the travel prediction model during the tailgate opening process, so as to continuously predict the distance between the tailgate and the obstacle; The tailgate control module is further configured to adjust the opening travel of the tailgate according to the continuously predicted distance.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a computer, the tailgate control method according to any one of claims 1 to 7 is executed.

10. A computer program product, characterized in that The method comprises a computer program, and when the computer program is executed by a computer, the tailgate control method according to any one of claims 1 to 7 is executed.

Images