Control method and device of electric pedal, computer equipment and storage medium

Abstract

The invention relates to the technical field of vehicle control, and discloses an electric pedal control method and device, computer equipment and a storage medium, the method comprises the following steps: when a vehicle door of a vehicle is opened, determining a target user having a getting-on and getting-off demand at present and target feature information of the target user; and adjusting the stretching state of the electric pedal at the corresponding position according to the position of the target user, and synchronously adjusting the suspension height of the suspension in the vehicle according to the target feature information. When the vehicle door is opened, passengers with getting-on and getting-off requirements and characteristics of the passengers are recognized, the heights of the electric pedal and the suspension at the corresponding positions are actively adjusted accordingly, the dilemma that the auxiliary function is not enabled due to negligence of a driver or inconvenient operation is avoided, and convenience and safety of getting-on and getting-off of the passengers such as old people and children needing assistance are improved.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and specifically to control methods, devices, computer equipment, and storage media for electric pedals. Background Technology

[0002] With socio-economic development and a faster pace of life, automobiles have become an important tool for improving travel convenience. Some vehicle types, such as SUVs, have higher chassis and are often equipped with electric running boards and adjustable suspension systems to facilitate passenger entry and exit, especially for elderly people with mobility issues or short children. Currently, these convenient features are primarily activated through manual control by the driver, such as via key, buttons, or the central control screen menu.

[0003] The core flaw of this technology lies in its passivity, lacking the ability to intelligently sense and assess passengers outside the vehicle. The system cannot automatically identify the specific attributes (such as age or height) of passengers about to board or alight, thus failing to autonomously activate the corresponding assistance functions in the most needed scenarios (such as when elderly people or children are traveling). This reliance on manual operation not only reduces the convenience and technological sophistication of use but also, in cases of driver negligence, may leave passengers in need of assistance facing difficulties getting on and off the vehicle, impacting the actual travel experience and safety. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a control method, device, computer equipment, and storage medium for electric pedals to solve the problem that related technologies rely on manual operation, which leads to passengers who need assistance still facing difficulties in getting on and off the vehicle, affecting the actual travel experience and safety.

[0005] In a first aspect, embodiments of the present invention provide a control method for an electric pedal, the method comprising: When the vehicle door is opened, identify the target user who needs to get on or off the vehicle and the target user's target feature information; Adjust the extension state of the electric pedal at the corresponding position according to the location of the target user, and synchronously adjust the suspension height of the vehicle's suspension according to the target feature information.

[0006] This invention identifies passengers needing to get on or off the vehicle and their characteristics when the door opens, and proactively adjusts the electric pedals and suspension height accordingly. This avoids the predicament of driver assistance functions not being activated due to driver negligence or operational inconvenience, improving the convenience and safety of elderly people, children, and other passengers requiring assistance when getting on and off the vehicle. In conjunction with the first aspect, in one implementation, determining the target user with current boarding / alighting needs and the target user's target feature information includes: If the door is opened externally, then at least one first user is identified outside the vehicle and within a preset range of the opened door; and / or, if the door is opened internally, then a second user is identified inside the vehicle and within a preset range of the opened door. The first user and / or the second user are identified as target users with boarding and alighting needs; The target feature information corresponding to the target user is obtained from the user feature database, wherein the user feature database includes user feature information of multiple candidate users, and the user feature information is collected by the candidate user outside the vehicle or before getting into the vehicle by an external camera.

[0007] The user determination logic of this invention comprehensively covers various real-world getting on and off vehicle scenarios. Whether it is a person getting on the vehicle alone outside, a person getting off the vehicle alone inside, or even a complex situation where people are getting on and off simultaneously from both sides of the vehicle, the target user can be accurately determined by the door opening method and preset range. This ensures that the electric pedal and suspension adjustment functions can be accurately triggered at all necessary moments, fundamentally avoiding misjudgment or omission, and significantly improving the reliability and user experience in actual use.

[0008] In conjunction with the first aspect or its corresponding implementation, in one implementation, the process of constructing the user feature database includes: Acquire video data captured by the external cameras configured in the vehicle; Identify video frame images in the video data, obtain at least one candidate user included in the video frame image, and determine the user tag corresponding to the candidate user; Extract the face and body images of the candidate users based on the video frame images; The predicted age range of the candidate user is obtained by analyzing the face image and the body image, and the predicted height data of the candidate user is determined by analyzing the body image. A user feature library is constructed based on the user tags corresponding to each candidate user, the predicted age range, and the predicted height data.

[0009] The user feature database construction process in this invention makes full use of existing vehicle exterior cameras without adding extra hardware costs. By analyzing the video data collected by the cameras, the facial and body images of candidate users are extracted, and their age range and height data are accurately predicted based on this. Combined with user tags, a complete feature profile is formed, realizing the automated and non-contact collection and updating of user features, ensuring that the suspension and pedal assistance can accurately adapt to the actual needs of different users.

[0010] In conjunction with the first aspect or its corresponding implementation, in one implementation, the step of analyzing the face image and the body image to obtain the predicted age range corresponding to the candidate user includes: Extract key feature points from the face image, calculate the probability distribution of the candidate user's age based on the key feature points, and select the age range with the highest probability as the first age range of the candidate user. Identify the human body feature information of the candidate user from the human body image; Extract key points of the candidate user from the human feature information, and construct the skeleton sequence of the candidate user based on the key points; Gait features of the candidate users are extracted from the skeleton sequence, and a second age range of the candidate users is determined based on the gait features; Based on the first age range and the second age range, the predicted age range of the first candidate user is calculated.

[0011] This invention integrates facial features and gait features for dual age analysis. It not only calculates the first age range based on facial key points, but also extracts gait features from the human skeleton sequence to determine the second age range. Finally, it comprehensively obtains the prediction result, which improves the accuracy and reliability of age prediction, overcomes the errors that may exist based on a single basis, and makes the judgment of the user's age more accurate.

[0012] In conjunction with the first aspect or its corresponding implementation, in one implementation, the step of analyzing the human image to determine the predicted height data corresponding to the candidate user includes: Identify the body proportion data of the candidate user from the human body image; The predicted height of the candidate user is calculated based on the human body proportion data.

[0013] This invention does not rely on specific reference objects or sensors. Instead, it analyzes human images, identifies the user's body proportion data, and directly calculates the predicted height based on this data. This non-contact measurement method simplifies hardware requirements, reduces costs, and ensures that height estimation can be completed quickly and conveniently within the field of view of a conventional vehicle-mounted camera.

[0014] In conjunction with the first aspect or its corresponding implementation, in one implementation, adjusting the extension state of the electric pedal at the corresponding position according to the position of the target user, and synchronously adjusting the suspension height of the vehicle suspension according to the target feature information, includes: If the target user is currently in need of getting into the vehicle, the extension state of the electric pedal at the corresponding position is adjusted according to the target user's location, predicted age range, and / or predicted height data, and the suspension height of the vehicle's suspension is adjusted synchronously according to the target user's predicted age range and / or predicted height data. If the target user is currently getting off the vehicle, then it is determined whether the user's facial image is contained in the user feature database; if so, the extension state of the electric pedal at the corresponding position is adjusted according to the target user's location, and the suspension height of the vehicle is simultaneously adjusted according to the predicted age range corresponding to the target user; if not, the extension state of the electric pedal at the corresponding position is adjusted according to the target user's location and predicted age range, and the suspension height of the vehicle is simultaneously adjusted according to the predicted age range corresponding to the target user and / or predicted height data.

[0015] This invention distinguishes between two scenarios: getting into and getting out of the vehicle, and implements targeted adjustment strategies. Especially when getting out, it prioritizes querying a feature database to confirm the user's identity. If the user is registered, it directly uses their pre-stored features for precise adjustments; if the user is unfamiliar, it provides adaptive assistance based on real-time analysis of age and height data. This differentiated intelligent decision-making ensures that whether the passenger is a familiar family member or a temporary passenger, they receive the most suitable pedal and suspension assistance, enhancing its practicality in real-world driving scenarios.

[0016] In conjunction with the first aspect or its corresponding implementation, in one implementation, the step of synchronously adjusting the suspension height of the vehicle's suspension based on the predicted age range and / or predicted height data corresponding to the target user includes: Obtain a three-dimensional mapping table, wherein the three-dimensional mapping table includes the mapping relationship between age range, height data and suspension height reference value; Query the predicted age range and / or the height benchmark value corresponding to the predicted height data from the three-dimensional mapping table; Calculate the difference between the height reference value and the current height value of the suspension, and obtain the vehicle body posture parameters corresponding to the door position where the target user is located; The target suspension height is calculated using the difference and the vehicle body attitude parameters.

[0017] This invention introduces a three-dimensional mapping table to quantify predicted user characteristics into specific suspension height benchmark values. It further combines the current suspension height with real-time vehicle posture parameters for comprehensive calculation, ultimately obtaining a target suspension height suitable for the current scenario. This ensures that the adjustment of the suspension height not only meets the user's physical needs but also compensates for changes in vehicle posture caused by load and road conditions, thus providing stable and convenient support for users getting in and out of the vehicle under any circumstances.

[0018] Secondly, embodiments of the present invention provide a control device for an electric pedal, the device comprising: The information determination module is used to determine the target user who needs to get on or off the vehicle and the target user's target feature information when the vehicle door is opened. The pedal adjustment module is used to adjust the extension state of the electric pedal at the corresponding position according to the position of the target user, and to synchronously adjust the suspension height of the vehicle suspension according to the target feature information.

[0019] Thirdly, embodiments of the present invention provide a computer device, including: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the electric pedal control method of the first aspect or any corresponding embodiment described above.

[0020] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer instructions for causing a computer to execute the electric pedal control method described in the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0021] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating a control method for an electric pedal according to some embodiments of the present invention; Figure 2 This is a schematic diagram of the installation position of an exterior camera according to some embodiments of the present invention; Figure 3 This is a structural block diagram of the control device for an electric pedal according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] According to an embodiment of the present invention, a control method for an electric pedal is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0025] This embodiment provides a control method for an electric pedal, which can be used in a vehicle's infotainment system. Figure 1 This is a flowchart of a control method for an electric pedal according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps: Step S101: When the vehicle door is opened, determine the target user who needs to get on or off the vehicle and the target user's target feature information.

[0026] Step S102: Adjust the extension state of the electric pedal at the corresponding position according to the target user's position, and simultaneously adjust the suspension height of the vehicle's suspension according to the target feature information.

[0027] The electric pedal control method provided in this embodiment includes: when the vehicle door is opened, identifying the target user who needs to get on or off the vehicle and the target user's target feature information; adjusting the extension state of the electric pedal at the corresponding position according to the target user's position, and simultaneously adjusting the suspension height of the vehicle's suspension according to the target feature information. This invention identifies passengers who need to get on or off the vehicle and their characteristics when the door is opened, and proactively adjusts the electric pedal and suspension height at the corresponding position accordingly. This avoids the predicament of driver assistance functions not being activated due to driver negligence or operational inconvenience, improving the convenience and safety of elderly people, children, and other passengers who need assistance getting on and off the vehicle.

[0028] In step S101, when the vehicle door is opened, the target user who needs to get on or off the vehicle and the target user's target feature information are determined.

[0029] In this embodiment of the invention, the need to get on or off the vehicle refers to a user's clear behavioral motivation to enter or leave the vehicle, which needs to be comprehensively determined by combining the user's posture, movement trajectory, and the status of the vehicle door. For example, if a pedestrian approaches the vehicle but walks straight past, it does not constitute a need to get on or off the vehicle; while if another pedestrian slows down and reaches out towards the door handle, it constitutes a clear need to get on the vehicle.

[0030] In this embodiment of the invention, the opening method refers to the door opening mode. This embodiment divides the door opening mode into two types: external opening and internal opening. External opening refers to opening triggered by external signals such as physical buttons, capacitive touch sensors, force sensors on the door handle, or UWB radar keys. Internal opening refers to opening triggered by internal signals such as door handles, electronic switch buttons, or voice commands.

[0031] In this embodiment of the invention, the preset range is characterized as a dynamically defined three-dimensional perception space centered on the target vehicle door. It is a virtual boundary defined based on the effective detection area of ​​the sensor. The sensor can be selected from millimeter-wave radar, ultrasonic sensors, or cameras, etc. For example, for a sliding door, the preset range may be a rectangular area extending 1.2 meters outside the door opening trajectory and with a width equal to the width of the door; for a hinged door, it may be a fan-shaped area with a radius of 1 meter centered on the door hinge.

[0032] In this embodiment of the invention, the first user and the second user are target users distinguished based on the door opening method and user location. The first user is characterized as one or more users located outside the vehicle and within a preset range when the door is opened from the outside. When the door is opened from the outside, an external opening signal is triggered, and the vehicle system calls upon the side cameras and radar outside the vehicle to determine whether a target conforming to a human silhouette, i.e., the first user, exists within the preset range of the door. The second user specifically refers to a user located inside the vehicle and within a preset range when the door is opened from the inside. When the door is opened from the inside, an internal opening signal is triggered, and the vehicle system calls upon the cameras and radar inside the vehicle to determine whether a target conforming to a human silhouette, i.e., the second user, exists within the preset range of the door.

[0033] The user feature library in this embodiment of the invention is a pre-established and continuously updated database used to store and manage feature templates for several candidate users. Each candidate user has a corresponding feature profile in the library. For example, the library may record user A's height, body shape, preferred welcome lighting, and preset seat position, etc.

[0034] In this embodiment of the invention, the user feature information refers to the feature information used to identify the user's identity and preferences. The vehicle system captures the user's features, such as full-body gait, facial features, and clothing habits, through an external camera before the user enters the vehicle and under natural lighting and posture conditions. This avoids the impact of insufficient lighting inside the vehicle or seat obstruction on the accuracy of feature recognition.

[0035] Suppose Mr. Wang drives to pick up his wife, Ms. Li, from her office. Ms. Li walks towards the passenger side door, and Mr. Wang's features are already in the user feature database. Ms. Li reaches out and touches the external hidden door handle of the passenger side door and opens the door. The vehicle's infotainment system receives the external opening signal and, using a camera mounted on the right rearview mirror and a radar on the B-pillar, determines that a user is present 0.5 meters outside the passenger side door, thus marking Ms. Li as the first user intending to get in the car. The system then initiates a recognition process, capturing a real-time image of Ms. Li through the external camera and comparing it with the user feature database. Although Ms. Li is not the car owner, her features were previously recorded as a candidate user, and a match was successful, thus acquiring Ms. Li's target feature information. When the vehicle arrives at its destination, Mr. Wang's child, sitting in the back seat, touches the internal left door handle and opens the door. The system receives the internal opening signal and determines that the child is the second user inside the car, on the left side, intending to get out. Since the child's features are already in the database, the system can directly acquire their target feature information.

[0036] In one embodiment of the present invention, by sensing and recognizing users when the car door is opened, the triggering object and its characteristics for getting on and off the vehicle are determined. Specifically, the scenario is first divided according to the type of car door opening method. When the car door is opened externally, i.e., triggered by a person outside the vehicle through a door handle, button, or sensor, the vehicle system will lock the external space of the vehicle and delineate a three-dimensional preset range centered on the opened car door using a distance sensor or visual detection algorithm. For example, a fan-shaped area with a radius of 1.5 meters and the door hinge point as the origin. At least one user identified within the three-dimensional preset range is designated as the first user. That is, the first user in this embodiment of the present invention is the user within the three-dimensional preset range centered on the opened car door when the car door is opened externally. Accordingly, if the door opens from the inside, i.e., triggered by an occupant inside the vehicle via the interior door handle or voice command, the vehicle system will then monitor the interior space of the vehicle, defining the second user within a preset range of the entire interior space of the passenger compartment. That is, in this embodiment of the invention, the second user is defined as the user inside the passenger compartment when the door opens from the inside.

[0037] It should be noted that these two judgment modes support independent or parallel operation, and achieve coverage of in-vehicle and out-of-vehicle interaction scenarios through "AND / OR" logical association. For example, when there are people inside the vehicle assisting in opening the door and people outside the vehicle approaching simultaneously, the vehicle system can simultaneously identify dual targets. After identifying the first user and / or the second user, the vehicle system will group the first user (i.e., the person waiting to get on the vehicle outside) and / or the second user (i.e., the person waiting to get off the vehicle inside) as target users with getting on and off needs.

[0038] In one embodiment, considering that private vehicles are primarily used by families, the vehicle-mounted system of this invention also includes a pre-built user feature database. This database is a dynamic database formed based on historical data collected by the vehicle-mounted system, storing feature profiles of multiple candidate users. Specifically, a surround-view camera array deployed outside the vehicle collects features of candidate users outside the vehicle or before they get in, avoiding feature distortion caused by changes in lighting or seat obstruction after the user enters the vehicle. Furthermore, feature profiles for each candidate user are constructed based on these collected feature information, generating a user feature database consisting of at least one feature profile. In addition, the user feature database in this embodiment also supports user customization. For example, users can create a new feature profile in the database, fill in the new user's feature information, and upload a photo of the new user. Users can also modify or delete existing feature profiles in the user feature database.

[0039] The vehicle-mounted system compares the current target user's features with those in each feature profile in the user feature database in real time, and outputs target feature information bound to the target user. The target user's target feature information typically includes multimodal biometrics (such as height profile, gait patterns, facial features, etc.), behavioral habits (such as common approach angles, typical standing distances, etc.), and soft features (such as frequently carried items recognition, clothing style patterns, etc.). If no target user's target feature information is found in the user feature database, for example, if the target user is a new user riding in a private car for the first time, then facial recognition, age prediction, and height prediction are performed on the target user to generate a feature profile and store it in the user feature database. Preferably, a pop-up window can also be displayed on the vehicle's central control screen to remind the driver whether to save the target user's feature profile. Specifically, the steps include the following: If the door opens externally, at least one first user is identified outside the vehicle and within a preset range of the opened door; and / or, if the door opens internally, a second user is identified inside the vehicle and within a preset range of the opened door; the first user and / or the second user are identified as target users with the need to get on and off the vehicle; target feature information corresponding to the target user is obtained from the user feature database, wherein the user feature database includes user feature information of multiple candidate users, and the user feature information is collected by the external camera of the candidate user outside the vehicle or before getting on the vehicle.

[0040] The user identification logic in this embodiment comprehensively covers various real-world vehicle entry and exit scenarios. Whether it's a person getting on the vehicle alone outside, a person getting off the vehicle alone inside, or even a complex situation where people are getting on or off from both sides of the vehicle simultaneously, the system can accurately identify the target user based on the door opening method and preset range. This ensures that the electric pedal and suspension adjustment functions are accurately triggered at all necessary moments, fundamentally avoiding misjudgments or omissions, and significantly improving reliability and user experience in actual use.

[0041] In one embodiment, when building the user feature database, video data of the user is first collected using external cameras installed in the vehicle. (See also...) Figure 2 The vehicle's external cameras can include multiple cameras such as left front-view (11), right front-view (12), left side-view (13), right side-view (14), left rear-view (15), and left rear-view (16). These cameras together constitute the vehicle's external perception network, capable of covering the complete behavioral trajectory of a user approaching the vehicle and getting in and out. Furthermore, by utilizing the vehicle's own sensors, data is collected in natural scenarios where the user interacts with the vehicle, ensuring that the constructed user feature database highly matches the vehicle's actual application environment and avoiding feature biases caused by data collected in laboratory or artificial environments. Figure 2 The external camera shown is merely an example, and the present invention does not specifically limit the number and location of the external camera.

[0042] After acquiring video data, the vehicle infotainment system uses an object detection algorithm running on the vehicle's computing unit to identify video frame images in the video data, obtain at least one candidate user included in the video frame image, and determine the user tag corresponding to the candidate user. The object detection algorithm can include convolutional neural network models based on YOLO or SSD, etc., and this embodiment of the invention does not specifically limit this. Further, the object detection algorithm performs frame-by-frame or frame-by-frame sampling analysis on the video data, locates and filters image regions belonging to candidate users from complex backgrounds, and assigns a unique identifier to each identified candidate user. The tag can be an internally generated temporary or permanent identifier bound to the vehicle key, mobile phone Bluetooth MAC address, or user-initiated account ID. For example, when the vehicle infotainment system identifies a new user, it can associate the new user with the key signal currently unlocking the vehicle, generating a tag such as "User corresponding to key 1"; or, through preliminary facial feature comparison, it can temporarily associate the user with a known but unconfirmed identity, such as "Unknown user A". Next, the vehicle infotainment system extracts the face image and body image of the candidate user based on the video frame images. Specifically, the vehicle-mounted system further utilizes face detection algorithms and human body region segmentation algorithms from video frames containing candidate users to extract the face image with the highest clarity and frontal representation, as well as the human body image containing the user's full-body outline, posture, and clothing. This embodiment does not impose specific limitations on the selection of face detection algorithms and human body region segmentation algorithms.

[0043] The vehicle infotainment system analyzes facial and body images to obtain the predicted age range for candidate users, and analyzes body images to determine the predicted height for candidate users. Further, facial image analysis yields features such as facial texture, wrinkles, and skeletal contours, and these features are used to predict the candidate user's age range, such as "youth (18-35 years old)" or "middle-aged (36-60 years old)". Based on the body image, the system analyzes the proportions of the human body within the image and calculates the predicted height for candidate users. The system creates or updates a feature profile in the database for each unique "user tag," containing at least the predicted age range and predicted height. The feature profiles of several candidate users ultimately constitute the user feature library. For example, a candidate user's feature profile might be: User Tag: "Key 1 - Owner", Predicted Age Range: "30-45", Predicted Height: "175cm". This allows the vehicle infotainment system to quickly retrieve and access this feature information in subsequent services to adjust the electric pedals and vehicle suspension. The process of building the user feature library is as follows: Acquire video data captured by the vehicle's external cameras; identify video frame images in the video data to obtain at least one candidate user included in the video frame image, and determine the user tag corresponding to the candidate user; extract the face image and body image of the candidate user based on the video frame image; analyze the face image and body image to obtain the predicted age range corresponding to the candidate user, and analyze the body image to determine the predicted height data corresponding to the candidate user; construct a user feature library based on the user tag, predicted age range, and predicted height data corresponding to each candidate user.

[0044] The user feature database construction process in this embodiment makes full use of existing vehicle exterior cameras without adding extra hardware costs. By analyzing the video data collected by the cameras, the facial and body images of candidate users are extracted, and their age range and height data are accurately predicted based on this. Combined with user tags, a complete feature profile is formed, realizing the automated and non-contact collection and updating of user features, ensuring that the suspension and pedal assist can accurately adapt to the actual needs of different users.

[0045] In one embodiment of the present invention, when analyzing facial and body images to obtain the predicted age range corresponding to candidate users, multi-dimensional biometric features are extracted from the user's facial and body images, and more reliable predicted age ranges and predicted height data are calculated accordingly. Specifically, the vehicle system first extracts key feature points from the facial image using a convolutional neural network target detection algorithm. Key feature points are characterized by stable and distinguishable features on the face, such as the corners of the eyes, corners of the mouth, tip of the nose, and facial contour points. Further, through transformations such as rotation, scaling, and translation, the face is adjusted to a standardized position to eliminate errors caused by head tilting or rotation. Probability distribution modeling is introduced into the convolutional neural network to calculate the probability distribution of the target user's age based on the key feature points. Specifically, the continuous age range is discretized into several ordered, adjacent age intervals, for example, one interval for every 10 years: 0-9, 10-19, 20-29, etc., and an output node is set for each age interval in the final fully connected layer of the convolutional neural network. During the training phase, the raw output values ​​of the output nodes are transformed into corresponding probability distributions using the Softmax function. The Softmax function ensures that the sum of the probabilities for all age intervals is 1, and the probability value for each interval is between 0 and 1. For example, for an input face, the output of the convolutional neural network might be: P(20-29 years old) = 0.65, P(30-39 years old) = 0.25, P(other intervals) = 0.10. 0.65, 0.25, and 0.1 represent the probability distribution of the candidate user's age. During the inference phase, the vehicle system selects the interval with the highest probability value as the first age interval, which in the above embodiment is 20-29 years old.

[0046] In another embodiment, the vehicle-mounted system first identifies the human features of candidate users from human images, including the overall outline and posture of the body. Then, based on a deep sequence model, it extracts key joints of the candidate users from the human feature information. Key joints refer to the connection points of the major bones in the human body, such as the shoulder, elbow, wrist, hip, knee, and ankle. The vehicle-mounted system tracks these joints in multiple consecutive frames of images and constructs a skeleton sequence of candidate users based on these key joints. This skeleton sequence describes the limb movement trajectory of the user while walking. Next, the vehicle-mounted system extracts gait features of candidate users from the skeleton sequence. Gait features are characterized by periodic patterns extracted from the walking posture of candidate users, including stride length, walking speed, arm and foot swing amplitude, trunk tilt, and joint flexion angles. Gait features are correlated with age; for example, younger people's gait is usually more flexible and has a longer stride, while older people's gait may be more cautious and have a shorter stride. A second age range for candidate users is further determined based on the gait features. Specifically, the acquired video sequence is analyzed frame by frame. For each frame of human image, a pre-trained pose estimation algorithm (such as OpenPose, HRNet, or a CNN-based model) is used to identify and locate the two-dimensional or three-dimensional pixel coordinates of key joints such as the hip, knee, and ankle. By tracking the spatial position of these joints across frames, the vehicle system generates a motion trajectory that changes over time for each joint. When extracting gait features, it is necessary to identify the swing curves of joints during walking and calculate the rate of change of velocity of key limb segments during movement, such as the lower leg and thigh. The gait of older individuals often exhibits different acceleration characteristics at ground contact and during swing. The rate of change of velocity is calculated in real time and the curve of change of joint angles within a gait cycle is also calculated. The joint angle changes of younger individuals are usually larger and the curves are smoother and more powerful. Based on the motion trajectory of the joints and the curve of change of joint angles, a skeleton sequence representing a complete gait cycle is constructed. This embodiment can also perform standardized preprocessing on the skeleton data, such as normalizing the coordinates of all joints with the hip center as the origin, thereby eliminating the initial influence of absolute position and user height scale. Furthermore, data such as normalized stride length, stride speed, cadence, and double-foot support time can be obtained. Double-foot support time is characterized by the time both feet are in contact with the ground simultaneously, and this time typically increases with age. The symmetry of the movement amplitude on both sides of the candidate user's body, as well as the amplitude of trunk swaying during walking, are also analyzed. The deep sequence model is trained on a large amount of age-labeled gait data to autonomously learn the complex mapping relationship between gait features and age. This embodiment does not impose specific limitations on the selection of the deep sequence model or the training process.

[0047] Furthermore, the vehicle-mounted system calculates the predicted age range of candidate users based on the first and second age ranges. In this embodiment, the vehicle-mounted system can assign appropriate weights to the two models based on their historical accuracy for weighted decision-making; or, when the two ranges overlap, their intersection is taken as the final range; when they do not overlap, the one with higher confidence is selected, or a more conservative, wider range covering both is chosen. This invention does not impose specific limitations. For example, when a person outside the vehicle is wearing a mask, age is mainly determined based on gait characteristics, therefore the predicted age range = second age range × 80% + first age range × 20%; when a person outside the vehicle is wearing a skirt or other clothing that obscures feature points, age is mainly calculated using facial features, therefore the predicted age range = first age range × 80% + second age range × 20%; when a person outside the vehicle is wearing a skirt and a mask, the predicted age range = first age range × 80% + second age range × 20%. Specifically, the following steps are included: Key feature points are extracted from facial images. The probability distribution of candidate users' ages is calculated based on these key feature points, and the age range with the highest probability is selected as the first age range for candidate users. Human body feature information of candidate users is identified from human body images. Key joint points of candidate users are extracted from the human body feature information, and a skeleton sequence of candidate users is constructed based on the key joint points. Gait features of candidate users are extracted from the skeleton sequence, and a second age range for candidate users is determined based on the gait features. Based on the first and second age ranges, the predicted age range of the first candidate user is calculated.

[0048] This embodiment uses a dual age analysis that integrates facial features and gait features. It not only calculates the first age range based on facial key points, but also extracts gait features from the human skeleton sequence to determine the second age range. Finally, it synthesizes the prediction results, which improves the accuracy and reliability of age prediction, overcomes the errors that may exist based on a single basis, and makes the judgment of the user's age more accurate.

[0049] In one embodiment, the vehicle-mounted system identifies the body proportion data of candidate users from human images. This body proportion data characterizes the relative dimensional relationships between different parts of the candidate user's body, such as the head-to-body ratio, the ratio of leg length to height, and the ratio of shoulder width to height. Body proportions exhibit certain statistical regularities in the adult population, and compared to absolute dimensions, they are less affected by distance and angle in images, making them more stable. Subsequently, the vehicle-mounted system calculates the predicted height data of the candidate user based on the body proportion data. For example, the height prediction model in the vehicle-mounted system identifies the candidate user's head-to-body ratio as approximately 1:7.5. Combining the distribution of this ratio in the statistical population with some prior knowledge of the scene, such as the known height of the vehicle door handle as a scale reference, the system estimates the user's absolute height. This avoids the challenges of absolute size measurement caused by camera perspective distortion and uncertain user distance, achieving high-precision height estimation by utilizing the inherent biometric proportions of the human body.

[0050] Specifically, when training a height prediction model, a large-scale, highly diverse image dataset is pre-collected. This dataset needs to contain a large number of people of different heights, covering the entire range from children to adults. Each person needs to be photographed at different distances to simulate real-world scenarios where users approach vehicles at varying distances. Each image is labeled with its actual height value, which must be accurately measured to serve as the ground truth in supervised learning. For example, a convolutional neural network (CNN) can be used as the backbone network, such as ResNet, VGG, or EfficientNet. During training, images are input into the CNN. The first few layers detect low-level features of the image, such as edges and corners. Middle layers combine these low-level features to form local contours, such as the shape of arms and legs. Deeper layers integrate these local contours into deep features related to height. These deep features represent the CNN's encoding of the overall proportions of the human body and the relative positional relationships between skeletal key points. For example, convolutional neural networks can learn head-to-body ratio, limb proportions, and relative distances between key points. The head-to-body ratio represents the ratio of total height to head height; limb proportions represent the ratio of leg length (from hip to ankle) to torso length (from shoulder to hip); and relative distances between key points represent, for example, the ratio of shoulder width to height, or arm length to height. The convolutional neural network is connected to one or more fully connected layers, and the output of each fully connected layer is the predicted height value. During training, a loss function (such as Mean Absolute Error (MAE) or Mean Squared Error (MSE)) is used to quantify the difference between the model's predicted height and the actual height labeled in the image, ultimately generating a height prediction model. After training, the height prediction model is deployed to the vehicle's infotainment system. When an external camera captures an image of a candidate user, human detection and cropping are performed first, and then the image of that human region is fed into the height prediction model. Specifically, the steps include: Identify the body proportion data of candidate users from human images; calculate the predicted height data of candidate users based on the body proportion data.

[0051] This embodiment does not rely on specific reference objects or sensors. Instead, it analyzes human images to identify the user's body proportion data and directly calculates the predicted height based on this. This non-contact measurement method simplifies hardware requirements, reduces costs, and ensures that height estimation can be completed quickly and conveniently within the field of view of a conventional vehicle-mounted camera.

[0052] In step S102, the extension state of the electric pedal at the corresponding position is adjusted according to the target user's position, and the suspension height of the vehicle's suspension is adjusted synchronously according to the target feature information.

[0053] In one embodiment, for a scenario where the target user is currently in need of boarding, the vehicle system will acquire the target user's location, predicted age range, and / or predicted height data. The target user's location refers to its spatial coordinates relative to a certain door, which is used to determine the specific door that needs to be served.

[0054] Furthermore, based on the user's location, the electric pedal under the corresponding door is activated to an appropriate extension state. The extension state includes the extent of extension or the step height. For example, for children or the elderly with lower predicted height, the pedal can extend further and be closer to the ground to significantly reduce the stepping height; while for taller adults, the extension distance can be shortened to avoid unnecessary movements, or the electric pedal can remain in its default state.

[0055] At the same time, the vehicle's infotainment system continues to adjust the suspension height based on target feature information. Suspension height represents the relative distance between the vehicle body and the tires, and is adjusted through active air suspension or hydraulic suspension systems. For example, for elderly users or children, the infotainment system will instruct the suspension to lower the vehicle body, coordinating with the extension of the electric pedals to create a smooth passage with minimal height difference from the ground to the cabin, optimizing the user's convenience when getting into the vehicle.

[0056] Secondly, for scenarios where the target user is currently getting off the vehicle, the vehicle's infotainment system first checks if the target user's facial image is present in the user feature database to confirm whether the user getting off is a candidate user in the database. If so, the system triggers the extension of the corresponding electric pedal based on the target user's location and simultaneously adjusts the suspension height according to the target user's predicted age range. For candidate users in the database, their ideal exit assist posture may have been learned and solidified in the user feature database's preference settings during historical interactions, allowing direct access to these preferences. If not, the system adjusts the extension of the electric pedal based on the target user's location and predicted age range. For example, a lower pedal is provided for older temporary passengers, and the vehicle's suspension height is simultaneously adjusted based on the target user's predicted age range and / or predicted height data. In this case, lacking the user's preset preferences, the system adjusts the electric pedal and vehicle suspension based on fundamental biometrics to ensure a baseline level of safety and convenience for any user. Specifically, this includes the following steps: If the target user is currently getting into the vehicle, the extension state of the electric pedal at the corresponding position is adjusted according to the target user's location, predicted age range, and / or predicted height data, and the suspension height of the vehicle is simultaneously adjusted according to the target user's predicted age range and / or predicted height data. If the target user is currently getting out of the vehicle, the system identifies whether the target user's face image is present in the user feature database. If so, the extension state of the electric pedal at the corresponding position is adjusted according to the target user's location, and the suspension height of the vehicle is simultaneously adjusted according to the target user's predicted age range. If not, the extension state of the electric pedal at the corresponding position is adjusted according to the target user's location and predicted age range, and the suspension height of the vehicle is simultaneously adjusted according to the target user's predicted age range and / or predicted height data.

[0057] This embodiment distinguishes between two scenarios: getting into and getting out of the vehicle, and implements targeted adjustment strategies. Especially when getting out of the vehicle, it prioritizes querying the feature database to confirm the user's identity; if the user is registered, it directly uses their pre-stored features for precise adjustments; if the user is unfamiliar, it provides adaptive assistance based on real-time analysis of age and height data. This differentiated intelligent decision-making ensures that whether the passenger is a familiar family member or a temporary passenger, they receive the most suitable pedal and suspension assistance, improving practicality in real-world driving scenarios.

[0058] In one embodiment, when adjusting the vehicle suspension based on the predicted age range and predicted height data of the target user, it is necessary to obtain a three-dimensional mapping table pre-stored in the vehicle's infotainment system. This three-dimensional mapping table is a database or lookup table pre-calibrated through experiments, simulations, and user studies. The three dimensions of the three-dimensional mapping table are: age range (e.g., "children: 0-12 years old", "teenagers: 13-18 years old", "youth: 19-35 years old", "middle-aged: 36-60 years old", "elderly: 61 years old and above"), height data (segments in centimeters, such as "below 150cm", "150-165cm", "165-180cm", "above 180cm"), and the corresponding suspension height reference value. The height reference value represents a theoretical target height that the vehicle suspension should achieve for an ideal user with a specific combination of age and height, to achieve the best getting in and out of the vehicle experience. For example, the three-dimensional mapping table may define a correspondence such that when the user characteristics are {age range: elderly, height data: 160-165cm}, the corresponding height reference value is "-40mm" (that is, 40mm lower than the standard posture).

[0059] After obtaining the predicted age range and predicted height data of the target user, the vehicle system queries the 3D mapping table for the corresponding height benchmark value for the predicted age range and / or predicted height data. Using the currently identified target user's predicted age range and predicted height data as input values, the system performs a matching query in the 3D mapping table and outputs the corresponding height benchmark value. It is important to emphasize that this invention employs an "AND / OR" logic when querying the height benchmark value based on the predicted age range and predicted height data. This means it can support using only age, only height, or both as query conditions to adapt to scenarios with varying data completeness, ensuring the robustness of the method.

[0060] Furthermore, the vehicle system calculates the difference between the baseline height value and the current suspension height value. This difference is used to determine the magnitude and direction of the suspension adjustment. The current suspension height value can be measured in real time by height sensors mounted on the suspension struts. Simultaneously, the vehicle system acquires the vehicle body attitude parameters corresponding to the target user's door position. These parameters characterize the vehicle's attitude changes when stationary on a non-level surface, such as roll and pitch angles. When a user prepares to get in or out of the vehicle from one door, if the vehicle is parked on a slope, the entire body will tilt. Simply adjusting all four suspension wheels to the baseline value might not provide a truly level platform for getting in and out of the target door. Therefore, the target suspension height needs to be compensated and corrected based on the specific vehicle tilt. Finally, the vehicle system calculates the target suspension height based on the difference and vehicle attitude parameters to ensure that, after adjustment, the vehicle body in the target door area reaches an ideal level and suitable height. This creates a convenient entry and exit contact point for the target user regardless of the surface the vehicle is parked on. The specific steps include the following: Obtain a 3D mapping table, which includes the mapping relationship between age ranges, height data, and suspension height reference values; query the predicted age range and / or the height reference value corresponding to the predicted height data from the 3D mapping table; calculate the difference between the height reference value and the current height value of the suspension, and obtain the vehicle body attitude parameters corresponding to the door position of the target user; calculate the target suspension height using the difference and the vehicle body attitude parameters.

[0061] This embodiment introduces a three-dimensional mapping table to quantify the predicted user characteristics into specific suspension height benchmark values. It further combines the current suspension height with real-time vehicle posture parameters for comprehensive calculation, and finally obtains the target suspension height that is suitable for the current scenario. This ensures that the adjustment of the suspension height not only meets the user's physical needs, but also compensates for changes in vehicle posture caused by load and road conditions, thereby providing stable and convenient support for users getting in and out of the vehicle under any circumstances.

[0062] This embodiment also provides a control device for an electric pedal, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0063] This embodiment provides a control device for an electric pedal, such as... Figure 3 As shown, it includes: The information determination module 301 is used to determine the target user who needs to get on or off the vehicle and the target user's target feature information when the vehicle door is opened.

[0064] The pedal adjustment module 302 is used to adjust the extension state of the electric pedal at the corresponding position according to the position of the target user, and to synchronously adjust the suspension height of the vehicle suspension according to the target feature information.

[0065] Information determination module 301 includes: The user acquisition unit is configured to, if the door opening method is external opening, determine at least one first user outside the vehicle and within a preset range of the opened door; and / or, if the door opening method is internal opening, determine a second user inside the vehicle and within a preset range of the opened door.

[0066] The user determination unit is used to determine the first user and / or the second user as target users with boarding and alighting needs.

[0067] The information acquisition unit is used to obtain target feature information corresponding to the target user from the user feature database. The user feature database includes user feature information of multiple candidate users. The user feature information is collected by the candidate user outside the vehicle or before getting into the vehicle by an external camera.

[0068] The user feature database is constructed through the following sub-units: The video acquisition subunit is used to acquire video data captured by the vehicle's external cameras.

[0069] The label determination subunit is used to identify video frame images in video data, obtain at least one candidate user included in the video frame image, and determine the user label corresponding to the candidate user.

[0070] The image extraction subunit is used to extract face images and body images of candidate users based on video frame images.

[0071] The data prediction subunit is used to analyze facial and human images to obtain the predicted age range of candidate users, and to analyze human images to determine the predicted height data of candidate users.

[0072] The feature library construction subunit is used to build a user feature library based on the user tags, predicted age range, and predicted height data corresponding to each candidate user.

[0073] The data prediction subunit includes: The probability calculation subunit is used to extract key feature points from the face image, calculate the probability distribution of the age of the candidate user based on the key feature points, and select the age range with the highest probability as the first age range of the candidate user.

[0074] The feature extraction subunit is used to identify human feature information of candidate users from human images.

[0075] The skeleton sequence construction subunit is used to extract key points of candidate users from human feature information and construct the skeleton sequence of candidate users based on the key points.

[0076] The gait extraction subunit is used to extract gait features of candidate users from the skeleton sequence and determine the second age range of candidate users based on the gait features.

[0077] The age prediction subunit is used to calculate the predicted age range of the first candidate user based on the first age range and the second age range.

[0078] The human body proportion acquisition subunit is used to identify the human body proportion data of candidate users from human body images.

[0079] The height prediction subunit is used to calculate the predicted height data of candidate users based on human proportion data.

[0080] Pedal adjustment module 302 includes: The first adjustment unit is used to adjust the extension state of the electric pedal at the corresponding position according to the target user's location, predicted age range, and / or predicted height data if the target user is currently in need of getting into the vehicle, and to simultaneously adjust the suspension height of the vehicle's suspension according to the target user's predicted age range and / or predicted height data.

[0081] The second adjustment unit is used to identify whether the user's facial image is in the user feature database if the target user is currently getting off the vehicle; if so, it adjusts the extension state of the electric pedal at the corresponding position according to the target user's position, and simultaneously adjusts the suspension height of the vehicle's suspension according to the target user's predicted age range; if not, it adjusts the extension state of the electric pedal at the corresponding position according to the target user's position and predicted age range, and simultaneously adjusts the suspension height of the vehicle's suspension according to the target user's predicted age range and / or predicted height data.

[0082] The first adjustment unit and the second adjustment unit also include: The mapping table acquisition sub-unit is used to acquire a three-dimensional mapping table, which includes the mapping relationship between age ranges, height data, and suspension height reference values.

[0083] The data query subunit is used to query the predicted age range and / or the height baseline value corresponding to the predicted height data from the three-dimensional mapping table.

[0084] The difference calculation subunit is used to calculate the difference between the height reference value and the current height value of the suspension, and to obtain the vehicle body attitude parameters corresponding to the door position of the target user.

[0085] The suspension height calculation subunit is used to calculate the target suspension height using the difference and vehicle attitude parameters.

[0086] In this embodiment, the control device for the electric pedal is presented in the form of a functional unit. Here, a unit refers to an ASIC circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0087] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0088] This invention also provides a computer device having the above-described features. Figure 3 The control device for the electric pedal shown.

[0089] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 4 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 4 Take a processor 10 as an example.

[0090] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0091] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0092] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device as shown by a landing page for an app. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0093] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0094] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0095] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0096] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A control method for an electric pedal, characterized in that, The method includes: When the vehicle door is opened, identify the target user who needs to get on or off the vehicle and the target user's target feature information; Adjust the extension state of the electric pedal at the corresponding position according to the location of the target user, and synchronously adjust the suspension height of the vehicle's suspension according to the target feature information.

2. The method according to claim 1, characterized in that, The process of determining the target users who currently have pick-up and drop-off needs, and the target user's target feature information, includes: If the door is opened externally, then at least one first user is identified outside the vehicle and within a preset range of the opened door; and / or, if the door is opened internally, then a second user is identified inside the vehicle and within a preset range of the opened door. The first user and / or the second user are identified as target users with boarding and alighting needs; The target feature information corresponding to the target user is obtained from the user feature database, wherein the user feature database includes user feature information of multiple candidate users, and the user feature information is collected by the candidate user outside the vehicle or before getting into the vehicle by an external camera.

3. The method according to claim 2, characterized in that, The process of constructing the user feature database includes: Acquire video data captured by the external cameras configured in the vehicle; Identify video frame images in the video data, obtain at least one candidate user included in the video frame image, and determine the user tag corresponding to the candidate user; Extract the face and body images of the candidate users based on the video frame images; The predicted age range of the candidate user is obtained by analyzing the face image and the body image, and the predicted height data of the candidate user is determined by analyzing the body image. A user feature library is constructed based on the user tags corresponding to each candidate user, the predicted age range, and the predicted height data.

4. The method according to claim 3, characterized in that, The step of analyzing the face image and the body image to obtain the predicted age range corresponding to the candidate user includes: Extract key feature points from the face image, calculate the probability distribution of the candidate user's age based on the key feature points, and select the age range with the highest probability as the first age range of the candidate user. Identify the human body feature information of the candidate user from the human body image; Extract key points of the candidate user from the human feature information, and construct the skeleton sequence of the candidate user based on the key points; Gait features of the candidate users are extracted from the skeleton sequence, and a second age range of the candidate users is determined based on the gait features; Based on the first age range and the second age range, the predicted age range of the candidate user is calculated.

5. The method according to claim 3, characterized in that, The step of analyzing the human image to determine the predicted height data corresponding to the candidate user includes: Identify the body proportion data of the candidate user from the human body image; The predicted height of the candidate user is calculated based on the human body proportion data.

6. The method according to claim 5, characterized in that, The step of adjusting the extension state of the electric pedal at the corresponding position according to the position of the target user, and synchronously adjusting the suspension height of the vehicle suspension according to the target feature information, includes: If the target user is currently in need of getting into the vehicle, the extension state of the electric pedal at the corresponding position is adjusted according to the target user's location, predicted age range, and / or predicted height data, and the suspension height of the vehicle's suspension is adjusted synchronously according to the target user's predicted age range and / or predicted height data. If the target user is currently getting off the vehicle, then it is determined whether the user's facial image is contained in the user feature database; if so, the extension state of the electric pedal at the corresponding position is adjusted according to the target user's location, and the suspension height of the vehicle is simultaneously adjusted according to the predicted age range corresponding to the target user; if not, the extension state of the electric pedal at the corresponding position is adjusted according to the target user's location and predicted age range, and the suspension height of the vehicle is simultaneously adjusted according to the predicted age range corresponding to the target user and / or predicted height data.

7. The method according to claim 6, characterized in that, The step of synchronously adjusting the suspension height of the vehicle based on the predicted age range and / or predicted height data corresponding to the target user includes: Obtain a three-dimensional mapping table, wherein the three-dimensional mapping table includes the mapping relationship between age range, height data and suspension height reference value; Query the predicted age range and / or the height benchmark value corresponding to the predicted height data from the three-dimensional mapping table; Calculate the difference between the height reference value and the current height value of the suspension, and obtain the vehicle body posture parameters corresponding to the door position where the target user is located; The target suspension height is calculated using the difference and the vehicle body attitude parameters.

8. A control device for an electric pedal, characterized in that, The device includes: The information determination module is used to determine the target user who needs to get on or off the vehicle and the target user's target feature information when the vehicle door is opened. The pedal adjustment module is used to adjust the extension state of the electric pedal at the corresponding position according to the position of the target user, and to synchronously adjust the suspension height of the vehicle suspension according to the target feature information.

9. A computer device, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the method of any one of claims 1 to 7.

Images