Vehicle door control method and apparatus based on millimeter-wave radar, and device

By using millimeter-wave radar-based pedestrian detection and trajectory tracking, and by recognizing user intent using angle and distance change rates, the problem of door control accuracy caused by sparse point cloud data has been solved, achieving seamless and precise door opening, thus improving user experience and safety.

WO2026232854A1PCT designated stage Publication Date: 2026-11-12CHERY AUTOMOBILE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2025/148185
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-07-15
Filing Date
2025-12-31
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Existing door control methods based on millimeter-wave radar suffer from low motion recognition accuracy due to sparse point cloud data, affecting response sensitivity, especially when lighting conditions change.

Method used

By using millimeter-wave radar for pedestrian detection and trajectory tracking, the system identifies pedestrian movement direction and speed changes by utilizing angle and distance change rates. Combined with preset conditions, the system controls door opening, including setting time windows and motion trajectory determination, to ensure that the doors open accurately when the user approaches.

Benefits of technology

It enables contactless and seamless door opening, improving the user experience, especially in situations where you are holding items or in harsh environments, avoiding key operation, and also improving control accuracy and security while protecting user privacy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2025148185_12112026_PF_FP_ABST
    Figure CN2025148185_12112026_PF_FP_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of the intelligent control of electric vehicles. Provided are a vehicle door control method and apparatus based on a millimeter-wave radar, and a device. The method comprises: in response to a detection signal indicating that a key enters an effective sensing range, triggering a millimeter-wave radar to start; receiving radar echo signals in real time, and performing pedestrian detection within a detection area defined in front of each vehicle door; if a pedestrian is detected within one or more detection areas, tracking a motion trajectory of the pedestrian; acquiring a change in the angle between the motion trajectory and a vertical direction of a corresponding vehicle door, and a rate of change of the distance between the pedestrian and the vehicle door; and when the change in the angle and the rate of change of the distance satisfy a set condition, controlling the vehicle door to open. The present invention can effectively solve the problems of inconvenient manual operation of vehicle doors and a low degree of intelligence in the prior art.
Need to check novelty before this filing date? Find Prior Art

Description

A vehicle door control method, device, and equipment based on millimeter-wave radar

[0001] Cross-references to related applications

[0002] This invention claims priority to Chinese Patent Application No. 202510974482.9, filed on July 15, 2025, entitled "A Door Control Method, Apparatus and Device Based on Millimeter-Wave Radar", the entire contents of which are incorporated herein by reference and constitute a part of this invention for all purposes. Technical Field

[0003] This application relates to the field of automotive safety technology, and more specifically to a door control method, device, and equipment based on millimeter-wave radar. Background Technology

[0004] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0005] Currently, keyless or manual automatic door opening controls mostly employ standardized actions. They rely on video or millimeter-wave radar to identify the user's movements in front of the vehicle, comparing them to standard actions in a database, and then using this identification method to control door opening. Although millimeter-wave radar is unaffected by lighting conditions, the point cloud data it generates is highly sparsity. This sparsity can lead to incomplete feature extraction, affecting the accuracy of action recognition and consequently, the sensitivity of the response. Summary of the Invention

[0006] In view of this, the present invention provides a vehicle door control method, device and equipment based on millimeter-wave radar to achieve non-contact and imperceptible vehicle door opening.

[0007] The first aspect of the present invention provides a vehicle door control method based on millimeter-wave radar, comprising the following steps:

[0008] The millimeter-wave radar is activated in response to a detection signal indicating that the key has entered the effective sensing range.

[0009] It receives radar echo signals in real time and performs pedestrian detection within the designated detection area in front of each vehicle door;

[0010] If a pedestrian is detected in one or more detection areas, the pedestrian's motion trajectory is tracked; the angle change between the motion trajectory and the corresponding door's vertical direction, as well as the rate of change of the distance between the pedestrian and the door, are obtained;

[0011] When the angle change and distance change rate meet the set conditions, the door is controlled to open.

[0012] In some embodiments, the set conditions are: the angle between the motion trajectory and the vertical direction of the door continuously decreases, the distance change rate is negative and within a set range.

[0013] In some embodiments, when tracking the pedestrian's movement trajectory, a time window is set. If the rate of change of the movement trajectory angle exceeds a set threshold within the time window, and the angle between the movement direction and the perpendicular direction of the car door exceeds a set threshold, the pedestrian will no longer be tracked.

[0014] In some embodiments, the detection area is rectangular, the width of which is set according to the width of an adult's body, and the length of which is set according to the adult's walking speed and the opening degree of the car door.

[0015] In some embodiments, when tracking the movement trajectory of a pedestrian, the pedestrian's walking speed is also detected in real time. If the speed is lower than a set speed threshold, the time required for the pedestrian to reach the door's movement space is predicted. Based on this time and the time required for the door to open, the door opening time and opening speed are determined.

[0016] A second aspect of the present invention provides a vehicle door control method based on millimeter-wave radar, comprising the following steps:

[0017] The millimeter-wave radar is activated in response to a detection signal indicating that the key has entered the effective sensing range.

[0018] It receives radar echo signals in real time and performs pedestrian detection within the designated detection area in front of each vehicle door;

[0019] If a pedestrian is detected in one or more detection areas, the specified movement parts are identified;

[0020] The movement trajectory of the moving part is detected, and when the movement trajectory meets the preset conditions, the corresponding door is controlled to open.

[0021] In some embodiments, a preset sequence of identifying features for a given action and preset conditions to be met are defined. The sequence of identifying features includes multiple consecutive positions that the action part needs to reach, and the conditions to be met include one or more of the following: action height condition, action speed condition, action time condition, and action stability condition. When the detected action trajectory meets the set sequence of identifying features and satisfies the preset conditions, the motion trajectory is determined to be valid, and the corresponding door is controlled to open.

[0022] A third aspect of the present invention provides a vehicle door control method based on millimeter-wave radar, characterized by comprising the following steps:

[0023] The millimeter-wave radar is activated in response to a detection signal indicating that the key has entered the effective sensing range.

[0024] It receives radar echo signals in real time and performs pedestrian detection within the designated detection area in front of each vehicle door;

[0025] If a pedestrian is detected in one or more detection areas, the pedestrian's movement trajectory is tracked;

[0026] Obtain the angle change between the motion trajectory and the corresponding vertical direction of the car door, as well as the rate of change of the distance between the pedestrian and the car door;

[0027] When the angle change and the rate of change of distance meet the set conditions, it is determined that the pedestrian has the intention to open the door;

[0028] The system identifies the designated movement parts of the pedestrian; detects the movement trajectory of the movement parts; and controls the corresponding vehicle door to open when the movement trajectory meets preset conditions.

[0029] In some embodiments, the set conditions are: the angle between the motion trajectory and the vertical direction of the door continuously decreases, the distance change rate is negative and within a set range.

[0030] In some embodiments, when tracking the pedestrian's movement trajectory, a time window is set. If the rate of change of the movement trajectory angle exceeds a set threshold within the time window, and the angle between the movement direction and the perpendicular direction of the car door exceeds a set threshold, the pedestrian will no longer be tracked.

[0031] In some embodiments, the detection area is rectangular, the width of which is set according to the width of an adult's body, and the length of which is set according to the adult's walking speed and the opening degree of the car door.

[0032] In some embodiments, when tracking the movement trajectory of a pedestrian, the pedestrian's walking speed is also detected in real time. If the speed is lower than a set speed threshold, the time required for the pedestrian to reach the door's movement space is predicted. Based on this time and the time required for the door to open, the door opening time and opening speed are determined.

[0033] In some embodiments, a preset sequence of identifying features for a given action and preset conditions to be met are defined. The sequence of identifying features includes multiple consecutive positions that the action part needs to reach, and the conditions to be met include one or more of the following: action height condition, action speed condition, action time condition, and action stability condition. When the detected action trajectory meets the set sequence of identifying features and satisfies the preset conditions, the motion trajectory is determined to be valid, and the corresponding door is controlled to open.

[0034] A fourth aspect of the present invention provides a vehicle door control device based on millimeter-wave radar, comprising:

[0035] The function trigger module is configured to trigger the millimeter-wave radar to start in response to a detection signal that the key has entered the effective sensing range;

[0036] The pedestrian detection module is configured to receive radar echo signals in real time and detect pedestrians around the vehicle.

[0037] The trajectory tracking module is configured to track the movement trajectory of a pedestrian when a pedestrian is detected. If the pedestrian is approaching a vehicle, the module determines the door the pedestrian is facing and defines a detection area in front of the door.

[0038] The door control module is configured to continuously track the movement trajectory of a pedestrian after detecting that the pedestrian has entered the detection area, obtain the angle change between the movement trajectory and the vertical direction of the door, and the rate of change of the distance between the pedestrian and the door. When the angle change and the rate of change of the distance meet the set conditions, the door is controlled to open.

[0039] A fifth aspect of the present invention provides an electronic device including a processor and a memory, wherein the memory stores computer instructions that, when executed by the processor, cause the electronic device to perform the method described thereon.

[0040] A sixth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method.

[0041] A seventh aspect of the present invention provides a computer program product comprising a computer program that, when executed by a processor, implements the method described herein.

[0042] One or more of the above technical solutions utilize millimeter-wave radar for pedestrian detection and trajectory tracking. By analyzing changes in the angle perpendicular to the car door and the rate of change in distance from the door, they identify changes in the pedestrian's direction of movement and speed, thus reflecting the pedestrian's intention to open the door. This achieves contactless and imperceptible door opening; the user only needs to walk normally to the door to control its opening. This is especially beneficial when holding items or in challenging environments, eliminating the need for key operation and significantly improving the user experience. Furthermore, the sparse point cloud data acquired in real-time allows for the calculation of pedestrian movement direction and speed changes, enhancing control accuracy. Attached Figure Description

[0043] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0044] Figure 1 illustrates an example of an implementation environment for an embodiment of the present invention;

[0045] Figure 2 shows a schematic diagram of the arrangement of the millimeter-wave radar in an embodiment of the present invention;

[0046] Figure 3 shows a flowchart of the first vehicle door control method based on millimeter-wave radar provided by an embodiment of the present invention;

[0047] Figure 4 shows a schematic diagram of the detection area in the first door control method provided by an embodiment of the present invention;

[0048] Figure 5 shows a flowchart of the second door control method based on millimeter-wave radar provided by an embodiment of the present invention;

[0049] Figure 6 shows a schematic diagram of the detection area for the foot-stepping action as a prescribed action in the second door control method provided by the embodiment of the present invention.

[0050] Figure 7 shows a schematic diagram of the detection area for the foot sweeping motion as a prescribed motion in the second door control method provided by the embodiment of the present invention;

[0051] Figure 8 shows a schematic diagram of the detection area for the foot-stepping action as a prescribed action in the third door control method provided by the embodiment of the present invention. Detailed Implementation

[0052] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While some embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this application. It should be understood that the drawings and embodiments of this application are for illustrative purposes only and are not intended to limit the scope of protection of this application.

[0053] In the description of the embodiments of this application, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on".

[0054] The term "vehicle door" as used in the embodiments of this application can refer to a hood, a tailgate, a rotating side door, a gull-wing door, a sliding door, etc., and is not specifically limited to these.

[0055] Figure 1 illustrates an example of an implementation environment for one or more embodiments of the present invention. This implementation environment includes a vehicle controller and a door control system connected to the vehicle controller. The door control system is connected to a millimeter-wave radar control module and a door actuator. When the vehicle network is in sleep mode, the millimeter-wave radar is off. When a legitimate smart key or other authorized device enters the effective sensing range around the vehicle, the vehicle controller receives a valid signal from the key sensing module and sends a start command to the millimeter-wave radar module via the vehicle network. The millimeter-wave radar then enters working mode and begins transmitting and receiving signals. The vehicle controller receives radar echo signals in real time, analyzes the trajectory, speed, and direction of the target within the detection area, and determines whether to open the door. After generating a door opening command, it sends it to the door actuator via the door control system. Upon receiving the command, the door actuator drives the motor to start working, automatically opening the door. Simultaneously, the door control system also sends signals to control modules such as speakers and lights to trigger preset welcome programs, such as playing soft music or illuminating interior lights to create a warm and comfortable atmosphere.

[0056] Figure 2 shows a schematic diagram of the arrangement of millimeter-wave radars in one or more embodiments of the present invention. Millimeter-wave radars are respectively arranged on the inner side 101 of the front bumper, the inner side 102 of the rear bumper, the inner side of the front door trim panel or the front door sill beam 103 (symmetrically arranged on the left front and right front), and the inner side of the rear door trim panel or the rear door sill beam 104 (symmetrically arranged on the left rear and right rear), and are used to control the opening of the hood, the tailgate, the front door, and the rear door.

[0057] As described in the background section, manually operating the key or door handle can lead to a poor user experience, especially if the user is carrying items or is in a hurry to get into the vehicle in the rain. Furthermore, if identity authentication is based on prescribed actions, the highly sparse point cloud data generated by millimeter-wave radar can result in incomplete feature extraction, affecting the accuracy of action recognition. To address these issues, this invention provides identity authentication based on action trajectories. When the action trajectory meets set requirements, the vehicle door is opened. The action trajectory can be a straight line approaching the door, or a prescribed gesture or foot movement. Judgment indicators are set for the action trajectory; the action trajectory parameters required for these indicators can be accurately calculated based on the sparse point cloud, thus solving the accuracy problem caused by the sparse point cloud. The different action trajectories mentioned above are applicable to different scenarios: Scenario 1: When the vehicle parking area is relatively open, the driver or passenger does not need to detour to get on the vehicle, such as temporary parking to pick up people, taxi or ride-sharing services, or business travel receptions, the vehicle is parked in a spacious area such as the entrance of a building or factory, making it convenient for passengers to get on the vehicle; Scenario 2: There are no restrictions on the vehicle parking area, parking lots and underground garages are both acceptable, which is suitable for daily driving scenarios.

[0058] Figure 3 shows a flowchart of a vehicle door control method based on millimeter-wave radar provided by one or more embodiments of the present invention, applicable to scenario one. The method is applied to a vehicle controller and includes the following steps:

[0059] S101. In response to the detection signal that the key has entered the effective sensing range, the millimeter-wave radar is triggered to start;

[0060] When a smart car key or other authorized device enters the effective sensing range of the vehicle (typically less than or equal to 5 meters), the vehicle controller sends a valid key sensing signal to the door control module. Upon receiving the valid signal from the key sensing module, the door control module sends a detection request signal to the millimeter-wave radar module, activating the millimeter-wave radar to emit millimeter-wave signals. This design ensures that the radar only activates when a user is preparing to approach the vehicle, thereby reducing unnecessary energy consumption. Simultaneously, the key sensing signal also serves as an identification function, enhancing security.

[0061] S102. Receive radar echo signals in real time and perform pedestrian detection within the designated detection area in front of each vehicle door;

[0062] Once activated, the millimeter-wave radar module begins emitting high-frequency radio waves (millimeter waves). These waves reflect off obstacles, creating echo signals. The radar system processes these echo signals in real time, analyzing frequency, phase, and amplitude changes to determine the presence and characteristics of objects within a defined detection area in front of the car door. Millimeter-wave radar can identify the size and shape of target objects to some extent based on the intensity and distribution of signal reflection. Other targets, such as electric vehicles and bicycles, differ significantly in size and shape from pedestrians walking towards the car door. Machine learning or deep learning algorithms are used to train the radar data for pedestrian recognition.

[0063] S103. If a pedestrian is detected in one or more detection areas, the movement trajectory of the pedestrian is tracked.

[0064] When a pedestrian is detected within the detection area, advanced tracking algorithms (such as Kalman filtering) are used to continuously track the target's movement. By continuously tracking the pedestrian's trajectory, the system can more accurately determine their behavioral intentions, providing a basis for subsequent operational decisions.

[0065] S104. Obtain the angle change between the motion trajectory and the corresponding vertical direction of the door, as well as the rate of change of the distance between the pedestrian and the door;

[0066] The angle change of a pedestrian's trajectory relative to the vehicle door, and the change in distance between the pedestrian and the door per unit time, are two crucial parameters for determining whether a pedestrian is moving towards the door. For example, if the angle change approaches zero degrees or decreases slightly, while the rate of change in distance remains negative (i.e., the pedestrian is gradually approaching the door), it indicates that the pedestrian is likely attempting to enter the vehicle. This refined analysis can effectively distinguish between a pedestrian's true intentions and accidental passing by. Furthermore, the sparse point cloud acquired in real-time allows for the calculation of the pedestrian's direction of movement and speed changes, improving control accuracy.

[0067] S105. When the angle change and distance change rate meet the set conditions, control the door to open.

[0068] Once the detected angle change and distance change rate meet preset conditions (such as the angle difference between the pedestrian's trajectory and the vertical direction of the door being less than a preset threshold, and the pedestrian's approach speed being within a certain range), the system will recognize this as a valid door opening request. Subsequently, the door control module will receive the door opening signal from the millimeter-wave radar, and then automatically open the door via the drive motor or actuator. Simultaneously, the system will also consider whether multiple people are approaching simultaneously, prioritizing the response to the legitimate user closest to the door to prevent confusion and ensure safety.

[0069] The aforementioned method utilizes millimeter-wave radar for pedestrian detection and trajectory tracking. By analyzing changes in the angle perpendicular to the car door and the rate of change in distance from the door, it identifies changes in the pedestrian's direction of movement and speed, thus reflecting the pedestrian's intention to open the door. This achieves contactless and seamless door opening; users simply need to walk to the door to control its opening, especially useful when holding items or in challenging environments, eliminating the need for key operation and significantly improving the user experience. Based on this method, any user can seamlessly enter the vehicle through any door as long as the key is nearby. Furthermore, compared to visual recognition solutions such as cameras, millimeter-wave radar focuses more on detecting dynamic information of moving objects rather than static features. This means it does not capture personal privacy information such as facial features, thus better protecting user privacy while achieving intelligent door control.

[0070] In step S103, in order to enable the opening of any car door, a detection area is set for each car door. The detection area can be defined as any shape, as long as it is within the radar detection envelope. The detection area can be configured by software within the millimeter-wave radar envelope. In order to make the radar detection more effective, the center normal of the detection area width coincides with the radar normal.

[0071] The detection area should not be set too large, as an excessively large area may allow multiple moving targets to enter, such as two people entering the detection area simultaneously, affecting the accuracy of motion trajectory tracking and increasing the risk of false triggering. Therefore, to improve the accuracy of seamless door opening, some embodiments use a rectangular detection area. The width of the rectangle is set according to the width of an adult's body, and the length is set according to the adult's walking speed and the door opening degree. By limiting the size of the detection area, the millimeter-wave radar can effectively capture the user's approach to the door, improving recognition accuracy.

[0072] For example, as shown in Figure 4, a detection area 202 is set in front of the radar 201. The detection area is a rectangle with a width of L1 and a length of L2, preferably L1 = 800mm and L2 = 2000mm. Setting the width to 800mm can fully cover the body width of a normal adult, while leaving a certain margin to accommodate differences in height and body shape and the possibility of carrying items. This means that whether the user approaches from the front or passes by side, they can be effectively detected by the system, thereby ensuring that the triggering of the door opening action will not fail due to body deviation. Setting the length to 2000mm is based on the understanding of adult walking speed and the deceleration habit when approaching the car door. Under normal circumstances, people naturally slow down as they approach their destination. The average adult walking speed is 4-6 km / h (1.11-1.67 m / s). As they approach the car door, they slow down, moving approximately 1 meter in about 1 second. Therefore, the system allows sufficient "observation" distance (1000mm) and time (about 1 second) to fully recognize the user's approaching intention before the door begins to open. This ensures the door opening action is neither delayed nor premature, thus improving overall smoothness. In the final 1000mm distance, the door opens synchronously, matching the user's movement rhythm. By the time the user reaches the door, the opening angle is already sufficient for easy boarding, eliminating extra waiting time and providing a seamless boarding experience.

[0073] Understandably, although a rectangular detection area is described here, its shape and size can be flexibly adjusted depending on different vehicle models and usage scenarios. For example, special designs such as gull-wing doors or sliding doors may require a larger detection area to cover approach paths from more angles, or the detection area may be reduced in places with limited space to decrease false triggers.

[0074] In step S103, the pedestrian's movement trajectory is tracked using tracking algorithms such as Kalman filtering. By predicting the pedestrian's position at the next moment and updating the tracking based on newly detected data, stable tracking of the pedestrian is ensured, avoiding target loss and misjudgment. Trajectory prediction allows for the prediction of the pedestrian's future movement trajectory. Analysis of the relationship between the movement trajectory and the door's position increases the credibility of the target moving towards the door.

[0075] In step S104, the set conditions for angle change and distance change rate are: the angle between the movement trajectory and the perpendicular direction of the door continuously decreases, and the distance change rate is negative and within the set range. The angle change rate condition focuses on determining whether the user is adjusting their direction to face the door, while the distance change rate condition emphasizes that the user is approaching the door at an appropriate speed. Through dual verification of directionality and speed, the system can more accurately adapt to the user's actual behavior. As shown in Figure 4, assuming the system identifies a moving target entering from side a, and continuously tracks its position after entry, if it meets the conditions of approaching the door and decreasing in speed within 500ms without touching the left and right side boundaries (sides b / d), it is determined to be a valid trigger, and the door is controlled via CAN signal.

[0076] Monitoring the rate of change of angle helps determine whether the target user is adjusting their orientation to eventually face the door. If the relative angle between the target object and the radar continuously decreases (i.e., the angle outward from the radar normal decreases), it indicates that the user is turning towards the door. Continuous monitoring of the rate of change of angle reduces errors in door opening decisions caused by momentary angle misreading. In addition, it can exclude objects that are moving but not facing the door, thereby reducing false triggering.

[0077] The distance change rate quantifies the speed at which the distance between two points changes. By setting a reasonable distance change rate threshold, the system can quickly identify that a target user is approaching the car door at an appropriate speed. This means that once the user enters the detection range and begins to decelerate, the system can react quickly and prepare to open the door. Combined with the design of the detection area size, this improves the user experience of seamless door opening.

[0078] When a pedestrian crosses the road, their trajectory typically changes perpendicular to or at a large angle to the direction of the car door. Based on this, by analyzing the rate of change of the target object's angle and the rate of change of its velocity direction, if the rate of change of angle is large within a short period and the angle between the velocity direction and the direction of the car door exceeds a certain threshold, it is determined that a pedestrian is crossing the road. Specifically, when tracking the pedestrian's trajectory, a time window is set. If, within this time window, the rate of change of the trajectory's angle exceeds a set threshold, and the angle between the movement direction and the perpendicular direction of the car door exceeds a set threshold, the pedestrian is no longer tracked. Technically, this embodiment effectively filters out moving targets that do not intend to open the door by setting a time window and an angle change rate threshold, especially distinguishing between pedestrian crossings and other actions not intended to open the door.

[0079] To ensure that the car door is already open to the appropriate degree before the user arrives, meaning that the user's walking and the door opening actions are synchronized without interference, the timing and speed of the door opening need to be determined based on the user's approach speed. In some embodiments, the space and time required for the door to open to a set degree are determined and denoted as the door action space and the opening time. In step S103, when tracking the pedestrian's motion trajectory, the pedestrian's walking speed is also detected in real time. If the speed is lower than a set speed threshold, the time required for the pedestrian to reach the door action space is predicted. This time can be understood as the time it takes for the car to complete the opening action, allowing the door to open to the appropriate degree. Based on this time and the door opening time, the door opening timing and speed are determined. Furthermore, the pedestrian's speed is monitored in real time before controlling the door opening; if the speed exceeds a set speed threshold, the door opening is not controlled.

[0080] The speed threshold is set because, under normal circumstances, even if a user approaches the car door quickly due to a hurry to get in, they will inevitably decelerate after entering the detection area before the door opens and they get in. Therefore, if a user has already entered the detection area and their speed continues to exceed the set threshold, it is assumed that they do not intend to get in the car. Furthermore, to ensure that the car door is already open to the appropriate degree when the user arrives, based on the above method, the faster the user approaches, the faster the door opening speed needs to be increased. Considering the safety of the car door's structural components and the surrounding environment, the door opening speed also needs to be limited.

[0081] Figure 5 shows a flowchart of a second vehicle door control method based on millimeter-wave radar provided by one or more embodiments of the present invention. Unlike the embodiments described above, this method performs authentication based on motion trajectories and is applicable to scenario one or scenario two. The method is applied to a vehicle controller and specifically includes the following steps:

[0082] S201: In response to the detection signal that the key has entered the effective sensing range, the millimeter-wave radar is triggered to start;

[0083] This step describes the initial triggering conditions of the system, namely, when a legitimate key enters the preset effective sensing range around the vehicle, the millimeter-wave radar module is activated and begins to prepare for environmental monitoring.

[0084] S202: Receives radar echo signals in real time and performs pedestrian detection within the designated detection area in front of each vehicle door;

[0085] Once activated, the radar will continuously transmit and receive electromagnetic waves. By analyzing the echo signals, it will be able to monitor and identify pedestrians or other objects within the preset detection area in real time.

[0086] S203: If a pedestrian is detected in one or more detection areas, identify the specified movement parts;

[0087] Once a pedestrian is detected within the detection area, the system begins analyzing the target's movement trajectory, which can be the movement trajectory of specific parts such as hand gestures or foot movements.

[0088] S204: Detect the movement trajectory of the moving part, and when the movement trajectory meets the preset conditions, control the corresponding door to open.

[0089] In step S202, a detection area of ​​a predetermined size is defined in front of each vehicle door. The size of the detection area is set according to the radar detection envelope range, as long as it can accurately identify the user-specified motion trajectory.

[0090] For example, the prescribed action is a foot-stepping action, as shown in Figure 6. A detection area 302 is set in front of the radar 301, with a width of L3 and a length of L4. The center normal of the detection area 302 coincides with the radar normal. It is recommended that L3 = L4 = 300mm. This area size can cover the size of an adult male's foot, making the overall detection success rate higher.

[0091] For example, the specified action is a foot sweeping motion, as shown in Figure 7. A detection area 402 is set in front of the radar 401. The detection area 402 is divided into two rectangles with a width of L5 and a length of L6. In order to make the radar detection more effective, the center normals of the two detection areas A and B are kept to coincide with the radar normal.

[0092] In step S203, point cloud data is generated for the detected pedestrians, and the action parts of the prescribed actions are identified based on the point cloud data. For example, taking the foot as the action part, the specific method for identifying the action parts of the prescribed actions based on the point cloud data includes: (1) extracting pedestrian point cloud data at each time moment from continuous point cloud data, and removing ground points using a ground segmentation algorithm (such as plane fitting based on RANSAC); (2) converting the pedestrian point cloud data at each time moment into a two-dimensional image; (3) performing preliminary positioning of the foot according to the position of the foot in the body part; (4) identifying the foot based on a pre-trained foot recognition model. The foot recognition model is trained based on a convolutional neural network. The dataset used for training includes an annotated dataset, which includes feet wearing different types of shoes and feet in different orientations. The source of the dataset may include images converted from point cloud data, or images containing feet that have been grayscaled. More specifically, the confidence level of the foot recognition model's output is obtained, foot targets at continuous time points are acquired based on continuous point cloud data, and the reasonableness of foot targets extracted at adjacent time points is judged based on the continuity of foot movements. Based on this, it is possible to distinguish between feet and similar objects (such as suitcases, animals, etc.).

[0093] To improve the accuracy of motion trajectory recognition, in step S204, the motion trajectory is detected based on the motion parts at consecutive time points. A sequence of identifying features for the motion and preset conditions are set. The sequence of identifying features includes multiple consecutive positions that the motion part must reach, and the conditions include motion height, motion speed, motion time, and motion stability. When the motion height, motion speed, and / or motion time all meet the set conditions, the motion trajectory is deemed valid. By imposing these conditions, the identifying characteristics of the specified motion can be improved, and false triggering can be avoided. It is understood that different specified motions will have different identifying feature sequences and condition settings. By comprehensively analyzing the target's motion height, speed, and / or time, it is possible to more accurately determine whether the target has performed a valid door-opening motion.

[0094] For example, the specified action is a foot-stomping action, and the defined characteristic sequence includes lifting the foot, pressing the foot down, and lifting the foot again. The defined conditions include: lifting height and hovering time, and pressing down and holding time, etc. For example, the foot-stomping action must last for more than 0.5 seconds, and the total duration is 1-1.5 seconds. When a foot target satisfying the above characteristic sequence is detected from a continuous sequence of foot targets, it is considered to satisfy the complete "press-hold-lift" action trajectory. Furthermore, when the action height, action time, and action stability all meet the defined conditions, the action is considered valid.

[0095] For example, the specified action is a foot sweeping motion. The defined characteristic sequence includes the foot moving to one side of the inspection area, the foot reaching the middle of the detection area, the foot moving to the other side of the inspection area, the foot reaching the middle of the detection area again, and the foot returning to the side of the initial action within the inspection area, such as sweeping from point A to point B, and then sweeping from point B back to point A (as shown in Figure 7). The defined conditions include: speed, rate of change of height during the action, and stability conditions. For example, the action speed must be 0.2-1.5 m / s, the foot movement trajectory must be from designated area A to B, and then from B to A, with an azimuth change > 15°; the stability condition is: maintaining a closeness trend with a distance change < 0.2 m within 0.5 seconds for 5 consecutive frames. When a foot target satisfying the above characteristic sequence is detected from foot targets at consecutive time points, the action trajectory is considered satisfactory. Simultaneously, if the speed, height change, and stability all meet the defined conditions, the action is considered valid.

[0096] Compared to the method of action recognition based on a standard action library, the above method, by setting a sequence of identifier features, only needs to capture a few identifier positions in the action trajectory to make a judgment on the standardization of the action. Therefore, it is not necessary to accurately identify the entire action trajectory, thus improving the sensitivity of detection.

[0097] It is understandable that the specific values ​​in the above conditions can all be determined through calibration.

[0098] When multiple pedestrians are identified within the inspection area, priority is given to responding to the target closest to the radar, while other targets are suppressed.

[0099] In specific scenarios, there is a need to help others open vehicle doors. For example, a passenger carrying items needs to board from the rear but cannot easily open the door manually, or they want to open the tailgate but the space in front of it is too small for radar to detect and recognize foot movements. In these cases, some embodiments establish a complete standard action library, assigning different actions to each door, thereby enabling the recognition of user actions based on the specific requirements of different doors. Specifically, the standard action library includes the corresponding identifier feature sequence and the conditions that must be met for each door. The recognized action trajectory is compared with multiple pre-stored identifier feature sequences and conditions; if the comparison is successful, the corresponding door is opened. Based on this, to open a door for a passenger, the driver or other passengers can perform a specified action in front of their own side of the door. In other embodiments, user-defined action trajectories can also be used, allowing users to set door-opening actions according to their personal habits, further improving the system's flexibility and user satisfaction.

[0100] Figure 8 shows a flowchart of a third vehicle door control method based on millimeter-wave radar provided by one or more embodiments of the present invention, applicable to scenario one or scenario two. The method is applied to a vehicle controller and specifically includes:

[0101] S301: In response to the detection signal that the key has entered the effective sensing range, the millimeter-wave radar is triggered to start;

[0102] This step marks the start of system activity; when a legitimate key enters the car's sensing area, the millimeter-wave radar is activated and ready to begin operation.

[0103] S302: Receives radar echo signals in real time and performs pedestrian detection within the designated detection area in front of each vehicle door;

[0104] Once activated, the radar will continuously transmit and receive electromagnetic waves. By analyzing the echo signals, it will be able to monitor and identify pedestrians or other objects within the preset detection area in real time.

[0105] S303: If a pedestrian is detected in one or more detection areas, track the movement trajectory of the pedestrian;

[0106] Once a pedestrian is detected, the system will use tracking algorithms, such as Kalman filtering, to continuously monitor and record the pedestrian's movement trajectory, ensuring that its movement path can be accurately tracked.

[0107] S304: Obtain the angle change between the motion trajectory and the corresponding door's vertical direction, as well as the rate of change of the distance between the pedestrian and the door;

[0108] The changes in the angle of a pedestrian's movement trajectory relative to the vertical direction of the vehicle door, as well as the changes in the distance between the pedestrian and the vehicle door per unit time, are used to determine whether a pedestrian intends to board the vehicle and to distinguish them from casual passing by.

[0109] S305: When the angle change and the distance change rate meet the set conditions, it is determined that the pedestrian has the intention to open the door;

[0110] Based on the above analysis, if the angle continues to decrease (tending towards the direction of the door normal) and the rate of change of distance is within the preset negative range, it indicates that the pedestrian is rapidly and directly approaching the door, and the system infers that the pedestrian may have the intention to open the door.

[0111] S306: Identify the specified movement parts of the pedestrian; the specified movement can be either the hand or the foot.

[0112] S307: Detect the movement trajectory of the moving part, and when the movement trajectory meets the preset conditions, control the corresponding door to open.

[0113] The preset conditions ensure that the door will only open when the user performs the exact opening action. This effectively prevents misoperation and accidental opening, especially when there are other moving objects around the vehicle. Through precise analysis of motion trajectories, the system can accurately distinguish between user actions and unintentional interference, thereby enhancing vehicle safety.

[0114] The above method first determines whether a pedestrian intends to board the vehicle by tracking their trajectory, and then performs authentication through prescribed actions, which further improves the accuracy of the judgment, enables more precise control of the vehicle door opening, avoids unnecessary misoperation, and solves the problem of illegal intrusion that may occur in public places.

[0115] The calculation and judgment conditions for trajectory tracking, angle change and distance change rate in steps S303-S305 are the same as those in steps S103-S105. The process of motion trajectory detection and judgment in steps S306-S307 is the same as that in steps S203-S204, and will not be repeated here.

[0116] Based on the first door control method described above, one or more embodiments of the present invention also provide a door control device based on millimeter-wave radar, comprising: a function trigger module configured to trigger the millimeter-wave radar to start in response to a detection signal indicating that the key has entered the effective sensing range; a pedestrian detection module configured to receive radar echo signals in real time and detect pedestrians around the vehicle; a trajectory tracking module configured to track the movement trajectory of a pedestrian when a pedestrian is detected, and if the pedestrian is approaching the vehicle, determine the door the pedestrian is facing and define a detection area in front of the door; and a door control module configured to continuously track the movement trajectory of a pedestrian after the pedestrian enters the detection area, obtain the angle change between the movement trajectory and the vertical direction of the door, and the rate of change of the distance between the pedestrian and the door, and control the door to open when the angle change and the rate of change of the distance meet set conditions.

[0117] Based on the second door control method described above, one or more embodiments of the present invention also provide a door control device based on millimeter-wave radar, comprising: a function trigger module configured to trigger the millimeter-wave radar to start in response to a detection signal indicating that a key has entered the effective sensing range; a pedestrian detection module configured to receive radar echo signals in real time and detect pedestrians around the vehicle; a motion part recognition module configured to recognize a specified motion part when a pedestrian is detected in one or more detection areas; and a door control module configured to detect the motion trajectory of the motion part and control the corresponding door to open when the motion trajectory meets preset conditions.

[0118] Based on the third door control method described above, one or more embodiments of the present invention also provide a door control device based on millimeter-wave radar, comprising: a function trigger module configured to trigger the millimeter-wave radar to start in response to a detection signal indicating that a key has entered the effective sensing range; a pedestrian detection module configured to receive radar echo signals in real time and detect pedestrians around the vehicle; a trajectory tracking module configured to track the movement trajectory of a pedestrian if a pedestrian is detected in one or more detection areas; a door opening intention control module configured to acquire the angle change between the movement trajectory and the vertical direction of the corresponding door, and the rate of change of the distance between the pedestrian and the door; determine that the pedestrian has the intention to open the door when the angle change and the rate of change of the distance meet set conditions; and a door opening control module configured to identify a specified action part of the pedestrian; detect the movement trajectory of the action part, and control the corresponding door to open when the movement trajectory meets preset conditions. One or more embodiments of the present invention also provide an electronic device that can be used to implement the door control method in the above embodiments. The electronic device includes one or more processors, one or more memories coupled to the processors, and a communication module coupled to the processors.

[0119] The memory may include one or more non-volatile memories and one or more volatile memories. Examples of non-volatile memories include, but are not limited to, at least one of the following: read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, hard disk, compact disc (CD), digital video disc (DVD), or other magnetic and / or optical storage. Examples of volatile memories include, but are not limited to, at least one of the following: random access memory (RAM), or other volatile memories that do not persist during the duration of a power outage. The computer program may be stored in the ROM. When the processor executes the computer program, it implements the above-described door control method.

[0120] In some embodiments, the program may be tangibly contained in a computer-readable medium, which may include a device (such as in memory) or other storage device accessible by the device. The program may be loaded from the computer-readable medium into RAM for execution. The computer-readable medium may include any type of tangible non-volatile memory, such as ROM, EPROM, flash memory, hard disk, whereby the computer-readable storage medium stores a computer program that, when executed by a processor, implements the aforementioned door control method.

[0121] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a server or terminal, they generate all or part of the processes or functions described in the embodiments of this application. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic cable, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to the server or terminal, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, and magnetic tape), an optical medium (e.g., digital video disk (DVD), etc.), or a semiconductor medium (e.g., solid-state drive).

[0122] Furthermore, although the operations are described in a specific order, this should be understood as requiring that such operations be performed in the specific order shown or in sequential order, or requiring that all illustrated operations be performed to achieve the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this application. Certain features described in the context of individual embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations.

[0123] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A vehicle door control method based on millimeter-wave radar, characterized in that, Includes the following steps: The millimeter-wave radar is activated in response to a detection signal indicating that the key has entered the effective sensing range. It receives radar echo signals in real time and performs pedestrian detection within the designated detection area in front of each vehicle door; If a pedestrian is detected in one or more detection areas, the pedestrian's motion trajectory is tracked; the angle change between the motion trajectory and the corresponding door's vertical direction, as well as the rate of change of the distance between the pedestrian and the door, are obtained; When the angle change and distance change rate meet the set conditions, the door is controlled to open.

2. The vehicle door control method based on millimeter-wave radar as described in claim 1, characterized in that, The set conditions are: the angle between the movement trajectory and the vertical direction of the car door continuously decreases, the distance change rate is negative and within the set range.

3. The vehicle door control method based on millimeter-wave radar as described in claim 1, characterized in that, When tracking the pedestrian's movement trajectory, a time window is set. If the rate of change of the movement trajectory angle exceeds a set threshold within the time window, and the angle between the movement direction and the perpendicular direction of the car door exceeds a set threshold, the pedestrian will no longer be tracked.

4. The vehicle door control method based on millimeter-wave radar as described in claim 1, characterized in that, The detection area is rectangular, with the width of the rectangle set according to the width of an adult's body and the length of the rectangle set according to the adult's walking speed and the opening degree of the car door.

5. The vehicle door control method based on millimeter-wave radar as described in claim 4, characterized in that, When tracking the movement trajectory of pedestrians, the system also detects the pedestrian's walking speed in real time. If the speed is lower than the set speed threshold, it predicts the time required for the pedestrian to reach the door's movement space. Based on this time and the time required for the door to open, it determines the door's opening time and speed.

6. A vehicle door control method based on millimeter-wave radar, characterized in that, Includes the following steps: The millimeter-wave radar is activated in response to a detection signal indicating that the key has entered the effective sensing range. It receives radar echo signals in real time and performs pedestrian detection within the designated detection area in front of each vehicle door; If a pedestrian is detected in one or more detection areas, the specified movement parts are identified; The movement trajectory of the moving part is detected, and when the movement trajectory meets the preset conditions, the corresponding door is controlled to open.

7. The vehicle door control method based on millimeter-wave radar as described in claim 6, characterized in that, The system predetermines a sequence of identifying features for a given action and predetermined conditions to be met. The sequence of identifying features includes multiple consecutive positions that the action part needs to reach, and the conditions to be met include one or more of the following: action height, action speed, action time, and action stability. When the detected action trajectory meets the set sequence of identifying features and the predetermined conditions, the motion trajectory is deemed valid, and the corresponding door is controlled to open.

8. A vehicle door control method based on millimeter-wave radar, characterized in that, Includes the following steps: The millimeter-wave radar is activated in response to a detection signal indicating that the key has entered the effective sensing range. It receives radar echo signals in real time and performs pedestrian detection within the designated detection area in front of each vehicle door; If a pedestrian is detected in one or more detection areas, the pedestrian's movement trajectory is tracked; Obtain the angle change between the motion trajectory and the corresponding vertical direction of the car door, as well as the rate of change of the distance between the pedestrian and the car door; When the angle change and the rate of change of distance meet the set conditions, it is determined that the pedestrian has the intention to open the door; Identify the specific movement parts of the pedestrian; The movement trajectory of the moving part is detected, and when the movement trajectory meets the preset conditions, the corresponding door is controlled to open.

9. The vehicle door control method based on millimeter-wave radar as described in claim 8, characterized in that, The set conditions are: the angle between the movement trajectory and the vertical direction of the car door continuously decreases, the distance change rate is negative and within the set range.

10. The vehicle door control method based on millimeter-wave radar as described in claim 8, characterized in that, When tracking the pedestrian's movement trajectory, a time window is set. If the rate of change of the movement trajectory angle exceeds a set threshold within the time window, and the angle between the movement direction and the perpendicular direction of the car door exceeds a set threshold, the pedestrian will no longer be tracked.

11. The vehicle door control method based on millimeter-wave radar as described in claim 8, characterized in that, The detection area is rectangular, with the width of the rectangle set according to the width of an adult's body and the length of the rectangle set according to the adult's walking speed and the opening degree of the car door.

12. The vehicle door control method based on millimeter-wave radar as described in claim 11, characterized in that, When tracking the movement trajectory of pedestrians, the system also detects the pedestrian's walking speed in real time. If the speed is lower than the set speed threshold, it predicts the time required for the pedestrian to reach the door's movement space. Based on this time and the time required for the door to open, it determines the door's opening time and speed.

13. The vehicle door control method based on millimeter-wave radar as described in claim 8, characterized in that, The system predetermines a sequence of identifying features for a given action and predetermined conditions to be met. The sequence of identifying features includes multiple consecutive positions that the action part needs to reach, and the conditions to be met include one or more of the following: action height, action speed, action time, and action stability. When the detected action trajectory meets the set sequence of identifying features and the predetermined conditions, the motion trajectory is deemed valid, and the corresponding door is controlled to open.

14. A vehicle door control device based on millimeter-wave radar, characterized in that, include: The function trigger module is configured to trigger the millimeter-wave radar to start in response to a detection signal that the key has entered the effective sensing range; The pedestrian detection module is configured to receive radar echo signals in real time and detect pedestrians around the vehicle. The trajectory tracking module is configured to track the movement trajectory of a pedestrian when a pedestrian is detected. If the pedestrian is approaching a vehicle, the module determines the door the pedestrian is facing and defines a detection area in front of the door. The door control module is configured to continuously track the movement trajectory of a pedestrian after detecting that the pedestrian has entered the detection area, obtain the angle change between the movement trajectory and the vertical direction of the door, and the rate of change of the distance between the pedestrian and the door. When the angle change and the rate of change of the distance meet the set conditions, the door is controlled to open.

15. An electronic device, characterized in that, The device includes a processor and a memory, the memory storing computer instructions that, when executed by the processor, cause the electronic device to perform the method of claim 1.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method of claim 1.

17. A computer program product, the computer program product comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in claim 1.