Vehicle control method and device, vehicle and storage medium

Abstract

The invention discloses a vehicle control method and device, a vehicle and a storage medium, and relates to the technical field of vehicle control. The scheme comprises the following steps: acquiring perception data of at least two data channels of a back row of a target vehicle; based on the perception data of each data channel, determining a channel score value corresponding to each data channel, the channel score value being used for representing a child existence probability determined based on the perception data of the data channel; determining a target score according to the channel score values of the at least two data channels; and under the condition that the target score exceeds a preset threshold value, a control instruction is generated to control the target vehicle to execute child protection operation. According to the scheme, the accuracy of child existence judgment is improved through multi-data channel fusion, so that child protection operation can be automatically triggered, and safety protection of children in the rear row of the vehicle is enhanced.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and in particular to a vehicle control method, device, vehicle, and storage medium. Background Technology

[0002] With the continuous improvement of vehicle intelligence and safety standards, the safety of child passengers is receiving increasing attention. To effectively prevent children from accidentally opening vehicle doors while the vehicle is in motion or temporarily parked, existing technologies generally rely on child safety locks, which lock the internal opening mechanisms of the rear doors through physical or electronic means. Therefore, ensuring that child safety locks can be reliably and promptly activated when necessary is one of the core aspects of improving children's passive safety.

[0003] However, relevant technologies for vehicle child safety locks mainly fall into two categories: mechanical and electronic. Mechanical safety locks typically require manual switching on the rear door side using a key or special tool; electronic locks can be manually activated or deactivated by the driver via physical or touch buttons on the center console. A common limitation of both types is that their activation depends entirely on the driver's active recognition, memory, and manual operation. In real-world driving scenarios, drivers may fail to activate the child safety locks due to negligence, forgetfulness, or failure to notice when temporarily carrying a child, thus creating a safety hazard. Summary of the Invention

[0004] In view of this, this application aims to provide a vehicle control method, device, vehicle, and storage medium that can automatically trigger child protection operations to enhance the safety of children in the rear seats of a vehicle.

[0005] A first aspect of this invention provides a vehicle control method, comprising: acquiring perception data from at least two data channels targeting the rear seats of a target vehicle; determining a channel score value corresponding to each data channel based on the perception data from each data channel; wherein the channel score value is used to characterize the probability of a child being present in the rear seats of the target vehicle and is related to the type of rear seat occupant characterized by the corresponding perception data; for the same data channel, the channel score value corresponding to a rear seat occupant type including a child is higher than the channel score value corresponding to a rear seat occupant type including only an adult; the channel score value corresponding to a rear seat occupant type including only an adult is higher than the channel score value corresponding to a rear seat occupant type being unoccupied; determining a target score based on the channel score values ​​from the at least two data channels; and generating a control command for controlling the target vehicle to perform child protection operations if the target score exceeds a preset threshold. Traditional solutions use binary classification to directly determine the presence or absence of a child. When the sensor misjudges at the boundary between a child and an adult, it directly leads to an incorrect final judgment. This application introduces an intermediate adult score, ensuring that even when a channel misidentifies a child as an adult, its output intermediate score remains higher than in the unoccupied state, while correct scores from other channels compensate for this bias. Similarly, when an adult is misidentified as a child, their score remains lower than that of a real child. This allows the target score after multi-channel fusion to more accurately reflect the actual situation, effectively solving the misjudgment problem at recognition boundaries in traditional solutions and significantly improving the accuracy and reliability of child safety lock and other protective functions.

[0006] Optionally, generating control instructions for controlling the target vehicle to perform child protection operations includes: generating control instructions for controlling the child safety lock of the target vehicle based on monitoring results of the occupant in the driver's seat of the target vehicle. By monitoring the occupant in the driver's seat of the target vehicle and obtaining monitoring results, and generating control instructions for controlling the child safety lock of the target vehicle based on the monitoring results, targeted control of the child safety lock is achieved. This solution adjusts the opening and closing of the child safety lock according to the actual state of the occupant in the driver's seat, avoiding the inconvenience and safety hazards caused by the child safety lock being fixed on or off. When there is an occupant in the driver's seat, the child safety lock can be activated to prevent children in the rear seat from accidentally opening the door. The lock is automatically released when the driver leaves the vehicle, which can prevent children from being accidentally locked inside the vehicle, thus improving the rationality and reliability of the child safety lock control.

[0007] Optionally, the step of generating control commands for the child safety lock of the target vehicle based on the monitoring results of the driver's seat occupant includes at least one of the following: generating a first control command to open the child safety lock when the child safety lock is in the closed state and the monitoring results indicate that an occupant is detected in the driver's seat; generating a second control command when the child safety lock is in the open state and the monitoring results indicate that the driver's seat occupant intends to leave the vehicle; the second control command is used to indicate the presence of a child in the vehicle when the child safety lock is in the open state; and generating a third control command to close the child safety lock when the monitoring results indicate that the driver's seat occupant has left the target vehicle. By monitoring the driver's seat occupant of the target vehicle and obtaining the monitoring results, and generating control commands for the child safety lock of the target vehicle based on the monitoring results, targeted control of the child safety lock is achieved. This solution adjusts the opening and closing of the child safety lock based on the actual status of the driver and passenger, avoiding the inconvenience and safety hazards caused by the child safety lock being fixed on or off. When there is a passenger in the driver's seat, the child safety lock can be activated to prevent children in the rear seat from accidentally opening the door. The lock is automatically released when the driver leaves the vehicle, which can prevent children from being accidentally locked inside the vehicle, thus improving the rationality and reliability of the child safety lock control.

[0008] Optionally, determining the target score based on the channel score values ​​of the at least two data channels includes: determining a weight value corresponding to each data channel based on the current driving scenario; and performing a weighted summation of the channel score values ​​and corresponding weight values ​​for each data channel to obtain the target score. The reliability of each data channel is evaluated based on the current driving scenario, and its weight value is dynamically allocated. The channel score values ​​are then weighted and summed to calculate the target score. This method can adaptively increase the weight ratio of reliable data channels according to environmental changes, making the target score more accurately reflect the actual state of the rear-seat occupants. This effectively overcomes the misjudgment problem caused by the failure of certain sensor channels in specific scenarios under fixed-weight schemes, significantly improving the robustness and environmental adaptability of child state recognition.

[0009] Optionally, the control command is used to activate the child safety lock of the target vehicle. Before generating the control command to control the target vehicle to perform child protection operations, the method further includes: determining a preset threshold based on the current driving scenario of the target vehicle, wherein the value of the preset threshold is negatively correlated with the door opening risk level corresponding to the current driving scenario. By determining a preset threshold based on the current driving scenario of the target vehicle before generating the control command to activate the child safety lock, and the value of the preset threshold is negatively correlated with the door opening risk level corresponding to the current driving scenario, and then determining whether to activate the child safety lock based on the comparison result between the target score and the preset threshold, a lower preset threshold value can make it easier for the target score to reach the critical value, ensuring that the child safety lock is activated in a timely manner and effectively avoiding the high safety risk caused by accidental opening of the rear door. In scenarios with low door opening risk, a higher preset threshold value can reduce the accidental activation of the child safety lock and avoid causing inconvenience to the user. This allows the activation logic of the child safety lock to be precisely matched with real-time safety needs, effectively preventing the high risk of accidental opening of the car door while driving, while also taking into account ease of use and user experience.

[0010] Optionally, acquiring perception data from at least two data channels for the rear seats of the target vehicle includes acquiring at least two of the following: image data from an image data channel, audio data from an audio data channel, rear seat pressure values ​​from a seat pressure data channel, and millimeter-wave radar data from a millimeter-wave radar data channel. Correspondingly, determining a channel score value corresponding to each data channel based on the perception data from each of the data channels includes: when the data channels include an image data channel, determining a channel score value for the image data channel based on the human morphological features represented by the image data; when the data channels include an audio data channel, determining a channel score value for the audio data channel based on at least one of the pitch features, timbre features, and volume features contained in the audio data; when the data channels include a seat pressure data channel, determining a channel score value for the seat pressure data channel based on the weight information represented by the rear seat pressure values; and when the data channels include a millimeter-wave radar data channel, determining a channel score value for the millimeter-wave radar data channel based on the human motion features represented by the millimeter-wave radar data. By acquiring data from at least two of the following channels—image data, audio data, seat pressure data, and millimeter-wave radar data—as perceptual data, and determining the channel score for each data channel based on the human morphological features, tone features, timbre features, volume features, weight information, and human motion features represented by the perceptual data from each channel, human morphological features from image data can intuitively distinguish children from adults through facial features and head-to-body ratio; tone, timbre, and volume features from audio data can help determine the occupant's status through voice differences; weight information from seat pressure data can initially distinguish children from adults through weight ranges; and human motion features from millimeter-wave radar data can help confirm whether there are people in the back seat and their activity status. The complementary combination of multi-channel data effectively avoids misjudgments caused by perceptual biases in a single data channel, compensates for the limitations of single-channel feature recognition, and ensures that when data from one channel is affected by environmental interference, other channels can still provide reliable information, significantly improving robustness and accuracy in complex environments.

[0011] Optionally, determining the channel score value of the image data channel based on the human morphological features represented by the image data includes: inputting the image data into a pre-trained classification model to obtain a classification result output by the classification model corresponding to the image data; and determining the channel score value of the image data channel based on the classification result. The classification model is trained on a training set containing image samples of children, adults, and vehicles with no occupants in the back seat. By introducing a classification model trained on a large number of samples to process the image data and output classification results, the channel score value is determined. This scheme utilizes machine learning methods, avoiding the limitations of traditional methods that rely on manually set thresholds and rules for feature judgment. It can more robustly handle changes in lighting, occupant posture, and shooting angle, directly learning and generalizing key visual features that distinguish children, adults, and empty states from the data, thereby improving the accuracy and environmental adaptability of the image channel score value.

[0012] According to a second aspect of this application, a vehicle control device is provided, comprising: The data acquisition module is used to acquire perception data from at least two data channels in the rear of the target vehicle. The channel scoring module is used to determine a channel score value corresponding to each data channel based on the perception data of each data channel; wherein, the channel score value is used to characterize the probability that there is a child in the back seat of the target vehicle, and is related to the type of rear-seat occupant characterized by the corresponding perception data; for the same data channel, the channel score value corresponding to the rear-seat occupant type including children is higher than the channel score value corresponding to the rear-seat occupant type including only adults; the channel score value corresponding to the rear-seat occupant type including only adults is higher than the channel score value corresponding to the rear-seat occupant type being unoccupied; The target scoring module is used to determine the target score based on the channel score values ​​of the at least two data channels; The instruction generation module is used to generate control instructions when the target score exceeds a preset threshold; the control instructions are used to control the target vehicle to perform child protection operations.

[0013] According to a third aspect of this application, a computer-readable storage medium is provided, the storage medium storing a computer program for performing the methods described in any of the above embodiments.

[0014] According to a fourth aspect of this application, an electronic device is provided, comprising: a processor; a memory for storing processor-executable instructions; the processor being configured to perform the method described in any of the above embodiments.

[0015] According to a fifth aspect of this application, a vehicle is provided, including the aforementioned electronic equipment.

[0016] This application provides a vehicle control method, apparatus, device, vehicle, and storage medium. The solution includes: acquiring perception data from at least two data channels targeting the rear seats of a target vehicle; determining a channel score value corresponding to each data channel based on the perception data from each data channel, wherein the channel score value characterizes the probability of a child's presence determined based on the perception data from that data channel; determining a target score based on the channel score values ​​from at least two data channels; and generating a control command to control the target vehicle to perform child protection operations when the target score exceeds a preset threshold. This solution improves the accuracy of child presence determination through multi-data-channel fusion, thereby automatically triggering child protection operations and enhancing the safety of children in the rear seats of the vehicle. Attached Figure Description

[0017] Figure 1 The diagram shown is a flowchart of a vehicle control method provided in one embodiment of this application.

[0018] Figure 2 The diagram shown is a block diagram of a vehicle control device provided in one embodiment of this application.

[0019] Figure 3 The diagram shown is a structural block diagram of an electronic device provided in one embodiment of this application. Detailed Implementation

[0020] 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, and 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.

[0021] Application Overview In the field of automatic control technology for vehicle child safety locks, a common approach to automatically identify and trigger protection functions for children in the rear seats is to rely on a single sensor data source for determination. Specifically, this solution collects data using a single type of sensor deployed in the rear of the vehicle and outputs a binary result of "child present" or "child not present" based on preset, fixed rules. Its basic working principle is to compare the raw sensor data with a static threshold, thereby driving the child safety lock's actuator. This approach was adopted in early solutions primarily because it allows for basic function triggering with low system complexity and direct logic control.

[0022] However, when this solution is applied to real-world, ever-changing vehicle usage environments, its reliability faces significant challenges. In its pursuit of simplicity and low cost, the solution's inherent single data source and fixed decision-making logic inevitably compromise the robustness and accuracy of status determination, potentially leading to false or missed triggers in specific scenarios. Specifically, in situations such as insufficient lighting at night, obstructed occupant posture, or temporary sensor coverage, information from a single sensor may become ineffective or ambiguous. For instance, a camera may fail to accurately analyze facial features due to insufficient light, and a pressure sensor may be unable to distinguish between a child and a heavy object placed on the seat, resulting in an inaccurate identification of a child's presence and making the activation of the child protection function unreliable.

[0023] To address the aforementioned issues, this application provides a vehicle control method. The method acquires perception data from at least two data channels in the rear of a target vehicle; determines a corresponding channel score based on the perception data of each data channel, where the channel score characterizes the probability of a child's presence under the perception data of that channel, with the highest score for a child's state, followed by an adult's state, and the lowest score for an unoccupied state for the same data channel; determines a target score based on the channel score of each data channel; and generates a control command to control the vehicle to perform child protection operations when the target score exceeds a preset threshold.

[0024] When the sensor detects an older child who is close to an adult, although a single sensor may give an intermediate score close to that of an adult, the intermediate scores from multiple sensors, after being fused, can still reach the threshold for child identification through probability accumulation, thus avoiding missed detections. Conversely, for a true adult, since the score is consistently lower than that of a child, the accumulated intermediate scores from multiple sensors can still remain below the threshold, avoiding false detections. This significantly improves the accuracy and robustness of child presence detection, effectively avoiding false triggering and missed triggering problems caused by blurred occupant feature edges, and contributing to the intelligent and dynamic control of child safety locks.

[0025] After introducing the basic principles of this application, various non-limiting embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0026] Exemplary methods Figure 1 This is a schematic flowchart of a vehicle control method provided in one embodiment of this application. Figure 2 The method described is executed by an onboard computing device, which may be an electronic control unit (ECU), a domain controller, an onboard central computing platform, a mobile terminal communicating with the vehicle, or other onboard computing devices with data processing capabilities. This application does not limit the specific onboard computing device to these devices. Figure 1 As shown, the method includes the following: Step S110: Acquire perception data for at least two data channels in the rear of the target vehicle.

[0027] In this embodiment of the application, the target vehicle refers to a motor vehicle equipped with the sensing sensors, on-board computing devices and child locks corresponding to the vehicle control method described in this application. The specific power type, size and specifications of the vehicle are not limited, as long as they can support the layout and continuous operation of the relevant sensors.

[0028] In this embodiment, the data channel refers to a data stream collected from a single type of sensor and transmitted to the processing unit. Each data channel corresponds to a specific perception mode, and its data reflects the state information of a certain dimension of the rear seats. For example, the image data channel corresponds to an in-vehicle camera, the audio data channel corresponds to an in-vehicle microphone, the seat pressure data channel corresponds to a seat pressure sensor, and the millimeter-wave radar data channel corresponds to millimeter-wave radar.

[0029] In the embodiments of this application, the perception data refers to the raw data collected by various sensors to characterize the state of the rear passengers, such as rear passenger image data collected by vehicle cameras, rear passenger audio data collected by vehicle microphones, rear seat pressure values ​​detected by seat pressure sensors, and rear human motion characteristic data collected by millimeter-wave radar.

[0030] Step S120: Based on the perception data of each data channel, determine the channel score value corresponding to each data channel; wherein, the channel score value is used to characterize the probability that there is a child in the back seat of the target vehicle, and is related to the type of rear passenger represented by the corresponding perception data; for the same data channel, the channel score value corresponding to the rear passenger type including children is higher than the channel score value corresponding to the rear passenger type including only adults; the channel score value corresponding to the rear passenger type including only adults is higher than the channel score value corresponding to the rear passenger type being unoccupied.

[0031] In this embodiment, the channel score is a quantified value calculated based on the perception data of a single data channel. It is used to characterize the probability of a child being present in the back seat of a target vehicle under the perception dimension of that channel. The channel score is positively correlated with the probability of a child's presence. Specifically, the channel score generally refers to a numerical value, level, or probability used to quantify the likelihood of a child's presence inferred from the perception information of a certain data channel. For example, it can be represented as a score between 0 and 100, where a high score represents a high probability of a child's presence; it can also be represented as three levels: high, medium, and low; or it can be a probability value between 0 and 1.

[0032] In this embodiment, the rear passenger type refers to the category of occupants in the rear of the target vehicle as represented by the perception data. Specifically, the rear passenger type is mainly divided into three categories: including children, including adults only, and no one. Including children means there is at least one child in the rear seat, possibly accompanied by an adult; including adults only means there are only adult occupants in the rear seat, without any children; and no one means no occupants are detected in the rear seat. For example, in the image data channel, if the perception data identifies one child and one adult in the rear seat, the corresponding rear passenger type is including children; if only one adult is identified, the corresponding rear passenger type is including adults only; and if no one is identified, the corresponding rear passenger type is no one.

[0033] In this embodiment, the channel score is highest when the rear passenger type includes children; second highest when the rear passenger type includes only adults; and lowest when there is no one in the rear passenger area. This arrangement reflects the order of probability of a child's presence under different rear passenger types. The presence of a child is the core condition for triggering protection operations; therefore, the highest channel score should be assigned when the sensing data points to a child. When the rear passenger type includes only adults, the sensing data may fall within the boundary between children and adults, such as when a child's weight is close to that of an adult, or when the child is obscured. This could lead to misidentification of a child as an adult. While the probability of a child's presence is significantly lower than in a scenario where a child is clearly identified, it is still higher than in a completely unoccupied state, hence an intermediate score is assigned. An unoccupied state indicates that the channel has not detected any occupants, resulting in the lowest probability of a child's presence and the lowest corresponding score.

[0034] For example, in the seat pressure data channel, the channel score for children with weight is higher than that for adults with weight, and the channel score for adults with weight is higher than that for no pressure.

[0035] It is important to note that channel scores can be generated either by identifying the occupant type or directly mapped from the raw features of the perceived data. In the image data channel, the occupant type can be determined first and then assigned a score, or a score can be directly output based on facial feature parameters, body contour dimensions, etc., using a scoring algorithm. If the features match the typical range for children, a higher score is output; if they match the range for adults, a medium score is output; and if no relevant features are detected, a lower score is output. Similarly, in the audio data channel, scores can be directly calculated based on the frequency, amplitude, and other features of the audio signal. Features belonging to the typical range of children's speech correspond to higher scores, adult speech ranges correspond to medium scores, and the lowest score is output when there is no voice.

[0036] Step S130: Determine the target score based on the channel score values ​​of the at least two data channels.

[0037] In this embodiment, the target score generally refers to a single quantitative indicator used for final decision-making, obtained by comprehensively processing multiple channel score values. For example, it could be the total score after weighted summation, the median of multiple scores, or calculated using a more complex fusion algorithm.

[0038] Step S140: If the target score exceeds a preset threshold, a control command is generated to control the target vehicle to perform child protection operations.

[0039] In this embodiment, the preset threshold refers to the target score threshold value that determines the presence of a child in the back seat of the target vehicle and triggers child protection operation. This threshold is preset by the data processing unit according to actual application needs and is used to distinguish whether the status of the back seat occupants corresponding to the target score needs to be activated for protective measures. For example, in daytime vehicle driving scenarios, the preset threshold can be set to 70. When the target score reaches or exceeds 70, it is determined that there is a child in the back seat and subsequent operations are triggered. In nighttime scenarios, the threshold can be adjusted in combination with sensor weights to adapt to different perception conditions.

[0040] In this embodiment of the application, the control command is used to control the target vehicle to perform child protection operations.

[0041] In this embodiment of the application, the child protection operation refers to specific measures taken to ensure the safety of children in the rear seat, mainly including: automatically activating the child safety locks on the rear doors, activating the child left-in-vehicle monitoring system, and restricting the window raising and lowering functions, etc.

[0042] In this embodiment, sensory data is acquired from multiple independent data channels. Each channel outputs a standardized channel score based on its sensory characteristics, following the scoring rule of "children higher than adults, adults higher than unoccupied." The target vehicle is controlled to perform child protection operations by fusing the channel scores into a target score and comparing it with a preset threshold. Traditional solutions use binary classification to directly determine the presence of a child. When the sensor misjudges at the boundary between a child and an adult, it directly leads to an incorrect final judgment. This application introduces an intermediate adult score, ensuring that when a channel misjudges a child as an adult, its output intermediate score is still higher than the unoccupied state, while the correct scores from other channels compensate for this deviation. Similarly, when an adult is misjudged as a child, its score is still lower than the score of a real child. This allows the multi-channel fused target score to more accurately reflect the actual situation, effectively solving the misjudgment problem at the identification boundary in traditional solutions, and significantly improving the accuracy and reliability of child safety lock and other protective functions.

[0043] based on Figure 1In addition to the method described in the embodiments of this specification, some specific implementation schemes of the method are also provided, which will be described below.

[0044] Optionally, generating control instructions for controlling the target vehicle to perform child protection operations includes: Based on the monitoring results of the occupant in the driver's seat of the target vehicle, a control command is generated to control the child safety lock of the target vehicle.

[0045] In this embodiment of the application, the driver's seat refers to the position inside the target vehicle where the driver operates the vehicle, usually located on the left or right side of the front row; for example, the left seat in the front row of a family sedan is the driver's seat.

[0046] In this embodiment, the monitoring result refers to the judgment conclusion obtained through the detection and analysis of the occupant's status in the driver's seat, including but not limited to: occupant in the driver's seat, occupant intending to leave the vehicle, occupant having left the vehicle, etc. The signals relied upon for monitoring can come from various means such as the pressure sensor of the driver's seat, the door opening and closing sensor, the driver monitoring system (DMS), or the sensing status of the smart key. For example, when the pressure value detected by the seat pressure sensor continuously exceeds a preset threshold, it can be determined that "occupant in the driver's seat is occupied".

[0047] In this embodiment, the child safety lock is a dedicated locking mechanism installed on the rear door of a vehicle. When activated, it temporarily disconnects the door handle from the door lock mechanism, thereby preventing accidental opening of the door by occupants (especially children) while driving or parked. The child safety lock described in this application specifically refers to an electronic locking device whose opening and closing can be controlled by electrical signals.

[0048] The control command refers to the electronic signal generated by the vehicle control system to operate the child safety lock, and generally contains command information to enable or disable the safety lock. For example, when the system detects that the driver's seat occupant has left, it generates a command to disable the child safety lock to restore normal door operation.

[0049] In this embodiment, by monitoring the occupant in the driver's seat of the target vehicle and obtaining the monitoring results, a control command is generated based on the monitoring results to control the child safety lock of the target vehicle, thereby achieving targeted control of the child safety lock. This solution adjusts the opening and closing of the child safety lock according to the actual state of the occupant in the driver's seat, avoiding the inconvenience and safety hazards caused by the child safety lock being fixed on or off. When there is an occupant in the driver's seat, the child safety lock can be activated to prevent children in the rear seat from accidentally opening the door. The lock is automatically released when the driver leaves the vehicle, which can prevent children from being accidentally locked inside the vehicle, thus improving the rationality and reliability of the child safety lock control.

[0050] Optionally, the step of generating control commands for controlling the child safety locks of the target vehicle based on monitoring results of the driver's seat occupant includes at least one of the following: When the child safety lock is in the closed state and the monitoring result indicates that there is an occupant in the driver's seat, a first control command is generated to open the child safety lock. When the child safety lock is in the open state and the monitoring result indicates that the occupant in the driver's seat intends to leave the vehicle, a second control command is generated; the second control command is used to alert the vehicle that a child is in the vehicle when the child safety lock is in the open state. If the monitoring results indicate that the driver occupant has left the target vehicle, a third control command is generated to deactivate the child safety lock.

[0051] In this embodiment, the first control command refers to an electronic command generated when it is determined that the child safety lock is in the closed state and there is an occupant in the driver's seat, used to activate the child safety lock locking function. This command can be sent to the door control unit via the vehicle bus to drive the rear door lock mechanism into the locked state, thereby disabling the interior door opening device.

[0052] In this embodiment, the first control command, while triggering the child safety lock activation, can also be configured to simultaneously output a notification message. This notification message informs the driver that the child safety lock status has changed. For example, it can be displayed on the dashboard as an icon indicating "Child safety lock activated," or a notification can be given via voice announcement.

[0053] In this embodiment of the application, the driver's seat occupant's intention to leave the vehicle refers to the state in which the driver's seat occupant is preparing to leave the driver's seat. This state is determined by the following methods: the driver's seat door switch state changes from locked to open, the seat belt buckle state changes from fastened to unfastened, and the driver monitoring system recognizes the driver's action or posture of turning towards the outside of the vehicle.

[0054] In this embodiment, the second control command refers to a prompting command generated when the monitoring results indicate that the driver intends to leave the vehicle and the child safety lock is currently in the open state. The second control command only takes effect when the child safety lock is in the open state. It does not change the mechanical state of the child safety lock, but is only used to remind that there is a child in the vehicle, so as to prevent the occupant from forgetting the child in the vehicle when leaving the vehicle. The prompting method may include vehicle-side prompting and mobile terminal prompting. For example, the vehicle-side can issue a voice prompt of "The child safety lock is open and there is a child in the vehicle" through the vehicle audio system, and at the same time send a corresponding prompt message to the driver's mobile terminal.

[0055] In this embodiment, the third control command refers to an electronic command generated when the monitoring result indicates that the driver's seat occupant has left the vehicle, used to deactivate the child safety lock. This command restores the rear door to its normal inward and outward opening function, preventing children from being accidentally locked inside the vehicle. "Having left the vehicle" refers to either: leaving the passenger compartment of the target vehicle, or being far away from the target vehicle.

[0056] Determining whether the driver's seat occupant has left the passenger compartment can be achieved through pressure sensors, door opening / closing sensors, driver monitoring systems, etc., without specific limitations here. For example, when the driver's seat pressure sensor detects that the pressure value has dropped below a threshold, and the door status sensor confirms that the driver's door has completed one opening / closing cycle, it is determined that the driver's seat occupant has left the passenger compartment of the target vehicle, and a third control command is generated.

[0057] Determining whether the driver's seat occupant has moved away from the vehicle can be achieved through methods such as external cameras, key signal monitoring, user mobile terminal signal detection, or communication with the user's mobile terminal; no specific limitations are specified here. For example, when the vehicle key's sensing signal is lost, or when the detected key signal strength is below a preset threshold, it is determined that the driver's seat occupant has moved away from the vehicle, and a third control command is generated.

[0058] It is important to note that if the driver has left the passenger compartment but has not moved far from the vehicle, no third control command will be generated. For example, if the driver briefly stays near the vehicle after exiting, talks to someone inside, or moves around temporarily near the vehicle, even though they have left the driver's seat and the driver's door has been opened and closed, if the driver is still detected to be near the vehicle by means of an external camera, key signal, or mobile device positioning, it can be determined that they have not moved far from the vehicle, and the child safety lock will not be disengaged. This approach avoids hastily unlocking the rear door when the driver has only temporarily left the vehicle and the child is still in the back seat, thus balancing convenience and safety.

[0059] In this embodiment of the application, the third control command can also simultaneously generate a status change prompt when performing the child safety lock release operation. This prompt may include methods such as issuing a prompt sound via the vehicle horn or sending an unlock notification to the owner's mobile phone via the vehicle networking module.

[0060] In this embodiment, by monitoring the occupant in the driver's seat of the target vehicle and obtaining the monitoring results, a control command is generated based on the monitoring results to control the child safety lock of the target vehicle, thereby achieving targeted control of the child safety lock. This solution adjusts the opening and closing of the child safety lock according to the actual state of the occupant in the driver's seat, avoiding the inconvenience and safety hazards caused by the child safety lock being fixed on or off. When there is an occupant in the driver's seat, the child safety lock can be activated to prevent children in the rear seat from accidentally opening the door. The lock is automatically released when the driver leaves the vehicle, which can prevent children from being accidentally locked inside the vehicle, thus improving the rationality and reliability of the child safety lock control.

[0061] Optionally, determining the target score based on the channel score values ​​of the at least two data channels includes: Based on the current driving scenario, determine the weight value corresponding to each of the data channels; The target score is obtained by weighted summation of the channel score values ​​and corresponding weight values ​​of each data channel.

[0062] In this embodiment, the current driving scenario refers to the set of real-time vehicle operating status and in-vehicle environmental conditions used to determine the presence of a child. This includes not only the vehicle's own dynamics, such as whether it is moving or stationary, but also environmental factors that affect the reliability of sensor data, such as in-vehicle light intensity and ambient noise levels. In low-light environments such as nighttime or tunnels, the reliability of the image data channel may decrease; while in noisy driving environments, the reliability of the audio data channel may be affected.

[0063] In this embodiment, the weight value refers to a coefficient preset for each data channel, used for weighted summation to calculate the target score. Its value is adjusted according to the reliability of the channel in the current driving scenario. The higher the reliability of the channel's perception data, the larger the corresponding weight value, and vice versa. The weight value can be allocated by querying a preset scenario weight table or by real-time allocation through a preset algorithm, such as reinforcement learning (RL) algorithm, which can adjust the weight of each channel in real time according to scene changes. For example, when the in-vehicle illumination is sufficient, the image data channel has high reliability, and its weight value is set to a higher value; when the in-vehicle noise is high, the audio data channel has reduced reliability, and its weight value is set to a lower value.

[0064] In this embodiment, the reliability of each data channel is assessed based on the current driving scenario, and their weight values ​​are dynamically allocated. The channel scores are then weighted and summed to calculate the target score. This method adaptively increases the weight ratio of reliable data channels according to environmental changes, making the target score more accurately reflect the actual state of the rear-seat occupants. This effectively overcomes the misjudgment problem caused by the failure of certain sensor channels in specific scenarios under fixed-weight schemes, significantly improving the robustness and environmental adaptability of child state recognition.

[0065] Optionally, the control command is used to activate the child safety lock of the target vehicle; before generating the control command for controlling the target vehicle to perform child protection operations, the method further includes: The preset threshold is determined based on the current driving scenario of the target vehicle, wherein the value of the preset threshold is negatively correlated with the degree of door opening risk corresponding to the current driving scenario.

[0066] In this embodiment, the risk level of door opening refers to the degree of danger that may occur if the rear door of the target vehicle is accidentally opened in the current driving scenario. Its level is directly related to factors such as the vehicle's operating status and the surrounding environment. For example, when the vehicle is traveling at high speed, the accidental opening of the rear door is likely to cause serious accidents such as passengers falling from the vehicle, resulting in a high risk level; when the vehicle is stationary and parked in a safe area, the probability of an accident caused by the accidental opening of the rear door is low, resulting in a low risk level.

[0067] In this embodiment, the preset threshold refers to a critical score value used to determine whether the child safety lock needs to be activated. The specific value of this threshold is negatively correlated with the assessed risk level of opening the door. Specifically, a lower threshold is used in scenarios with a high risk level to improve the sensitivity of the safety lock activation, while a higher threshold is used in scenarios with a low risk level to reduce false triggering. For example, when the vehicle is traveling at high speed, the preset threshold is set to a lower value to ensure that the target score is likely to exceed the threshold and the child safety lock is activated in a timely manner; when the vehicle is stationary, the preset threshold is set to a higher value to prevent the child safety lock from being activated falsely.

[0068] In this embodiment, before generating the control command to activate the child safety lock, a preset threshold is determined based on the current driving scenario of the target vehicle. The value of the preset threshold is negatively correlated with the door opening risk level corresponding to the current driving scenario. Then, based on the comparison between the target score and the preset threshold, a decision is made on whether to activate the child safety lock. In scenarios with a high door opening risk, a lower preset threshold value makes it easier for the target score to reach the critical value, ensuring timely activation of the child safety lock and effectively avoiding the high safety risk caused by accidental opening of the rear doors. In scenarios with a low door opening risk, a higher preset threshold value reduces accidental activation of the child safety lock, avoiding inconvenience to the user. This allows the activation logic of the child safety lock to accurately match real-time safety needs, effectively preventing the high risk of accidental door opening during driving while also considering ease of use and user experience.

[0069] Optionally, acquiring perception data from at least two data channels for the rear seats of the target vehicle includes: Acquire at least two of the following: image data from the image data channel, audio data from the audio data channel, rear seat pressure values ​​from the seat pressure data channel, and millimeter-wave radar data from the millimeter-wave radar data channel; Correspondingly, determining the channel score value corresponding to each data channel based on the perceived data of each data channel includes: When the data channel includes an image data channel, the channel score value of the image data channel is determined based on the human morphological features represented by the image data. When the data channel includes an audio data channel, the channel score value of the audio data channel is determined based on at least one of the pitch features, timbre features, and volume features contained in the audio data. When the data channel includes a seat pressure data channel, the channel score value of the seat pressure data channel is determined based on the weight information represented by the rear seat pressure value. When the data channel includes a millimeter-wave radar data channel, the channel score value of the millimeter-wave radar data channel is determined based on the human motion characteristics represented by the millimeter-wave radar data.

[0070] In this embodiment of the application, the image data channel refers to the data channel of the in-vehicle camera, which is used to collect image data of the rear seats of the target vehicle.

[0071] In this embodiment of the application, the human morphological features refer to various features in the image data that can characterize the shape and proportion of the human body, mainly including facial features and head-to-body ratio. Facial features can cover dimensions such as cheekbones, chin, forehead, distribution of facial features, and skin texture. The head-to-body ratio is the proportional relationship between the head and the body. Different occupant states correspond to different human morphological features. For example, children have low cheekbones, short chins, and a head-to-body ratio greater than 1:7, while adults have prominent cheekbones, long chins, and a head-to-body ratio between 1:7.5 and 1:8.

[0072] In this embodiment of the application, the audio data channel refers to the data channel of the corresponding vehicle microphone, which is used to collect audio data inside the target vehicle.

[0073] In this embodiment, the pitch feature refers to the characteristic representing the highness or lowness of sound in audio data, determined by sound frequency. Different occupants exhibit significant differences in pitch characteristics; children's pitches are typically higher than 200Hz, while adults' pitches are typically lower than 150Hz. This feature can be used to distinguish whether rear-seat occupants are children or adults. The timbre feature refers to the characteristic representing the unique quality of sound in audio data, capable of distinguishing the differences in sound from different sources. Children's timbre is clear and delicate, while adults' timbre is deep and resonant. For example, timbre characteristics can distinguish whether the sound coming from the back seat is a child's delicate cry or an adult's deep voice. The volume feature refers to the characteristic representing the intensity of sound in audio data, measured in decibels (dB). The volume range of children's and adults' voices differs; children's volume is typically between 50-60dB, while adults' volume is typically between 60-70dB.

[0074] In this embodiment, the seat pressure data channel refers to the data channel of the corresponding seat pressure sensor, used to collect the pressure value of the rear seat of the target vehicle. The pressure value can characterize the weight information of the rear passenger. For example, the pressure data collected by the rear seat pressure sensor can be used to determine whether there is a rear passenger and whether the passenger's weight is within the typical weight range of a child.

[0075] In this embodiment, the millimeter-wave radar data channel refers to the data channel of the corresponding millimeter-wave radar, used to collect millimeter-wave radar data from the rear of the target vehicle. For example, data collected by the vehicle's rear millimeter-wave radar can detect whether there are people in the rear seats, whether the occupants have moved, and the distance to the vehicle doors.

[0076] In this embodiment, the human movement characteristics refer to occupant activity information acquired by millimeter-wave radar, including respiratory rate, limb movement amplitude, and movement trajectory. Children typically exhibit a higher respiratory rate (20-30 breaths / minute) and more unconscious limb movements.

[0077] In this embodiment, at least two types of data from image data channels, audio data channels, seat pressure data channels, and millimeter-wave radar data channels are acquired as sensing data. Based on the human morphological features, tone features, timbre features, volume features, weight information, and human motion features represented by the sensing data from each data channel, a channel score value is determined for each data channel. The human morphological features in the image data can intuitively distinguish between children and adults through facial features and head-to-body ratio. The tone, timbre, and volume features in the audio data can assist in determining the occupant's status through voice differences. The weight information from the seat pressure data can preliminarily distinguish between children and adults through weight ranges. The human motion features from the millimeter-wave radar data can assist in confirming whether there are people in the back seat and their activity status. The complementary combination of multi-channel data effectively avoids misjudgment problems caused by the perception bias of a single data channel, compensates for the limitations of single-channel feature recognition, and ensures that when the data from one channel is affected by environmental interference, other channels can still provide reliable information, significantly improving robustness and accuracy in complex environments.

[0078] Optionally, determining the channel score value of the image data channel based on the human morphological features represented by the image data includes: The image data is input into a pre-trained classification model to obtain the classification result output by the classification model; Based on the classification results, determine the channel score value of the image data channel; The classification model is trained on a training set that includes image samples of children, images of adults, and images of vehicles with no occupants in the back seat.

[0079] In this embodiment, the training set refers to a set of labeled images used to train the classification model, which may contain three types of samples: images of child occupants, images of adult occupants, and images of empty seats. This dataset can cover different lighting conditions, shooting angles, and occupant postures, such as including sample images from various scenes including daytime / nighttime, frontal / side views, and sitting / leaning postures.

[0080] In this embodiment, the child image samples refer to rear-seat vehicle image samples containing children in the training set, used to allow the classification model to learn the human morphological features of children. The adult image samples refer to rear-seat vehicle image samples containing adults in the training set, used to allow the classification model to learn the human morphological features of adults. In this embodiment, the image samples with no occupants in the rear seats refer to image samples in the training set where no one is sitting in the rear seats, used to allow the classification model to learn the rear-seat image features in an empty state, such as images where only a backpack is placed in the rear seat or images where the rear seat is completely empty.

[0081] In this embodiment, the classification model refers to a pre-trained multi-class recognition algorithm capable of classifying input vehicle rear-seat image data. This model establishes a mapping relationship from image pixel features to three categories: "child," "adult," and "unmanned," by learning the inherent features and patterns of a large number of labeled image samples in the training set. During the inference phase, the rear-seat image is input into this model, which automatically completes feature extraction and classification decisions. For example, this model can be an image classifier built on a convolutional neural network (CNN).

[0082] In this embodiment, the classification result refers to the output generated by the classification model after processing the input image data. This result can be presented in two forms: one is a probability vector containing the probability values ​​corresponding to each occupant category (child, adult, unoccupied); the other is the category label corresponding to the highest probability value.

[0083] In this embodiment, when the classification result is a probability vector, a fully connected layer can be added to the end of the model to map the probability vector into a one-dimensional continuous value, thereby directly obtaining the channel score value of the image data channel. This fully connected layer encodes the rule "higher score for children with higher probability, moderate score for adults with higher probability, and lower score for no one with higher probability" into the mapping relationship through learned weights.

[0084] In this embodiment, when the classification result is a single category label, a corresponding channel score value is assigned to each category according to a preset mapping rule. This rule strictly follows the principle of "child score value > adult score value > no score value", setting a fixed value or a range of values ​​for each category.

[0085] In this embodiment, a classification model trained on a large number of samples is introduced to process image data and output classification results, thereby determining channel scores. This approach utilizes machine learning methods, avoiding the limitations of traditional methods that rely on manually set thresholds and rules for feature judgment. It can more robustly handle changes in lighting, occupant posture, and shooting angle, directly learning and generalizing key visual features that distinguish children, adults, and vacant states from the data, thereby improving the accuracy and environmental adaptability of image channel scores.

[0086] Exemplary device The apparatus embodiments of this application can be used to execute the method embodiments of this application. For details not disclosed in the apparatus embodiments of this application, please refer to the method embodiments of this application.

[0087] Figure 2 The diagram shown is a block diagram of a vehicle control device according to an embodiment of this application. Figure 2 As shown, the device 200 includes: The data acquisition module 210 is used to acquire perception data from at least two data channels in the rear of the target vehicle. The channel scoring module 220 is used to determine a channel score value corresponding to each data channel based on the perception data of each data channel; wherein, the channel score value is used to characterize the probability that there is a child in the back seat of the target vehicle, and is related to the type of rear-seat occupant characterized by the corresponding perception data; for the same data channel, the channel score value corresponding to the rear-seat occupant type including children is higher than the channel score value corresponding to the rear-seat occupant type including only adults; the channel score value corresponding to the rear-seat occupant type including only adults is higher than the channel score value corresponding to the rear-seat occupant type being unoccupied. The target scoring module 230 is used to determine a target score based on the channel scoring values ​​of the at least two data channels; The instruction generation module 240 is used to generate a control instruction when the target score exceeds a preset threshold; the control instruction is used to control the target vehicle to perform child protection operations.

[0088] Optionally, the instruction generation module 240 is used to generate control instructions for controlling the child safety lock of the target vehicle based on the monitoring results of the driver's seat occupant of the target vehicle.

[0089] Optionally, the instruction generation module 240 is used for: When the child safety lock is in the closed state and the monitoring result indicates that there is an occupant in the driver's seat, a first control command is generated to open the child safety lock. When the child safety lock is in the open state and the monitoring result indicates that the occupant in the driver's seat intends to leave the vehicle, a second control command is generated; the second control command is used to alert the vehicle that a child is in the vehicle when the child safety lock is in the open state. If the monitoring results indicate that the driver occupant has left the target vehicle, a third control command is generated to deactivate the child safety lock.

[0090] Optionally, the target scoring module 230 is used for: Based on the current driving scenario, determine the weight value corresponding to each of the data channels; The target score is obtained by weighted summation of the channel score values ​​and corresponding weight values ​​of each data channel.

[0091] Optionally, the control command is used to activate the child safety lock of the target vehicle; the device 200 further includes: The threshold determination module is used to determine the preset threshold based on the current driving scenario of the target vehicle, wherein the value of the preset threshold is negatively correlated with the degree of door opening risk corresponding to the current driving scenario.

[0092] Optionally, the data acquisition module 210 is used to acquire at least two of the following: image data from the image data channel, audio data from the audio data channel, rear seat pressure values ​​from the seat pressure data channel, and millimeter-wave radar data from the millimeter-wave radar data channel. Correspondingly, the channel scoring module 220 includes: An image scoring unit is used to determine a channel score value for an image data channel based on the human morphological features represented by the image data when the data channel includes an image data channel. An audio scoring unit is used to determine a channel score value for an audio data channel based on at least one of pitch features, timbre features, and volume features contained in the audio data, when the data channel includes an audio data channel. A pressure scoring unit is used to determine a channel score value for the seat pressure data channel based on the weight information represented by the rear seat pressure value, when the data channel includes a seat pressure data channel. A radar scoring unit is used to determine a channel score value for a millimeter-wave radar data channel based on the human motion characteristics represented by the millimeter-wave radar data, when the data channel includes a millimeter-wave radar data channel.

[0093] Optionally, the image scoring unit is used for: The image data is input into a pre-trained classification model to obtain the classification result output by the classification model corresponding to the image data; Based on the classification results, determine the channel score value of the image data channel; The classification model is trained on a training set that includes image samples of children, images of adults, and images of vehicles with no occupants in the back seat.

[0094] Exemplary electronic devices Below, for reference Figure 3 This describes an electronic device according to embodiments of the present application. Figure 3 A block diagram of an electronic device according to an embodiment of this application is illustrated.

[0095] like Figure 3 As shown, the electronic device 300 includes one or more processors 310 and memory 320.

[0096] The processor 310 may be another form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 300 to perform desired functions.

[0097] Specifically, processor 310 can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present invention. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. Processor 310 may also include a main processor, and may also include a baseband chip, a modem, etc.

[0098] The memory 320 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 310 may execute the program instructions to implement the vehicle control methods of the various embodiments of this application described above and / or other desired functions. Various contents, such as category correspondences, may also be stored in the computer-readable storage medium.

[0099] In one example, the electronic device 300 may also include an input device 330 and an output device 340, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).

[0100] In addition, the input device 330 can also be a device that receives user input data and information, such as a keyboard, mouse, camera, scanner, light pen, voice input device, touch screen, pedometer, or gravity sensor. The output device 340 can output various information to the outside. The output device 340 may include, for example, a display, speaker, printer, and communication network and its connected remote output devices.

[0101] Of course, for the sake of simplicity, Figure 3 Only some of the components of the electronic device 300 relevant to this application are shown in this illustration; components such as buses, input / output interfaces, etc., are omitted. In addition, the electronic device 300 may include any other suitable components depending on the specific application.

[0102] Exemplary vehicle In addition to the methods and devices described above, embodiments of this application may also include a vehicle, comprising a vehicle body and the electronic equipment.

[0103] Exemplary computer program products and computer-readable storage media In addition to the methods and devices described above, embodiments of this application may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the vehicle control methods according to various embodiments of this application as described in the "Exemplary Methods" section of this specification.

[0104] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0105] Furthermore, embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the vehicle control methods according to various embodiments of this application described in the "Exemplary Methods" section above.

[0106] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0107] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0108] For the foregoing method embodiments, in order to simplify the description, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0109] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For apparatus embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0110] The steps in the methods of the various embodiments of this application can be adjusted, merged, or deleted in order according to actual needs, and the technical features described in each embodiment can be replaced or combined.

[0111] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0112] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0113] The modules or submodules described as separate components may or may not be physically separate. The components that constitute a module or submodule may or may not be physical modules or submodules; that is, they may be located in one place or distributed across multiple network modules or submodules. Some or all of the modules or submodules can be selected to achieve the purpose of this embodiment's solution, depending on actual needs.

[0114] Furthermore, the functional modules or sub-modules in the various embodiments of this application can be integrated into one processing module, or each module or sub-module can exist physically separately, or two or more modules or sub-modules can be integrated into one module. The integrated modules or sub-modules described above can be implemented in hardware or in the form of software functional modules or sub-modules.

[0115] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0116] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software unit executed by a processor, or a combination of both. The software unit can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0117] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0118] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A vehicle control method, characterized in that, include: Acquire perception data from at least two data channels in the rear of the target vehicle; Based on the perception data of each data channel, a channel score value corresponding to each data channel is determined; wherein, the channel score value is used to characterize the probability that there is a child in the back seat of the target vehicle, and is related to the type of rear-seat occupant characterized by the corresponding perception data; for the same data channel, the channel score value corresponding to the rear-seat occupant type including children is higher than the channel score value corresponding to the rear-seat occupant type including only adults; the channel score value corresponding to the rear-seat occupant type including only adults is higher than the channel score value corresponding to the rear-seat occupant type being unoccupied; The target score is determined based on the channel score values ​​of the at least two data channels; If the target score exceeds a preset threshold, a control command is generated to control the target vehicle to perform child protection operations.

2. The method according to claim 1, characterized in that, The generation of control instructions for controlling the target vehicle to perform child protection operations includes: Based on the monitoring results of the occupant in the driver's seat of the target vehicle, a control command is generated to control the child safety lock of the target vehicle.

3. The method according to claim 2, characterized in that, The step of generating control commands for the child safety locks of the target vehicle based on monitoring results of the driver's seat occupant includes at least one of the following: When the child safety lock is in the closed state and the monitoring result indicates that there is an occupant in the driver's seat, a first control command is generated to open the child safety lock. When the child safety lock is in the open state and the monitoring result indicates that the occupant in the driver's seat intends to leave the vehicle, a second control command is generated; the second control command is used to alert the vehicle that a child is in the vehicle when the child safety lock is in the open state. If the monitoring results indicate that the driver occupant has left the target vehicle, a third control command is generated to deactivate the child safety lock.

4. The method according to claim 1, characterized in that, Determining the target score based on the channel score values ​​of the at least two data channels includes: Based on the current driving scenario, determine the weight value corresponding to each of the data channels; The target score is obtained by weighted summation of the channel score values ​​and corresponding weight values ​​of each data channel.

5. The method according to claim 1, characterized in that, The control command is used to activate the child safety lock of the target vehicle; Before generating the control instructions for controlling the target vehicle to perform child protection operations, the method further includes: The preset threshold is determined based on the current driving scenario of the target vehicle, wherein the value of the preset threshold is negatively correlated with the degree of door opening risk corresponding to the current driving scenario.

6. The method according to claim 1, characterized in that, The acquisition of perception data from at least two data channels targeting the rear seats of the target vehicle includes: Acquire at least two of the following: image data from the image data channel, audio data from the audio data channel, rear seat pressure values ​​from the seat pressure data channel, and millimeter-wave radar data from the millimeter-wave radar data channel; Correspondingly, determining the channel score value corresponding to each data channel based on the perceived data of each data channel includes: When the data channel includes an image data channel, the channel score value of the image data channel is determined based on the human morphological features represented by the image data. When the data channel includes an audio data channel, the channel score value of the audio data channel is determined based on at least one of the pitch features, timbre features, and volume features contained in the audio data. When the data channel includes a seat pressure data channel, the channel score value of the seat pressure data channel is determined based on the weight information represented by the rear seat pressure value. When the data channel includes a millimeter-wave radar data channel, the channel score value of the millimeter-wave radar data channel is determined based on the human motion characteristics represented by the millimeter-wave radar data.

7. The method according to claim 6, characterized in that, The process of determining the channel score value of the image data channel based on the human morphological features represented by the image data includes: The image data is input into a pre-trained classification model to obtain the classification result output by the classification model corresponding to the image data; Based on the classification results, determine the channel score value of the image data channel; The classification model is trained on a training set that includes image samples of children, images of adults, and images of vehicles with no occupants in the back seat.

8. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is used to perform the method according to any one of claims 1 to 7.

9. A vehicle, characterized in that, Including the electronic device as described in claim 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1 to 7.

Images