Child occupant sensing method, apparatus, child seat, and computer-readable storage medium

Abstract

Embodiments of the present application provide a child passenger sensing method and device, a child seat and a computer readable storage medium. The child seat is provided with a UWB radar device and a data acquisition device, and stores state analysis rules corresponding to each assembly state, wherein the assembly state represents the installation form of the child seat and the age group of the child passenger. In the case that the child seat is effectively seated by the child passenger, echo data sent by the UWB radar device and monitoring data collected by the data acquisition device are obtained. The assembly state of the child seat is determined according to the echo data and the monitoring data, and the three-dimensional coordinate information of the child passenger is generated based on the echo data. The target state analysis rule corresponding to the child passenger is determined according to the assembly state. The passenger state of the child passenger is determined based on the target state analysis rule, the echo data and the three-dimensional coordinate information. In this way, the problem of missed detection and false detection can be avoided, the reliability of child passenger sensing can be improved, and the safety risk can be reduced.

Description

Technical Field

[0001] This application relates to the field of child car seat technology, and more specifically, to a child occupant sensing method, device, child car seat, and computer-readable storage medium. Background Technology

[0002] Child seats are special restraint devices designed for and installed in the rear seats of cars for child occupants. As travel demand gradually increases, monitoring the status of child occupants in child seats is becoming an important part of ensuring child safety in cars.

[0003] In related technologies, vehicle-mounted radar or cameras are generally used to detect the status of child occupants through general perception algorithms or image recognition algorithms. However, this approach often has several problems. Specifically, using general perception algorithms for child occupant status detection often fails to meet the detection needs of children of different age groups, leading to problems such as missed detections and false detections. This reduces the reliability of child occupant detection and increases safety risks. On the other hand, image recognition algorithms are often more complex, require higher onboard computing power, and have higher application costs. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a child occupant sensing method, device, child seat, and computer-readable storage medium to avoid missed detections and false detections, improve the reliability of child occupant sensing, and reduce safety risks.

[0005] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows: Firstly, this application provides a child occupant perception method applied to a child car seat. The child car seat is equipped with a UWB radar device and a data acquisition device. The child car seat stores state analysis rules corresponding to each assembly state, and the assembly state represents the installation form of the child car seat and the age group of the child occupant. The method includes: When the child occupant is effectively seated in the child seat, the echo data sent by the UWB radar device and the monitoring data collected by the data acquisition device are acquired. The assembly status of the child seat is determined based on the echo data and the monitoring data, and the three-dimensional coordinate information of the child occupant is generated based on the echo data. Based on the assembly state, determine the target state analysis rules corresponding to the child occupant; Based on the target state analysis rules, the occupant state of the child occupant is determined according to the echo data and the three-dimensional coordinate information.

[0006] In an optional implementation, the data acquisition device includes an installation information acquisition device, a seat cushion pressure acquisition device, and a seat belt status acquisition device; the method further includes: The installation data, pressure data, seat belt data, and echo data from the installation information acquisition device, the seat cushion pressure acquisition device, and the seat belt status acquisition device, respectively, are acquired. If the installation data indicates that the child seat is installed in place, the pressure data is within the preset pressure range, the seat belt data indicates that the seat belt buckle is closed, and the echo data indicates that there is a living person on the child seat, then it is determined that the child seat has been effectively occupied by the child occupant.

[0007] In an optional implementation, the data acquisition device further includes a rotation positioning detection device, wherein determining the assembly status of the child seat based on the echo data and the monitoring data includes: The installation configuration of the child seat is determined based on the rotation positioning data sent by the rotation positioning detection device, and the age group of the child occupant is determined based on the echo data and the pressure data. The assembly status of the child seat is determined based on the installation configuration and the age group.

[0008] In an optional implementation, the UWB radar device includes multiple components, and the step of generating the three-dimensional coordinate information of the child occupant based on the echo data includes: Based on the pre-constructed electronic fence, the echo data of each of the UWB radar devices is spatially aligned and boundary-matched for filtering to obtain target echo data; The target echo data is processed using weighted least squares and extended Kalman filtering to obtain the three-dimensional coordinate information of the child occupant.

[0009] In an optional implementation, the UWB radar device includes a master UWB node and multiple slave UWB nodes. The master UWB node is connected to each of the slave UWB nodes via a CAN bus. The master UWB node is located at the headrest of the child seat, and the slave UWB nodes are respectively located at the backrest, base, chest buckle, and buckle of the child seat.

[0010] In an optional embodiment, the child seat is also communicatively connected to the vehicle's infotainment system, and the child seat is also equipped with a backup battery. The method further includes: Obtain the vehicle status sent by the vehicle system; When the vehicle is in a normal driving state, the master UWB node, each slave UWB node and each data acquisition device are controlled to operate at a first sampling frequency. When the vehicle is in a standby state with the engine off, control the main UWB node and the target data acquisition device to operate at the second sampling frequency; When the vehicle is in a locked sleep state, the backup battery is activated to power the child seat, and the main UWB node and the target data acquisition device are controlled to operate at the third sampling frequency. Wherein, the first sampling frequency is greater than the second sampling frequency, and the second sampling frequency is greater than the third sampling frequency.

[0011] In an optional implementation, the method further includes: When the vehicle is in a standby or locked-out state, the system determines whether the child occupant has not left the child seat based on the echo data of the main UWB node and the monitoring data of the target data acquisition device. If the child occupant does not leave the child seat, the master UWB node, each slave UWB node, and each data acquisition device are controlled to operate at a first sampling frequency and a child abandonment alarm is sent.

[0012] In an optional implementation, the target state analysis rule includes target filtering parameters, a target feature extraction algorithm, and a target state recognition model; the step of determining the occupant state of the child occupant based on the target state analysis rule, according to the echo data and the three-dimensional coordinate information, includes: The echo data is filtered based on the target filtering parameters, and the child's state features in the echo data are extracted based on the target feature extraction algorithm. The child's state features and the three-dimensional coordinate information are input into the target state recognition model for processing to obtain the occupant state of the child occupant; the occupant state includes the child occupant's action type and physiological state; The method further includes: Based on the danger level corresponding to the action type and the physiological state, a danger alarm message is output.

[0013] Secondly, this application provides a child occupant sensing device applied to a child seat. The child seat is equipped with a UWB radar device and a data acquisition device. The child seat stores state analysis rules corresponding to each assembly state, and the assembly state characterizes the installation form of the child seat and the age group of the child occupant. The device includes: The acquisition module is used to acquire echo data sent by the UWB radar device and monitoring data collected by the data acquisition device when the child seat is correctly installed and the child occupant is effectively seated. The determination module is used to determine the assembly status of the child seat based on the echo data and the monitoring data, and to generate the three-dimensional coordinate information of the child occupant based on the echo data; The determining module is further configured to determine the target state analysis rule corresponding to the child occupant based on the assembly state; The determining module is further configured to determine the occupant status of the child occupant based on the target status analysis rules, the echo data, and the three-dimensional coordinate information.

[0014] Thirdly, this application provides a child seat, including a main control unit, which is used to execute a computer program to implement the method described in any of the foregoing embodiments.

[0015] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a main control unit, implements the method described in any one of claims 1-8.

[0016] The child occupant sensing method, device, child seat, and computer-readable storage medium provided in this application embodiment include a child seat equipped with a UWB radar device and a data acquisition device, which stores state analysis rules corresponding to each assembly state. Based on this, the child seat can determine its own assembly state according to the echo data collected by the UWB radar device and the monitoring data collected by the data acquisition device, and generate three-dimensional coordinate information of the child occupant based on the echo data. Furthermore, it determines the target state analysis rule corresponding to the child occupant based on the assembly state, and analyzes the occupant's state by combining the echo data and three-dimensional coordinate information with this target state analysis rule. This allows for accurate identification of the current actual assembly state. Since the assembly state can characterize the installation form of the child seat and the age group of the child occupant, this method can select state analysis rules suitable for different age groups and different child seat installation situations, thereby avoiding missed detections and false detections, improving the reliability of child occupant sensing, and reducing safety risks. In addition, since visual analysis is not required, high onboard computing power is not needed, resulting in lower application costs and easier promotion and use.

[0017] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A block diagram of a child seat provided in an embodiment of this application is shown; Figure 2 This paper illustrates a flowchart of a child occupant perception method provided in an embodiment of this application. Figure 3 This illustration shows another flowchart of the child occupant perception method provided in an embodiment of this application; Figure 4 A functional block diagram of a child occupant sensing device provided in an embodiment of this application is shown. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can be arranged and designed in various different configurations.

[0021] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0022] It should be noted that relational terms such as "first" and "second" are used merely 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 process, method, article, or apparatus. Unless otherwise specified, 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 the element.

[0023] Child seats refer to specialized restraint devices designed for and installed in the rear seats of cars for child occupants. Child occupant sensing is used to detect in real time whether a child is in a child seat, whether they are sitting correctly, whether they are maintaining a safe posture, and whether there are any abnormal vital signs or dangerous movements. It can also trigger warnings or protective responses when necessary. Its core significance is to improve the safety of children throughout the entire process of traveling in a car, especially when the driver cannot directly observe the rear seats. It enables objective, continuous, and reliable monitoring of the child's status, thereby effectively preventing safety accidents caused by forgetting, accidental movement, obstruction, or equipment failure.

[0024] Two main approaches are used in related technologies to achieve child occupant perception: one is a radar-based perception approach, and the other is a camera-based visual perception approach.

[0025] In radar-based perception solutions, child occupant detection typically relies on a single radar module, usually installed in the center of the front of the vehicle, such as above the dashboard or the top of the front seat back, with its sensing range covering the entire rear area. This arrangement presents significant technical challenges for child occupant detection: when the child seat is installed rear-facing in a newborn configuration (Group 0), the newborn's chest is completely facing away from the single radar module in the center of the front. The electromagnetic waves emitted by this single radar module must penetrate the thick foam layer of the child seat, the soft tissue of the newborn's back, and any thick swaddling clothes that may be present before detection can be achieved. This results in severe signal attenuation and phase distortion, significantly reducing the accuracy of extracting subtle physiological features such as respiratory rhythm based on echo signals. Furthermore, when a child changes position in the seat, such as rolling over, lying on their side, or curling up, the single radar module, due to its fixed viewing angle and lack of spatial redundancy, is prone to losing target signals or blurring reflection characteristics, leading to the risk of missed detections.

[0026] In camera-based visual perception solutions, child occupant detection relies on a single camera installed in the center of the front row of the vehicle. This single camera captures images and runs image recognition algorithms to perceive the child occupant's status. However, this single camera also has significant technical problems: because the camera is installed far from the child seat and its optical field of view is obstructed by the vehicle's interior structure, it cannot obtain an effective imaging angle of the newborn's face or torso for the rear-facing newborn (Group 0), making recognition extremely difficult. In addition, the image recognition algorithm associated with this single camera is highly complex, requiring a high-performance onboard processor, which increases the cost of the entire vehicle's electronic system and hinders large-scale mass production applications. More importantly, this single camera has extremely poor resistance to obstruction. Once the child is covered by a thick blanket, swaddle, clothing, or other foreign object, the single camera loses effective visual information input, easily leading to false alarms or missed alarms, and failing to meet the all-weather stable operation requirements for child occupant detection.

[0027] In addition to the issues mentioned above, related technologies also suffer from a general lack of universality at the algorithm level. Current mainstream UWB perception algorithms are developed for adult or general occupant scenarios. However, child seats need to be adapted to children of different ages and various installation configurations in actual use. For example, Group 0 is rear-facing for newborns, Group 1 is forward-facing for toddlers, and Groups 2 and 3 are forward-facing for older children with booster seats. These differences directly lead to variations in the child's body position, breathing amplitude, movement characteristics, and even the spatial response characteristics of the UWB radar echo. Therefore, using a uniform data processing logic will make it difficult to simultaneously address the weak vital signs of newborns and the large limb movements required by older children. This results in insufficient sensitivity for detecting vital signs in newborns and poor robustness in recognizing dangerous movements in older children, ultimately leading to increased false alarm and false negative rates, making it difficult to support the reliable operation of child restraint systems in real-world vehicle scenarios. Furthermore, the complexity of image recognition algorithms and their high overall requirements for onboard computing power also result in high application costs and limited scalability.

[0028] Based on this, embodiments of this application provide a child occupant sensing method, device, child seat, and computer-readable storage medium to solve the above-mentioned problems. Figure 1 For a block diagram of the child seat provided in the embodiments of this application, please refer to... Figure 1 The child seat includes a main control unit, a UWB radar device, and a data acquisition device, and the main control unit is communicatively connected to the UWB radar device and the data acquisition device respectively.

[0029] In this embodiment, the UWB (Ultra-Wideband) radar device can transmit narrow pulse signals toward the child's seating area and receive echo signals reflected by the child's body, thereby generating echo data that can reflect the child's movements and vital signs; the data acquisition device can collect environmental information and information related to the child seat.

[0030] Installation type refers to the physical installation method of child seats in vehicles, specifically including two basic types: forward-facing installation and rear-facing installation; age group refers to the four occupant groups divided according to GB27887 standard: group 0 (newborns), group 1 (toddlers), group 2 (preschool children), and group 3 (older children).

[0031] In this embodiment, the main control unit can acquire various data sent by the UWB radar equipment and data acquisition equipment, and implement the child occupant perception method provided in this application embodiment by executing a computer program. Understandably, the main control unit must meet the ISO 26262 ASIL-B functional safety level requirements and possess safety mechanisms such as a hardware safety island to ensure the security of in-vehicle data. The main control unit is installed in the rear enclosed cavity of the child seat base. This location was chosen after considering both structural safety and functional implementation: on the one hand, this location is far from the child's seating area, avoiding direct contact risk to the child under normal use or collision conditions; on the other hand, this location also avoids the force path where the child seat mainly bears the impact load in a frontal collision, thus ensuring that the main control unit will not fail or shift due to structural deformation during a vehicle collision. All wiring harnesses connecting to the main control unit are pre-embedded inside the child seat frame, maintaining the integrity of the seat's appearance and preventing exposed wiring harnesses from being worn, snagged, or accidentally pulled and broken in a collision. It also completely avoids any physical connection or electrical coupling between the vehicle's main control unit and the original vehicle wiring.

[0032] In one possible implementation, the child seat's main structure may also include an adjustable headrest, a high-density EPS (Expanded Polystyrene) cushioned backrest, a base, an ISOFIX rigid connection mechanism, an upper tether strap, and a five-point harness system (consisting of shoulder straps, chest buckle, crotch strap, and bottom buckle). The overall structure meets the requirements for force transmission and energy absorption under vehicle collision conditions. The ISOFIX rigid connection mechanism is used to rigidly fix the child seat to the vehicle's seat anchor points, the upper tether strap is used to inhibit the seat from tilting forward and rolling during a collision, and the five-point harness system distributes the force on the child's torso through multi-directional restraint.

[0033] It should be understood that, Figure 1 The structure shown is only a schematic diagram of a child car seat; child car seats may also include components larger than those shown. Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown. Figure 1 The components shown can be implemented using hardware, software, or a combination thereof.

[0034] The following is based on the above. Figure 1 The child seat in this application is the main implementer. The child occupant perception method provided in this application embodiment is described by way of example in conjunction with the flowchart.

[0035] Specifically, Figure 2 Please refer to the flowchart of a child occupant perception method provided in this application embodiment. Figure 2 The method includes: Step S20: With the child occupant effectively seated in the child seat, acquire the echo data sent by the UWB radar device and the monitoring data collected by the data acquisition device.

[0036] In this embodiment, before activating the sensing function, the child seat needs to first confirm that it is in a valid working condition, that is, it is securely installed in the vehicle and a child occupant is actually sitting in it. Under this condition, the echo data sent by the UWB radar device and the monitoring data collected by the data acquisition device can be obtained in real time, and the status of the child occupant can be sensed based on the echo data and the monitoring data.

[0037] Among them, UWB radar equipment can capture minute body surface displacements of child occupants by emitting nanosecond-level extremely narrow pulse signals and receiving the echoes reflected from the human body. The time difference information contained in the echo data can be used to calculate distance, amplitude changes can reflect limb obstruction or posture changes, and phase changes directly correspond to the periodic fluctuations of the chest cavity caused by breathing and heartbeat.

[0038] For example, the frequency of chest cavity movements caused by breathing is approximately 20 to 50 times per minute, or 0.3 to 0.8 Hz, while the frequency of heartbeats is approximately 60 to 150 times per minute, or 1.0 to 2.5 Hz. Specifically, the respiratory rate of newborns under 28 days old is approximately 40 to 50 breaths per minute; the respiratory rate of infants under one year old is approximately 30 to 40 breaths per minute; the respiratory rate of toddlers from one year old to preschool age is approximately 20 to 35 breaths per minute; and the respiratory rate of toddlers from primary school age to 18 years old is approximately 20 to 25 breaths per minute. These physiological rhythms leave identifiable periodic disturbances in the phase of the echo signal. Therefore, echo data can reflect not only the types of movements of child occupants, such as rolling over, climbing, and sliding, but also basic vital signs such as respiratory rate and heart rate.

[0039] The monitoring data can reflect information related to the use of child seats, such as whether the child seat is installed properly, the pressure of the seat cushion, and the ambient temperature.

[0040] Step S21: Determine the assembly status of the child seat based on the echo data and monitoring data, and generate the three-dimensional coordinate information of the child occupant based on the echo data.

[0041] Step S22: Determine the target state analysis rules corresponding to the child occupants based on the assembly state.

[0042] Step S23: Based on the target state analysis rules, determine the occupant state of the child occupant according to the echo data and three-dimensional coordinate information.

[0043] In this embodiment, the assembly state refers to the combined state of the child seat's current installation form and the child occupant's age group. The installation form can be interpreted as the orientation of the child seat relative to the vehicle, i.e., forward-facing or rear-facing installation. The age group is divided according to the GB27887 standard, including group 0 (newborns), group 1 (toddlers), group 2, and group 3 (older children). The three-dimensional coordinate information reflects the child occupant's real-time position.

[0044] In this embodiment, the child seat can determine the current assembly status through echo data and monitoring data, and retrieve the corresponding target status analysis rule from the pre-stored status analysis rule library. Based on the target status analysis rule, the child seat can collaboratively process UWB echo data and three-dimensional coordinate information to finally output the occupant status of the child occupant.

[0045] In one possible implementation, the occupant status can include two dimensions: the child occupant's actions and physiological state, such as "sitting upright but with the seatbelt loose" or "rear-facing with a sudden drop in respiratory rate."

[0046] The child occupant perception method provided in this application embodiment includes a child seat equipped with a UWB radar device and a data acquisition device, which stores state analysis rules corresponding to each assembly state. Based on this, the child seat can determine its own assembly state according to the echo data collected by the UWB radar device and the monitoring data collected by the data acquisition device, and generate the three-dimensional coordinate information of the child occupant based on the echo data. Furthermore, it determines the target state analysis rule corresponding to the child occupant based on the assembly state, and analyzes the occupant's state by combining the echo data and the three-dimensional coordinate information with this target state analysis rule. This allows for accurate identification of the current actual assembly state. Since the assembly state can characterize the installation form of the child seat and the age group of the child occupant, this method can select state analysis rules suitable for different age groups and different child seat installation situations, thereby avoiding missed detections and false detections, improving the reliability of child occupant perception, and reducing safety risks. In addition, since visual analysis is not required, high onboard computing power is not needed, resulting in lower application costs and easier promotion and use. In addition, the UWB radar device is installed inside the child seat, so it can be integrated with the child seat's collision safety structure, avoiding the problem of easy detachment that exists with externally attached UWB devices, preventing secondary injuries in the event of a collision, and improving the safety of the child seat.

[0047] Considering that related technologies generally use a single UWB radar device to detect the presence of objects on child seats, but this method often cannot distinguish between child occupants and inanimate objects such as toys, backpacks, and pets, leading to false alarms for objects or false triggering of empty seats, this embodiment can also solve the above problems by combining multiple monitoring data for judgment.

[0048] It should be noted that the relevant solutions usually rely solely on the presence of strong reflection characteristics in the UWB echo signal to determine whether there is a child in the seat. However, in daily use, child seats are often accompanied by plush toys, folded clothes, backpacks, or even pets that briefly stay in the seat. These objects may also exhibit similar micro-movements or breathing frequency phase disturbances under UWB detection, leading to misjudgment that there is a child in the seat, thus affecting the user experience.

[0049] Based on this, this embodiment can be equipped with multiple data acquisition devices to combine echo data from these devices and UWB radar equipment to jointly determine whether the child seat is correctly installed and whether the child occupant is effectively seated. In one possible implementation, the data acquisition devices may include installation information acquisition devices, seat cushion pressure acquisition devices, and seat belt status acquisition devices. Figure 2 On this basis, Figure 3 For another flowchart illustrating the child occupant perception method provided in this application embodiment, please refer to [link / reference]. Figure 3 The method also includes: Step S10: Acquire installation data, pressure data, seat belt data, and echo data from the installation information acquisition device, seat cushion pressure acquisition device, and seat belt status acquisition device, respectively.

[0050] Step S11: If the installation data indicates that the child seat is installed in place, the pressure data is within the preset pressure range, the seat belt data indicates that the seat belt buckle is closed, and the echo data indicates that there is a living person on the child seat, then it is determined that the child seat has been effectively occupied by the child occupant.

[0051] In this embodiment, the installation information acquisition device can specifically be an ISOFIX installation position sensor integrated inside the ISOFIX rigid interface of the seat. This sensor is mechanically linked with the locking mechanism to detect in real time whether the ISOFIX interface is fully engaged with the original vehicle anchor point. The seat cushion pressure acquisition device is specifically a distributed seat cushion pressure sensor array installed under the surface fabric of the seat cushion in the seat base. It is composed of pressure sensors and is used to collect the pressure distribution and total pressure value applied to the seat cushion when the child is seated. The seat belt status acquisition device is specifically a seat belt buckle closure sensor integrated inside the bottom buckle and chest buckle of the five-point seat belt to detect whether the buckle and chest buckle have completed the physical closure action.

[0052] Optionally, the effective seating of a child in a child seat includes two aspects: first, the child seat is correctly installed; and second, there is a living child occupant safely seated.

[0053] During this process, the child seat can acquire installation data, pressure data, and seat belt data sent by the three types of devices mentioned above, and simultaneously acquire echo data from the UWB radar device. Among them, the installation data indicates that the ISOFIX interface is fully locked, that is, the child seat is correctly installed; the pressure data is within the preset pressure range, which corresponds to the typical weight range of children in groups 0 to 3 in the GB27887 standard, to exclude accidental pressure caused by non-living objects such as dolls and backpacks; the seat belt data indicates that the seat belt buckle and chest buckle are fully closed, indicating that the child has been basically restrained and is safely seated; and the echo data needs to be preliminarily calculated to confirm the presence of breathing micro-movements and body surface undulation signals that conform to the characteristics of a living person, thus indicating that there is indeed a child occupant in the seat.

[0054] In one possible implementation, after the child seat is installed, secured, and powered on, it first checks whether its power supply is stable. If the power supply is stable, it obtains the aforementioned parameters for judgment. If the child seat is not installed correctly, or is accidentally touched by other objects, or the seat belt is not properly fastened, it can send a prompt message, such as outputting a prompt voice, so that the user can check based on the prompt message.

[0055] In this embodiment, effective location judgment can be made by using information sent from components from different acquisition devices and different installation locations. Therefore, the false alarm rate caused by single-point failure or external interference can be controlled at an extremely low level.

[0056] In this way, the child car seat can autonomously identify typical abnormal scenarios such as an empty seat, incorrect installation, and no seat belt. It will only activate the subsequent perception and analysis process when there is a real monitoring foundation, which not only ensures the rigor of the perception logic, but also improves the reliability of daily use for users.

[0057] The following section provides a possible approach to determining the assembly status of a child car seat based on echo data and monitoring data.

[0058] In this embodiment, the data acquisition device also includes a rotation position detection device. The child seat can also determine the installation form of the child seat based on the rotation position data sent by the rotation position detection device, and determine the age group of the child occupant based on the echo data and pressure data, thereby determining the assembly status of the child seat based on the installation form and age group.

[0059] In this embodiment, the installation configuration of the child seat can be determined by a built-in rotation positioning detection device. In one possible implementation, a mechanical pressure switch can be installed at the latch position of the seat rotation mechanism. When the seat is installed in either the forward or reverse orientation and locked in place, the latch triggers the pressure switch to conduct, thereby generating rotation positioning data characterizing the forward or reverse installation.

[0060] Meanwhile, the child car seat can collect echo data from UWB radar equipment and determine the age group based on the echo data.

[0061] In one possible implementation, multiple UWB radar devices can be installed at the headrest, backrest, base, chest buckle, and buckle of the child seat. The child seat can obtain the UWB echo energy distribution of each UWB radar device, calculate the proportion of the echo intensity of each UWB radar device to the total echo intensity of all UWB radar devices, determine the current echo energy distribution characteristics based on the echo intensity proportion of each UWB radar device, and match it with the preset echo distribution templates corresponding to each age group to determine the age group of the child occupant.

[0062] The echo distribution templates for each age group are constructed based on the height ranges and corresponding riding morphological characteristics of children as specified in the new national standard GB 27887-2024. Specifically, they include three typical height ranges: newborns and infants (height 40cm~87cm), school children / toddlers (height 80cm~105cm), and older children (height 100cm~150cm).

[0063] Because newborns and infants are primarily in a lying position when rear-facing, with shorter torsos and larger heads, their echo energy is mainly concentrated in the UWB radar area at the backrest. School-aged children and toddlers are primarily in a sitting position, with moderate torso lengths, and their echo energy is relatively evenly distributed between the UWB radar at the headrest and the backrest. Older children, with increased height and a larger proportion of their upper body and head, show a significant increase in echo energy at the headrest, creating a more pronounced spatial distribution difference compared to school-aged children and toddlers. Therefore, the proportion of echo intensity presented by children of different heights at different locations on the UWB radar is significantly different, enabling accurate age group identification of children.

[0064] In addition, considering that there may be overlapping areas in the height of children at different stages, the child seat can also collect children's weight data through a distributed pressure sensor array on the seat cushion, and combine the weight range of each age group to further supplement the determination of the children's age group based on the proportion of echo intensity.

[0065] For example, when a child passenger's height falls within both the height range of newborns and infants and the height range of school children / toddlers, or both the height range of school children / toddlers and the height range of older children, height information alone cannot uniquely determine the applicable age group. In this case, the pressure data collected by the seat pressure acquisition device serves as a key supplementary criterion to further narrow down the range of group determination.

[0066] In one example, group 0 corresponds to a weight of no more than 10 kg, group 1 corresponds to a weight between 9 and 18 kg, group 2 corresponds to a weight between 15 and 25 kg, and group 3 corresponds to a weight between 22 and 36 kg.

[0067] It should be noted that children's weight data is only used to provide auxiliary decision support in scenarios where height ranges overlap. In this case, the age group determined based on echo intensity ratio and the age group determined based on weight can be weighted and fused according to preset height and weight weights to determine the final age group. In addition, if weight also falls into the overlapping area and cannot assist in the judgment, a conservative strategy can be adopted first. That is, the target state analysis rule corresponding to the age group with higher safety constraints and more timely alarm response can be selected, or a smaller age group can be selected to ensure that the child occupant's perception results always meet the functional safety redundancy requirements.

[0068] In this embodiment, the height weight and weight weight can be configured by the user in advance, or dynamically adjusted based on the data confidence of the seat pressure acquisition device and the UWB radar device. This application does not impose any further limitations on this.

[0069] Optionally, the data acquisition device may also include a seat NTC temperature sensor, which may be located in the headrest, for example, inside a UWB radar device in the headrest, facing into the vehicle cabin, to detect the ambient temperature inside the vehicle and determine whether there is a risk of heatstroke.

[0070] In one possible implementation, ventilation modules can be installed inside both the base and backrest of the child seat. When the child seat determines that the interior temperature exceeds the acceptable level based on temperature data sent by a temperature sensor, the ventilation modules can be activated to lower the temperature of the seat surface and prevent the child from being burned by high temperatures. Furthermore, the interior temperature can be monitored in real time in situations where the child is forgotten after the car is locked, so that an alarm can be issued promptly if the interior temperature becomes too high.

[0071] Understandably, due to the inherent limitations of deploying a single UWB radar device, the stability and accuracy of the perception of the child occupant's status are insufficient. Therefore, in this embodiment, multiple UWB radar devices can be set up.

[0072] Based on this, the child seat can spatially align and compare the echo data of each UWB radar device according to the pre-built electronic fence to obtain the target echo data; by processing the target echo data through weighted least squares method and extended Kalman filter, the three-dimensional coordinate information of the child occupant can be obtained.

[0073] In this embodiment, multiple UWB radar devices are deployed in different physical locations on the child seat using a distributed architecture of one master and multiple slaves. All nodes share the same 19.2MHz temperature-compensated crystal oscillator as the global clock source, thereby achieving sub-nanosecond time synchronization accuracy. This ensures that each node completes signal transmission and reception at the same time, avoiding echo data misalignment or fusion inaccuracy due to clock deviation. The master UWB node can periodically send synchronization frames with precise timestamps to each slave node via the CAN Lite bus. Each slave node can use this to calibrate the offset and frequency drift of its local clock in real time, forming a stable and reliable timing system.

[0074] Each slave node can send echo data to the master node, which then integrates the echo data from each slave node with its own echo data and sends the integrated echo data to the child seat. The child seat then performs spatial alignment of the echo data from each node.

[0075] In one possible implementation, spatial alignment can be achieved through pre-calibration and coordinate transformation. Specifically, during the production stage, high-precision calibration equipment can be used to determine the fixed spatial position of each UWB node in a unified seat coordinate system. This coordinate system has the center of the seat base as the origin, with the X-axis pointing to the left and right direction of the seat, the Y-axis pointing to the front and back direction, and the Z-axis pointing to the up and down direction. The raw echo data collected by each node is mapped to this global coordinate system according to the preset coordinate transformation matrix, thereby completing spatial alignment and ensuring the spatial consistency of echo data from different nodes.

[0076] In this embodiment, a static safety electronic fence can be defined in three-dimensional space based on the physical structural boundaries of the child seat itself. This fence is strictly limited to the effective seating area of ​​the seat. The child seat can project all the echo data that has been spatially aligned uniformly into the spatial range defined by the electronic fence, retaining only the effective echo signals falling inside the fence, while eliminating interference signals from outside the seat, front occupants, debris in the vehicle or other reflectors. This allows for boundary comparison and filtering of the echo data to obtain the target echo data.

[0077] Based on this, child seats can process target echo data using weighted least squares and extended Kalman filtering to obtain the three-dimensional coordinate information of the child occupant.

[0078] In one possible implementation, the child seat can use a weighted least squares method combined with an extended Kalman filter to fuse and solve the target echo data, thereby obtaining the child occupant's three-dimensional coordinate information. Understandably, this three-dimensional coordinate information can represent the child's spatial position within the seat in real time and its trajectory over time.

[0079] In one possible implementation, the UWB radar device may include a master UWB node and multiple slave UWB nodes. The master UWB node and each slave UWB node are connected via a CAN bus. The master UWB node is located at the headrest of the child seat, and the slave UWB nodes are located at the backrest, base, chest buckle, and seat belt buckle of the child seat, respectively.

[0080] In this embodiment, the main UWB node refers to the UWB radar device integrated into the EPS buffer cavity inside the child seat headrest. Its antenna faces the seating area and is positioned to avoid the force-bearing area that the child's head might come into contact with during a collision. This node uses a 24mm × 18mm directional planar antenna with a horizontal beamwidth of 60 degrees and a vertical beamwidth of 40 degrees. When installed forward-facing, this node can detect the child's head and upper body movements, as well as vital signs such as breathing and heartbeat, and can identify dangerous actions such as standing and climbing. It is suitable for all scenarios except for the reverse-facing installation scenario (group 0). Understandably, in a multi-UWB radar device setup, this node acts as the array's main controller, responsible for timing control and echo data aggregation of all slave UWB nodes.

[0081] The UWB node comprises four independent units. One of them is located in the gap between the frame in the lower middle part of the seat back, with the antenna facing the seating area. It is wrapped in a knitted fabric and does not directly contact the high-density energy-absorbing foam inside the backrest. It uses a wide-beam omnidirectional PCB antenna with a beamwidth of 120 degrees. It is mainly used to detect respiratory micro-movements from the back of a newborn in the 0-group reverse installation scenario, and to assist in recognizing changes in body position such as turning over, lying on the side, or curling up.

[0082] Secondly, it is located in a closed cavity below the lap area of ​​the seat base and above the ISOFIX interface. The antenna surface is vertically upward and it uses a narrow beam directional antenna with a beamwidth of only 30 degrees. The coverage area is strictly limited to the seat area. It is used to detect abnormal movements of the child's lower body, loose lap belts, or body slippage. At the same time, it can detect breathing signals from bottom to top when covered by a thick blanket. This node can supplement the blind spots of the headrest / backrest node.

[0083] Thirdly, it is integrated into the five-point safety belt chest buckle shell and is injection molded as a whole with the chest buckle structure. It adopts a miniaturized flexible PCB antenna, which can report the precise three-dimensional coordinates of the child's upper body in real time. It is used to identify behaviors such as touching the chest buckle, unfastening the chest buckle without permission, and abnormal sitting posture. It is suitable for one set of forward-facing installation for toddlers.

[0084] Fourthly, it is integrated into the bottom buckle of the five-point seat belt. It also uses a miniaturized omnidirectional antenna and completes integrated injection molding. It mainly works with the seat belt buckle closure sensor to achieve dual verification, while continuously reporting the child's torso reference coordinates to help determine whether the child has been taken out of the seat or left in the car.

[0085] In practical applications, all UWB nodes utilize the NXP DW3110 Ultra-Wideband (UWB) chip. This chip meets automotive-grade requirements, operates in a temperature range of -40°C to 105°C, uses the nationally approved 7.25 GHz to 8.75 GHz frequency band with a bandwidth of 1.5 GHz, achieves a distance resolution of 10 cm, and supports both positioning tag and impulse radar operating modes. It is feasible for mass production and cost control. The master UWB node at the headrest is connected to the main control unit inside the child seat via an SPI (Serial Peripheral Interface) bus. The other four slave UWB nodes communicate with the master node via a CAN Lite bus. All wiring harnesses are pre-embedded within the seat frame, neither exposed nor affecting seat adjustment and installation, while also meeting the vehicle's electromagnetic compatibility (EMC) requirements.

[0086] Furthermore, the single UWB radar solution used in these technologies suffers from a trade-off between power consumption and battery life. Specifically, the UWB radar in these technologies is often installed in the center of the front seat or mounted on the headrest, relying solely on the vehicle's cigarette lighter or the original vehicle bus for power. Once the vehicle is turned off, the power supply is lost, and the UWB radar immediately stops working. Even in some solutions equipped with backup power, all sensing modules maintain the same high-frequency sampling and end-to-end data processing even after the engine is off, resulting in persistently high power consumption. Backup batteries often run out within hours, making it impossible to support continuous monitoring for extended periods, such as more than 72 hours. Simultaneously, due to the lack of vehicle status recognition capabilities, child seats cannot distinguish between a temporarily off vehicle and a locked vehicle, nor can they dynamically adjust their power consumption accordingly. Consequently, in scenarios where prolonged monitoring is required and a child is left behind, there is neither continuous monitoring capability nor an effective wake-up mechanism, rendering the system passively ineffective.

[0087] Based on this, this embodiment can obtain the vehicle status sent by the vehicle system by interacting with the vehicle system, and control the operating frequency of UWB nodes and data acquisition equipment in a hierarchical manner accordingly to solve the above problems.

[0088] Specifically, the child seat also communicates with the vehicle's infotainment system via a standardized interface and includes a backup battery. The child seat can obtain vehicle status information from the infotainment system. When the vehicle is in normal driving mode, it controls the main UWB node, all slave UWB nodes, and all data acquisition devices to operate at a first sampling frequency. When the vehicle is in standby mode with the engine off, it controls the main UWB node and target data acquisition devices to operate at a second sampling frequency. When the vehicle is in a locked sleep mode, it activates the backup battery to power the child seat and controls the main UWB node and target data acquisition devices to operate at a third sampling frequency.

[0089] The first sampling frequency is greater than the second sampling frequency, and the second sampling frequency is greater than the third sampling frequency.

[0090] In this embodiment, the child seat adopts a two-stage power supply architecture combining auxiliary power supply and a backup battery. The auxiliary power supply is achieved through a DC005 interface located on the side of the seat, supporting power from the original vehicle's cigarette lighter adapter or the original vehicle's USB Type-C interface. This interface is uniformly configured by the child seat itself and does not rely on modifications to the original vehicle's wiring or active cooperation from the vehicle's power supply system, thus it is compatible with more than 99% of existing vehicle models. When a collision causes a power outage, or when the main power is interrupted after the vehicle is turned off, it can automatically switch to the built-in backup battery. This backup battery is an automotive-grade wide-temperature lithium iron phosphate battery with a capacity of 2600mAh and an operating temperature range covering -40℃ to 85℃. It is encapsulated in a closed cavity inside the seat base, meeting both collision safety requirements and ensuring stable discharge performance in extreme environments.

[0091] In this embodiment, the vehicle system can send vehicle status information to the child seat in real time, and the child seat can dynamically adjust the working mode of each unit according to the obtained vehicle status information.

[0092] When the vehicle is in normal driving condition, the child seat can activate all sensing and data acquisition functions. That is, the main UWB node, each slave UWB node and all data acquisition devices operate at the first sampling frequency (e.g., 50Hz) to ensure high-time fusion of multi-node echo data and sensor data, and meet the needs of rapid identification of dangerous actions and abnormal sitting postures of children.

[0093] When the vehicle is turned off but not locked, the child seat enters a standby state. At this time, only the main UWB node at the headrest and the target data acquisition device operate at the second sampling frequency. In this embodiment, the target data acquisition device refers to the core data acquisition device, such as an installation information acquisition device, a seat cushion pressure acquisition device, a seatbelt status acquisition device, and a seat NTC temperature sensor. The second sampling frequency is lower than the first sampling frequency, correspondingly reducing overall power consumption, but still maintaining the basic ability to determine whether the child is still in the seat, preventing the child from accidentally slipping out or leaving the seat unnoticed during temporary vehicle parking.

[0094] Once the vehicle is locked, the child seat enters a locked sleep state. At this time, the backup battery is activated to power the entire system, and the sampling frequency of the main UWB node and the aforementioned target data acquisition device is reduced to the third sampling frequency. The child seat is in an ultra-low frequency sweep mode, which is only used to periodically detect the presence of respiratory micro-motion signals from living beings. All other UWB nodes and non-core sensors enter deep sleep. The overall sleep current can be controlled to 10μA or less, thereby maximizing the backup battery's runtime.

[0095] Furthermore, to facilitate timely monitoring and feedback of scenarios where children are left behind, an emergency wake-up mode needs to be set up. In this embodiment, the child seat can also determine whether the child occupant has not left the child seat based on the echo data of the main UWB node and the monitoring data of the target data acquisition device when the vehicle is in a standby state with the engine off or in a sleep state with the vehicle locked; if the child occupant has not left the child seat, the main UWB node, each slave UWB node, and each data acquisition device are controlled to operate at a first sampling frequency, and a child loss alarm is sent.

[0096] In this embodiment, when the vehicle enters a standby or locked-down state, the main UWB node inside the child seat maintains low-power operation, continuously collecting echo data. Simultaneously, the target data acquisition device also synchronously collects monitoring data at a lower frequency. This data is sent to the main control unit at the seat for joint analysis to determine whether the child occupant is still within the seat's seating area and has not been removed. Specifically, the echo signal acquired by the main UWB node is used to identify the presence of a living target and its spatial positional changes; installation data is used to determine if the child seat is still properly installed; pressure data is used to verify whether a pressure load within the child's weight range is continuously present; and the seatbelt status helps confirm whether the restraint system is still in a closed, locked state. If the combined information from these multiple sources indicates that the child occupant has not left the seat, the child seat is determined to have been abandoned.

[0097] In addition, data from temperature sensors can be used to collaboratively determine whether there is a risk of high-temperature heat radiation, in order to help assess the dangerous situation of a child in the event of being left behind.

[0098] In this embodiment, when it is determined that a child has been left behind, the child seat can switch from a low-power, low-frequency sampling operation mode to a full-power operation state, wake up all UWB nodes and all data acquisition devices, and restart high-frequency data acquisition and processing according to the first sampling frequency, so as to ensure that subsequent action recognition, vital sign assessment and early warning response have a real-time basis.

[0099] Meanwhile, the child seat can send child loss alarms, such as pushing emergency alarm information and vehicle location to the car owner's mobile app, phone, and SMS via the seat's built-in communication module, or activating a high-decibel sound and light alarm and simultaneously turning on the emergency ventilation function to reduce the seat surface temperature; if no effective response is received from the user within a set time, it can also automatically dial the preset emergency contact number.

[0100] In addition, child seats can transmit information such as whether a child has been left behind, whether vital signs are abnormal, and whether the seat belt is fastened to the vehicle's infotainment system in real time via a standardized communication interface. After receiving this data, the infotainment system can take further enhanced protective measures based on its own functions, such as flashing hazard lights, automatically sounding the horn, leaving windows slightly ajar, adjusting the air conditioning temperature, or displaying a pop-up notification on the central control screen.

[0101] In this way, after the vehicle is turned off or locked, the child seat can maintain basic monitoring capabilities with extremely low power consumption, and can quickly resume full-function operation and issue multi-channel alarms in a timely manner after confirming that the child seat has been left behind, thus balancing battery life reliability and emergency response effectiveness.

[0102] Optionally, the child seat can also be equipped with active safety features. Specifically, the child seat itself integrates an IMU (Inertial Measurement Unit) to collect real-time data on the child seat's three-axis acceleration and angular velocity during vehicle operation, in order to determine whether the vehicle is experiencing emergency braking.

[0103] When the IMU detects that the acceleration value exceeds a preset safety threshold, the child seat determines that an emergency braking has occurred and activates active restraint protection accordingly. At this time, the child seat can increase the sampling frequency of all multi-node UWB sensing arrays to a fourth sampling frequency, for example, to 100Hz, to achieve high-frequency dynamic tracking of the child's head and chest positions. Based on this, the main control unit can calculate the child's head and chest displacement in real time using the fused, centimeter-level three-dimensional coordinates. When the displacement at either position exceeds the preset safety threshold, the child seat immediately activates the pre-tensioning seatbelt actuator to automatically tighten, and simultaneously locks the headrest adjustment mechanism, thereby optimizing the child's body restraint posture during a collision and effectively reducing the risk of secondary injury due to restraint failure. The active restraint protection will then terminate when the vehicle restarts normally or after the vehicle has remained stationary for a preset period.

[0104] In addition, child seats can be ventilated according to actual conditions, including emergency ventilation protection and intelligent adaptation. Among them, emergency ventilation protection means that in the event of a child being left behind, if the temperature measured by the ambient temperature sensor in the vehicle exceeds 40 degrees Celsius, the child seat must automatically activate the ventilation module installed inside the seat base and backrest. This ventilation module delivers airflow to the area in contact with the child's torso through a built-in fan and air duct to accelerate the dissipation of heat from the seat surface and prevent the child from suffering heatstroke or skin burns due to prolonged exposure to a high-temperature, enclosed environment.

[0105] Intelligent adaptation at the seat end means that when the child seat determines that the child occupant is asleep based on echo data and three-dimensional coordinate information, it can control the seat's built-in ventilation module to adjust the airflow speed to a gentle level to maintain the child's comfort; when it detects that the occupant is crying continuously, the seat's built-in speaker will automatically play pre-stored soothing nursery rhymes.

[0106] The following section provides a possible approach to determining the occupant status of a child occupant based on target state analysis rules, echo data, and three-dimensional coordinate information.

[0107] In this embodiment, the target state analysis rules include target filtering parameters, target feature extraction algorithm, and target state recognition model. The child seat can filter the echo data based on the filtering parameters and extract the child state features from the echo data based on the feature extraction algorithm. The child state features and three-dimensional coordinate information are input into the state recognition model for processing to obtain the occupant state of the child. The occupant state includes the child's action type and physiological state.

[0108] Based on this, danger alarm information can be output according to the danger level and physiological state corresponding to the action type.

[0109] In this embodiment, the child seat can retrieve a set of rules corresponding to the current assembly state from a set of pre-stored state analysis rules. The target filtering parameters are used to suppress interference signals that are unrelated to the physiological characteristics or movement characteristics of the children in this group. For example, a combination of low-frequency gain enhancement and high-frequency noise suppression is designed for the weak breathing signals of newborns in group 0. The target feature extraction algorithm sets the feature extraction threshold based on the movement amplitude, breathing rhythm and body position change patterns unique to the children in this group, so as to stably capture the basic feature quantities reflecting their true state from the filtered echo data. The target state recognition model is a lightweight fusion model trained based on the group-specific action judgment rules. For example, it can be a neural network structure that works in conjunction with CNN (Convolutional Neural Network) and LSTM (Long Short-Term Memory). Its input end receives the child's state features and three-dimensional coordinate information obtained by the aforementioned steps, and its output end gives the joint judgment result of action type and physiological state.

[0110] In this embodiment, during the model training phase, technicians can manually label massive amounts of measured action data based on the prohibited action ranges and safety allowable boundaries corresponding to children in groups 0, 1, 2, and 3. For example, turning over at an angle exceeding 15 degrees is defined as an abnormal action in group 0, and standing and climbing actions are defined as high-risk actions in groups 2 or 3. This forms an age- and grouped action feature library, and based on this, four state recognition models are trained respectively.

[0111] During implementation, the child seat first identifies the installation form through a rotation positioning detection device, and then combines the weight data collected by the seat pressure sensor with the echo energy distribution ratio of each UWB node to complete the group determination; then, the target filtering parameters and target feature extraction algorithm of the corresponding group are automatically loaded to preprocess and extract features from the real-time echo data; finally, the standardized features and three-dimensional coordinate information are input into the target state recognition model dedicated to the group to obtain the occupant state output including the action type and physiological state.

[0112] In this embodiment, the child seat can determine whether a dangerous situation has occurred or predict whether a dangerous situation is about to occur based on the occupant's status, such as whether the child has made a dangerous move or whether there is an abnormal physiological state (such as rapid breathing), and issue corresponding danger alarm information according to the degree of danger of different dangerous situations.

[0113] In one example, if the occupant's condition is a precursor to some dangerous actions, such as incorrect sitting posture or slight displacement of the seat belt, a Level 1 warning can be issued. At this time, the child seat's own LED indicator light will emit a soft light signal, and the child seat's built-in soft buzzer will be activated at the same time to emit a low-decibel warning sound. This will remind the driver to pay attention to the current child's sitting status without interfering with the driver's normal driving, thereby avoiding potential risks in advance.

[0114] If the occupant's actions constitute a specific dangerous act or the child's vital signs are abnormal—for example, if the child unbuckles their seatbelt, stands, climbs, or leans out of the child seat while the vehicle is in motion—a Level 2 warning will be issued. In this case, the child seat's built-in red audible and visual warning device will be activated immediately, manifesting as a high-decibel buzzer and a bright LED flashing simultaneously. This clearly conveys an escalation of danger signal, prompting the driver to recognize the urgency of the situation and immediately choose a safe location to stop. Understandably, the warning intensity and response effect of a Level 2 warning should be higher than that of a Level 1 warning.

[0115] In addition, a three-level warning system can be set. For example, if a child is left behind, the temperature inside the vehicle reaches or exceeds a preset temperature (e.g., 40 degrees Celsius), or the displacement of the child in the child seat exceeds a preset safety threshold after a collision, a three-level warning can be issued. At this time, an emergency alarm message containing real-time vehicle location information should be pushed simultaneously to the owner's mobile APP, a preset phone number, and an SMS platform through the communication module integrated into the child seat. At the same time, the child seat itself will continuously activate a high-decibel audible and visual alarm and simultaneously activate the emergency ventilation function built into the child seat. If no manual confirmation response is received within the set time, the emergency contact number will be automatically dialed to ensure emergency protection can still be achieved even if the vehicle system is not connected or the vehicle system fails.

[0116] To perform the corresponding steps in the above embodiments and various possible methods, an implementation of a child occupant sensing device is given below. Optionally, the child occupant sensing device can adopt the above-described... Figure 1 The device structure of the child seat is shown. For further details, please refer to... Figure 4 , Figure 4 This is a functional block diagram of a child occupant sensing device provided in an embodiment of this application. It should be noted that the basic principle and technical effects of the child occupant sensing device provided in this embodiment are the same as those in the above embodiments. For the sake of brevity, any parts not mentioned in this embodiment can be referred to the corresponding content in the above embodiments. The child occupant sensing device includes: an acquisition module and a determination module.

[0117] This acquisition module is used to acquire echo data sent by the UWB radar device and monitoring data collected by the data acquisition device when the child seat is correctly installed and the child occupant is effectively seated.

[0118] Understandably, this acquisition module can also be used to perform the above step S20.

[0119] This determination module is used to determine the assembly status of the child seat based on echo data and monitoring data, and to generate three-dimensional coordinate information of the child occupant based on the echo data; to determine the target status analysis rules corresponding to the child occupant based on the assembly status; and to determine the occupant status of the child occupant based on the target status analysis rules, echo data, and three-dimensional coordinate information.

[0120] Understandably, this determining module can also be used to perform the above steps S21 to S23.

[0121] Optionally, the acquisition module is also used to acquire installation data, pressure data, seat belt data, and echo data from the installation information acquisition device, seat pressure acquisition device, and seat belt status acquisition device, respectively; if the installation data indicates that the child seat is installed in place, the pressure data is within the preset pressure range, the seat belt data indicates that the seat belt buckle is closed, and the echo data indicates that there is a living person on the child seat, then it is determined that the child seat has been effectively occupied by the child occupant.

[0122] Optionally, the determining module is also used to determine the installation configuration of the child seat based on the rotational positioning data sent by the rotational positioning detection device, and to determine the age group of the child occupant based on the echo data and pressure data; and to determine the assembly status of the child seat based on the installation configuration and age group.

[0123] Optionally, the determining module is also used to perform spatial alignment and boundary comparison filtering on the echo data of each UWB radar device according to the pre-built electronic fence to obtain target echo data; and to process the target echo data by weighted least squares method and extended Kalman filter to obtain the three-dimensional coordinate information of the child occupant.

[0124] Optionally, the child occupant sensing device also includes a control module.

[0125] This control module is used to acquire the vehicle status sent by the vehicle system; when the vehicle is in normal driving condition, it controls the main UWB node, each slave UWB node, and each data acquisition device to operate at a first sampling frequency; when the vehicle is in standby mode with the engine off, it controls the main UWB node and the target data acquisition device to operate at a second sampling frequency; when the vehicle is in lock-up and sleep mode, it activates the backup battery to power the child seat and controls the main UWB node and the target data acquisition device to operate at a third sampling frequency; wherein, the first sampling frequency is greater than the second sampling frequency, and the second sampling frequency is greater than the third sampling frequency.

[0126] Optionally, the control module is also used to determine whether a child occupant has not left the child seat based on the echo data of the main UWB node and the monitoring data of the target data acquisition device when the vehicle is in a standby state with the engine off or a sleep state with the vehicle locked; if the child occupant has not left the child seat, the module controls the main UWB node, each slave UWB node and each data acquisition device to operate at a first sampling frequency and sends a child abandonment alarm.

[0127] Optionally, the determining module is further configured to filter the echo data based on the target filtering parameters and extract the child's state features from the echo data based on the target feature extraction algorithm; input the child's state features and three-dimensional coordinate information into the target state recognition model for processing to obtain the occupant's occupant state; the occupant state includes the child occupant's action type and physiological state; and output danger alarm information according to the danger level corresponding to the action type and physiological state.

[0128] Optionally, the above modules can be stored in the form of software or firmware. Figure 1 The main control unit shown may be embedded in the operating system (OS) of the child seat, and can be controlled by... Figure 1 The main control unit executes the commands. Meanwhile, the data and program code required to execute the above modules can be stored in memory.

[0129] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a main control unit, implements the child occupant perception method provided in this application.

[0130] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0131] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0132] If a function is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0133] The above are merely preferred embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for sensing child occupants, characterized in that, The method is applied to child car seats, which are equipped with UWB radar and data acquisition devices. The child car seats store state analysis rules corresponding to each assembly state, and each assembly state represents the installation configuration of the child car seat and the age group of the child occupant. The method includes: When the child occupant is effectively seated in the child seat, the echo data sent by the UWB radar device and the monitoring data collected by the data acquisition device are acquired. The assembly status of the child seat is determined based on the echo data and the monitoring data, and the three-dimensional coordinate information of the child occupant is generated based on the echo data. Based on the assembly state, determine the target state analysis rules corresponding to the child occupant; Based on the target state analysis rules, the occupant state of the child occupant is determined according to the echo data and the three-dimensional coordinate information.

2. The method according to claim 1, characterized in that, The data acquisition equipment includes an installation information acquisition device, a seat cushion pressure acquisition device, and a seat belt status acquisition device; the method further includes: The installation data, pressure data, seat belt data, and echo data from the installation information acquisition device, the seat cushion pressure acquisition device, and the seat belt status acquisition device, respectively, are acquired. If the installation data indicates that the child seat is installed in place, the pressure data is within the preset pressure range, the seat belt data indicates that the seat belt buckle is closed, and the echo data indicates that there is a living person on the child seat, then it is determined that the child seat has been effectively occupied by the child occupant.

3. The method according to claim 2, characterized in that, The data acquisition device further includes a rotation positioning detection device, and the step of determining the assembly status of the child seat based on the echo data and the monitoring data includes: The installation configuration of the child seat is determined based on the rotation positioning data sent by the rotation positioning detection device, and the age group of the child occupant is determined based on the echo data and the pressure data. The assembly status of the child seat is determined based on the installation configuration and the age group.

4. The method according to claim 1, characterized in that, The UWB radar device includes multiple units, and the generation of the child occupant's three-dimensional coordinate information based on the echo data includes: Based on the pre-constructed electronic fence, the echo data of each of the UWB radar devices is spatially aligned and boundary-matched for filtering to obtain target echo data; The target echo data is processed using weighted least squares and extended Kalman filtering to obtain the three-dimensional coordinate information of the child occupant.

5. The method according to claim 4, characterized in that, The UWB radar device includes a master UWB node and multiple slave UWB nodes. The master UWB node is connected to each of the slave UWB nodes via a CAN bus. The master UWB node is located at the headrest of the child seat, and the slave UWB nodes are respectively located at the backrest, base, chest buckle, and buckle of the child seat.

6. The method according to claim 5, characterized in that, The child seat is also connected to the vehicle's infotainment system and is equipped with a backup battery. The method further includes: Obtain the vehicle status sent by the vehicle system; When the vehicle is in a normal driving state, the master UWB node, each slave UWB node and each data acquisition device are controlled to operate at a first sampling frequency. When the vehicle is in a standby state with the engine off, control the main UWB node and the target data acquisition device to operate at the second sampling frequency; When the vehicle is in a locked sleep state, the backup battery is activated to power the child seat, and the main UWB node and the target data acquisition device are controlled to operate at the third sampling frequency. Wherein, the first sampling frequency is greater than the second sampling frequency, and the second sampling frequency is greater than the third sampling frequency.

7. The method according to claim 6, characterized in that, The method further includes: When the vehicle is in a standby or locked-out state, the system determines whether the child occupant has not left the child seat based on the echo data of the main UWB node and the monitoring data of the target data acquisition device. If the child occupant does not leave the child seat, the master UWB node, each slave UWB node, and each data acquisition device are controlled to operate at a first sampling frequency and a child abandonment alarm is sent.

8. The method according to claim 1, characterized in that, The target state analysis rules include target filtering parameters, target feature extraction algorithms, and target state recognition models; the step of determining the occupant state of the child occupant based on the target state analysis rules, according to the echo data and the three-dimensional coordinate information, includes: The echo data is filtered based on the target filtering parameters, and the child's state features in the echo data are extracted based on the target feature extraction algorithm. The child's state features and the three-dimensional coordinate information are input into the target state recognition model for processing to obtain the occupant state of the child occupant; the occupant state includes the child occupant's action type and physiological state; The method further includes: Based on the danger level corresponding to the action type and the physiological state, a danger alarm message is output.

9. A child occupant sensing device, characterized in that, An application for child car seats, wherein the child car seat is equipped with a UWB radar device and a data acquisition device, and the child car seat stores state analysis rules corresponding to each assembly state, wherein the assembly state represents the installation form of the child car seat and the age group of the child occupant, the device comprising: The acquisition module is used to acquire echo data sent by the UWB radar device and monitoring data collected by the data acquisition device when the child seat is correctly installed and the child occupant is effectively seated. The determination module is used to determine the assembly status of the child seat based on the echo data and the monitoring data, and to generate the three-dimensional coordinate information of the child occupant based on the echo data; The determining module is further configured to determine the target state analysis rule corresponding to the child occupant based on the assembly state; The determining module is further configured to determine the occupant status of the child occupant based on the target status analysis rules, the echo data, and the three-dimensional coordinate information.

10. A child car seat, characterized in that, It includes a main control unit, which is used to execute a computer program to implement the method according to any one of claims 1-8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the main control unit, it implements the method described in any one of claims 1-8.

Images