A method and system for adjusting a car seat belt to fit a safety seat

Abstract

The application relates to a method and system for adjusting a car safety belt of an adaptive safety seat, and relates to safety belt adjustment, which comprises the following steps: determining individualized demand parameters based on a user identity; determining a safety belt restraint biomechanics model and a sitting posture comfort benchmark according to the individualized demand parameters; collecting real-time three-dimensional posture information and body pressure distribution information of the user; analyzing the real-time three-dimensional posture information and the body pressure distribution information according to the safety belt restraint biomechanics model and the sitting posture comfort benchmark to generate sitting posture adaptive optimization instructions; controlling an electric adjusting mechanism to execute the sitting posture adaptive optimization instructions and collecting a current state of the user; generating a safety belt wrapping path according to the current state of the user, the safety belt restraint biomechanics model and fixed geometric constraints; and driving an auxiliary operation device to complete automatic threading and buckling of the safety belt according to the safety belt wrapping path. The application has the effect of improving the convenience of user riding.

Description

Technical Field

[0001] This invention relates to the field of seat belt adjustment, and more particularly to a method and system for adjusting automotive seat belts to fit a child safety seat. Background Technology

[0002] Car seatbelt adjustment for child safety seats refers to a technological means that intelligently adjusts the seatbelt path, tightness, and anchor points for passengers with disabilities who have limited mobility, restricted limb movement, or special physical characteristics, so as to maximize their riding comfort, autonomy, and sense of dignity while ensuring collision safety.

[0003] Currently, commercially available car safety seats and their seat belt systems are mainly designed for occupants with standard sitting posture and independent movement capabilities, and their adjustment methods often require a certain degree of limb coordination and strength to operate.

[0004] For people with disabilities who have upper limb dysfunction, poor trunk stability, or who need to use wheelchairs for a long time, the existing seat belt system is difficult to wear and adjust, which reduces the convenience of riding and needs to be improved. Summary of the Invention

[0005] To improve the convenience of users, this invention provides a method and system for adjusting car seat belts to fit child safety seats.

[0006] In a first aspect, the present invention provides a method for adjusting a car seat belt to fit a child safety seat, employing the following technical solution:

[0007] A method for adjusting a car seatbelt to fit a child safety seat includes:

[0008] Collect user identification and real-time body posture data;

[0009] Determine personalized requirements parameters based on user identity identifiers;

[0010] Based on individualized requirements, determine the seat belt restraint biomechanical model and seating comfort benchmark;

[0011] Collect real-time 3D posture information and body pressure distribution information of users;

[0012] Based on the seat belt restraint biomechanical model and sitting comfort benchmark, real-time three-dimensional posture information and body pressure distribution information are analyzed to generate sitting posture adaptive optimization instructions.

[0013] The system controls the preset electric adjustment mechanism of the child safety seat to execute posture adaptive optimization commands and collects the user's current status.

[0014] The system generates a seatbelt wrapping path based on the user's current status, the seatbelt constraint biomechanical model, and preset fixed geometric constraints.

[0015] Based on the seat belt wrapping path, the preset auxiliary operating device is driven to complete the automatic threading and fastening of the seat belt.

[0016] By adopting the above technical solution, personalized parameters are first invoked through user identification. Combined with real-time body posture and pressure data, the seat posture and support are dynamically optimized based on a biomechanical model, prioritizing comfortable and stable seating while ensuring safety. Based on the optimized posture and fixed-point constraints, the system intelligently plans a three-dimensional path for the seat belt that conforms to the body's curves and avoids sensitive areas. It also controls auxiliary equipment to automatically put on and fasten the seat belt. This transforms the traditionally complex dressing operation that requires active user intervention into a proactive and continuous personalized service from the system. This significantly reduces the dressing difficulty for users with limited upper limb or trunk mobility, improving their autonomy and convenience while ensuring safety.

[0017] Optionally, methods for determining seatbelt restraint biomechanical models and seating comfort benchmarks may also be included.

[0018] Based on personalized demand parameters, determine the occupant's disability type identification and body contour data;

[0019] Based on occupant disability type identification, the priority of restraint targets and the physical risk area map are determined;

[0020] Generate personalized weighting coefficients based on the priority of constraint objectives;

[0021] Constructing a seatbelt constraint biomechanical model based on personalized weighting coefficients;

[0022] Combining body contour feature data with body risk area maps to generate mandatory avoidance areas and preferred constraint areas;

[0023] A seating comfort benchmark is generated based on the mandatory avoidance area, the preferred constraint area, and the preset general comfort pressure threshold.

[0024] Optionally, the following steps are also included after the seatbelt is fastened:

[0025] Collect users' body pressure distribution data and body micro-movement data;

[0026] Based on body pressure distribution data, calculate the rate of change of local pressure and the duration of pressure peak;

[0027] Extracting micro-motion spectral features based on body micro-motion data;

[0028] The risk level of local compression and the type of abnormal fretting are determined by combining the pressure change rate, peak duration, and fretting spectrum characteristics.

[0029] A comprehensive adjustment command is generated based on the local compression risk level and the type of abnormal micro-motion.

[0030] Based on the comprehensive adjustment command, the take-up unit is controlled to perform corresponding dynamic tension adjustment.

[0031] Optionally, tension adjustment methods may also be included:

[0032] Tension adjustment parameters and zone control parameters are determined based on comprehensive adjustment commands;

[0033] Based on the tension adjustment parameters, a sequence of commands for the take-up motor speed and direction is generated;

[0034] Based on the area control parameters, a preset voltage application command sequence is generated for each independent unit in the electrostatic adsorption array;

[0035] Based on the retractor motor speed and steering command sequence, the retractor is controlled to perform multi-stage actions including tension release, hold and recovery phases. At the same time, based on the voltage application command sequence, the electrostatic adsorption units of the corresponding sections of the seat belt liner are controlled to apply adsorption voltage of a specific intensity, and the body pressure distribution data and body micro-motion data are updated.

[0036] The immediate effects of synergistic regulation were assessed based on updated body pressure distribution data and body micro-movement data.

[0037] When the immediate effect of the coordinated adjustment meets the preset adjustment completion effect, an adjustment completion confirmation signal is generated.

[0038] Optional, also includes:

[0039] Collect real-time electrical characteristic data of the interface between each electrostatic adsorption unit and the user's clothing;

[0040] Determine the current equivalent adsorption strength and effective contact state based on real-time electrical property data;

[0041] The voltage amplitude compensation parameters are calculated based on the current equivalent adsorption intensity and effective contact state.

[0042] The final driving voltage of each unit is generated based on the voltage amplitude compensation parameters and the voltage application command sequence.

[0043] The voltage application operation of the corresponding electrostatic adsorption unit is controlled based on the final driving voltage.

[0044] Optional, also includes:

[0045] Collect dynamic load signals acting on the safety seat and seat belt system;

[0046] Based on dynamic load signals, analyze the time-frequency characteristics and spatial distribution characteristics of the load;

[0047] Anomaly patterns are generated based on the time-frequency and spatial distribution characteristics of the load.

[0048] If the abnormal event mode is a preset sudden abnormal event, the seat belt will be adjusted according to the preset sudden handling method.

[0049] Optionally, the emergency response method includes:

[0050] Based on abnormal event patterns, spatial location information and event intensity level information of events are obtained;

[0051] The target control area is determined based on the spatial location information of the event;

[0052] Target property alteration strategies are generated using event intensity level information;

[0053] The smart material unit is determined based on the target control region;

[0054] Current control parameters are generated by changing the strategy based on the target physical properties;

[0055] Current-controlled parameters are used to drive smart material units to change their rheological state, providing dynamic constraint support and energy dissipation.

[0056] Optional, also includes:

[0057] Acquire the adjusted dynamic load signal;

[0058] Determine the event decay state based on the adjusted dynamic load signal;

[0059] The preset safe recovery threshold is determined based on the event decay status.

[0060] When the safety recovery threshold is reached, the smart material unit is detrimentalized and returns to its basic flexible state.

[0061] Optional, also includes:

[0062] Collect users' physiological rhythm signals and posture maintenance duration data;

[0063] A fatigue state prediction model was established based on physiological rhythm signals and posture maintenance duration data.

[0064] Based on the fatigue state prediction model, it is determined whether the preset fatigue risk period is about to begin;

[0065] When the fatigue risk period is detected, the control retractor and electric adjustment mechanism perform preset small-amplitude cyclic movements to relieve the static load on the muscles.

[0066] Secondly, this application provides a car seat belt adjustment system adapted to a child safety seat, employing the following technical solution:

[0067] A car seatbelt adjustment system adapted to a child safety seat, comprising:

[0068] The data acquisition module is used to collect user identification, real-time body posture data, real-time 3D posture information, body pressure distribution information, and the user's current status.

[0069] A memory for storing a program that implements a method for adjusting a car seatbelt to fit a child safety seat;

[0070] The processor is used to load and execute programs stored in memory.

[0071] In summary, this application includes at least one of the following beneficial technical effects:

[0072] 1. First, personalized parameters are retrieved through user identification. Combined with real-time body posture and pressure data, the seat posture and support are dynamically optimized based on a biomechanical model, prioritizing comfortable and stable seating while ensuring safety. Based on the optimized posture and fixed-point constraints, the system intelligently plans a three-dimensional path for the seat belt that conforms to the body's curves and avoids sensitive areas. It also controls auxiliary equipment to automatically put on and fasten the seat belt. This transforms the traditionally complex dressing operation that requires active user intervention into a proactive and continuous personalized service from the system. This significantly reduces the dressing difficulty for users with limited upper limb or trunk mobility, improving their autonomy and convenience while ensuring safety.

[0073] 2. The overall tension is controlled by the retractor motor, combined with the electrostatic adsorption array to adjust the local fit, simultaneously optimizing the contact pressure distribution between the seat belt and the body during tension adjustment. Multi-stage retraction avoids the discomfort of single tightening, while zoned electrostatic adsorption dynamically adjusts the restraint force in key areas such as the shoulders and chest based on real-time body pressure data. Thus, without damaging the seat belt fabric, it achieves intelligent, comfortable, and personalized dynamic restraint, especially meeting the special needs of users with sensitive bodies, maximizing adaptability for daily riding while ensuring safety.

[0074] 3. By identifying the location and intensity of events, the system accurately locates the body areas requiring regulation and determines material property change strategies based on event intensity. By controlling the rheological state of intelligent material units, the system can switch a specific area from a comfort mode to a high-strength protection mode within milliseconds. This provides dynamic and targeted restraint enhancement and impact energy dissipation in sudden situations such as collisions and sharp turns, minimizing unnecessary full-body restraint while ensuring safety, achieving an intelligent balance between safety and comfort under extreme conditions. Attached Figure Description

[0075] Figure 1 This is a flowchart of a method for adjusting a car seatbelt to fit a child safety seat;

[0076] Figure 2 This is a flowchart of the tension adjustment method. Detailed Implementation

[0077] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0078] Reference Figure 1 This application discloses a method for adjusting a car seat belt to fit a child safety seat, including the following steps:

[0079] S10: Collect user identification and real-time body posture data.

[0080] User identification refers to a code used to uniquely identify and distinguish different occupants. In practice, this can be a unique code (such as a user ID) stored in a personal IC card or mobile device, or a biometric template obtained through an in-vehicle biometric module (such as a fingerprint reader or facial recognition camera). In this embodiment, a near-field communication card reader located next to the vehicle door or seat reads the code from the user's personal IC card as the user identification.

[0081] Real-time body posture data refers to a set of physical quantity data that reflects the current body contour, posture, and contact state with the seat, acquired in real time by sensors. Specifically, it includes body contour point cloud data obtained through 3D vision sensor scanning, and body pressure distribution maps collected by a flexible thin-film pressure sensor array integrated into the seat cushion and backrest. In this embodiment, a ToF depth camera mounted on the roof acquires the three-dimensional coordinates of key skeletal points of the torso and limbs, while pressure data is collected by a 1024-point pressure sensor array within the seat, together forming the real-time body posture data.

[0082] S11: Determine personalized requirements parameters based on user identity identifiers.

[0083] Personalized requirements parameters refer to a set of preset parameters directly related to a specific occupant's physical condition, health status, and riding preferences. This parameter set is obtained by querying a preset user profile database. The database stores for each registered user historical data including disability type (e.g., spinal cord injury level, degree of muscle atrophy), baseline body dimensions (shoulder width, sitting height, and thigh length in a seated position), medically contraindicated areas (e.g., surgical wounds, pressure ulcer-prone areas), and seatbelt pressure sensitivity. After reading the user's identity identifier, the system retrieves the corresponding parameter set from the database using this identifier as an index, which serves as the personalized requirements parameters for this adjustment. The user profile database is established and maintained in advance by those skilled in the art, and the data may originate from the user's initial system usage calibration entries, continuous riding learning records, or secure interconnection with an authorized medical database. The structured storage and indexing techniques used in the database are common knowledge in the field and will not be elaborated upon here.

[0084] S12: Determine the seat belt restraint biomechanical model and seating comfort benchmark based on individual needs parameters.

[0085] A seatbelt restraint biomechanical model is a computer mathematical model used to simulate the distribution of restraint forces exerted by the seatbelt system on specific occupant parts such as the torso and pelvis, as well as the occupant's body dynamic response under conditions such as vehicle acceleration, deceleration, turning, and collision. The core parameters of this model (such as the inertial parameters of each body segment, joint stiffness and damping, and the contact friction coefficient between the seatbelt and the body) are dynamically adjusted according to the body dimensions and disability type in the personalized requirements parameters.

[0086] Postural comfort benchmarks refer to a set of technical indicators and thresholds used to quantitatively assess and guide whether an occupant's posture is safe, stable, and comfortable. These include permissible pressure ranges for various body contact areas (such as the ischial tuberosities, back, and shoulders), ideal angle ranges for the spine and pelvis, and specific postural requirements tailored to different types of disabilities (such as compensatory support angles for individuals with scoliosis). Benchmark values ​​are derived from ergonomic standards and clinical research data on people with disabilities, and are fine-tuned based on individualized needs.

[0087] The specific methods for determining the seat belt restraint biomechanical model and the seating comfort benchmark will be explained in detail in subsequent sections S20 to S25, and will not be repeated here.

[0088] S13: Collect the user's real-time three-dimensional posture information and body pressure distribution information.

[0089] Real-time 3D posture information refers to the real-time data reconstructed from sensors, showing the 3D spatial position, orientation, and joint angles of the occupant's major body segments (such as head, torso, pelvis, and thighs) relative to the vehicle coordinate system. In this embodiment, it is obtained in real-time by fusing data from a ToF depth camera and seat posture sensors, and using a skeletal tracking algorithm.

[0090] Body pressure distribution information refers to the two-dimensional spatial distribution data of the pressure per unit area on the contact surface between the occupant's body and the seat. This is determined by reading the output of each sensing unit in the flexible pressure sensor array to generate a "heat map" reflecting the pressure magnitude and location.

[0091] S14: Based on the seat belt restraint biomechanical model and sitting comfort benchmark, analyze real-time three-dimensional posture information and body pressure distribution information to generate sitting posture adaptive optimization instructions.

[0092] The posture adaptive optimization command refers to a series of digital commands used to control the coordinated action of multiple electric adjustment mechanisms on the child safety seat to adjust the occupant's posture to a state that better conforms to biomechanical safety and comfort benchmarks.

[0093] By inputting real-time 3D posture information into a seatbelt-restrained biomechanical model, the theoretical force values ​​of key body parts under the current posture are calculated and compared with the safe force range preset in the model from publicly available biomechanical research data to quantify the posture safety deviation. Simultaneously, the body pressure distribution information is compared with the comfort pressure thresholds for each body zone preset in the sitting comfort benchmark, based on ergonomic experimental statistics, to quantify the pressure comfort deviation. Then, based on the constraint target priority defined in the personalized demand parameters, corresponding weights are assigned to the above deviations to construct an optimization function aimed at minimizing the weighted total deviation. Finally, using the physical adjustable range of each electric adjustment mechanism (such as lumbar support travel and leg support angle) as constraints, the optimization function is solved to calculate the specific adjustment amount required for each adjustment mechanism to make the overall occupant posture closest to the safety and comfort benchmark state. This set of adjustment amounts is then encoded into corresponding drive commands, thus generating a sitting posture adaptive optimization command.

[0094] S15: Controls the preset electric adjustment mechanism of the child safety seat to execute the posture adaptive optimization command and collects the user's current status.

[0095] Electric adjustment mechanisms are actuators integrated inside child seats that can change their mechanical position in response to electrical signals. Common types include servo motor-driven linear actuators (for adjusting the lumbar support and leg support extension length), geared motors (for adjusting the backrest angle), and memory foam airbags (for adjusting the side wing coverage).

[0096] The user's current state refers to the latest real-time three-dimensional posture and body pressure distribution information collected again by sensors after the electric adjustment mechanism has completed its operation and stabilized. This state is used to verify the adjustment effect and serves as input for subsequent seat belt path planning.

[0097] The electric adjustment mechanism of the child safety seat executes posture adaptive optimization commands and collects the user's current state for subsequent steps.

[0098] S16: Generate the seatbelt wrapping path based on the user's current state, the seatbelt constraint biomechanical model, and preset fixed geometric constraints.

[0099] Fixed geometric constraints refer to the immovable three-dimensional spatial coordinates of the fixed mounting points of the seat belt system on the vehicle body structure. Fixed geometric constraints are predetermined by those skilled in the art and will not be elaborated upon here.

[0100] The seatbelt wrap path refers to the three-dimensional spatial curve that the seatbelt webbing must follow from the retractor, around or through the occupant's body, and finally to the buckle.

[0101] First, a 3D surface model of the occupant's torso and limbs is reconstructed based on the user's current 3D posture information. Second, constraint requirements for the current user are obtained from the seatbelt constraint biomechanical model. These requirements include at least the body parts the seatbelt should fit against (e.g., shoulders, hips) and areas of risk that must be avoided (e.g., wounds, medical equipment). Then, the 3D surface model, constraint requirements, and fixed geometric constraints are used as input conditions. A pre-defined path planning algorithm (e.g., the Rapid Expanding Random Tree (RRT) algorithm) iteratively solves the problem while satisfying the following multiple constraints: the path must pass through the specified fitting areas, completely avoid all risk areas, and the path's start and end points must coincide with the fixed geometric coordinates of the seatbelt retractor exit and buckle position, respectively. Finally, the smooth space curve output by the algorithm, satisfying all constraints, is the generated seatbelt wrapping path. In cases where multiple feasible paths exist, the system selects the optimal path based on the principle of the shortest total path length.

[0102] S17: Based on the seat belt wrapping path, drive the preset auxiliary operating device to complete the automatic threading and fastening of the seat belt.

[0103] The auxiliary operating device refers to a robotic arm installed inside the vehicle cabin, possessing multiple degrees of freedom of movement, and equipped with a specially designed gripper at its end for grasping the seatbelt latch. The system converts the seatbelt wrapping path into a sequence of motion trajectories for the robotic arm's end effector. The robotic arm is controlled to move near the retractor to grasp the latch, then moves precisely along the planned path, guiding the seatbelt webbing around the occupant's body, and finally aligning and inserting the latch into the buckle located on the other side of the seat. A built-in force sensor confirms a "click" locking sound or the achievement of a preset insertion force, thus completing the automatic threading and fastening process.

[0104] It also includes methods for determining seatbelt restraint biomechanical models and seating comfort benchmarks:

[0105] S20: Determine the occupant disability type identifier and body contour feature data based on personalized demand parameters.

[0106] The occupant disability type identifier is a code that uniquely identifies the category and level of a specific physical impairment of an occupant. This identifier is extracted directly from the personalized needs parameters.

[0107] Body profile data refers to the set of geometric parameters that describe the static dimensions of an occupant's body. This data is also extracted from personalized requirements parameters.

[0108] The personalized requirements parameters include occupant disability type identification and body contour data.

[0109] S21: Determine the priority of constraint targets and the physical risk area map based on the occupant disability type identification.

[0110] Constraint target priority refers to the ranking of the importance of the various safety protection tasks that the seat belt system needs to accomplish.

[0111] The priority of constraint targets is obtained by querying a pre-defined type strategy library. This strategy library was jointly built by rehabilitation medicine experts and automotive safety engineers, and a corresponding priority sequence is predefined for each type of disability. For example, for users with high-level paraplegia, the knowledge base defines "pelvic stability" as the highest priority and "chest decompression" as the second highest priority.

[0112] A body risk zone map refers to information on body surface areas that require special attention or avoidance from the seat belt system, marked in the form of graphical coordinate data.

[0113] This atlas was also obtained by querying the aforementioned strategy database. For specific disability types, the atlas will mark in detail the common sites of pressure ulcers (such as the ischial tuberosity and sacrum and coccyx), areas of sensory loss, surgical scar locations, or the surface projection range of implanted medical devices (such as spinal fixation devices).

[0114] S22: Generate personalized weight coefficients based on the priority of the constraint objectives.

[0115] Personalized weighting coefficients refer to a set of values ​​used in subsequent mathematical models to quantify the relative importance of different constraint objectives.

[0116] A linear normalization algorithm is used to assign a weight value of 1.0 to the highest-ranking target in the constraint target priority sequence, and assign values ​​between 0 and 1 to the remaining targets according to their relative relationship with the highest priority.

[0117] S23: Constructing a seatbelt constraint biomechanical model based on personalized weighting coefficients.

[0118] Based on a general human multibody dynamics model and a seatbelt force element model, personalized weight coefficients generated by S22 are embedded into the optimization objective function of this framework. Specifically, within the cost function used by the model to predict body displacement during a collision, the penalty terms for indicators such as pelvic displacement and thoracic vertebral compression are set as corresponding personalized weight coefficients. In this way, the general model is "personalized," resulting in a seatbelt-restrained biomechanical model.

[0119] S24: Combine body contour feature data with body risk area map to generate mandatory avoidance area and preferred constraint area.

[0120] The mandatory avoidance zone refers to the spatial area on the surface of the occupant's personalized 3D body model (generated based on body contour data) that must be completely avoided when planning the seatbelt path. This zone is generated by scaling and mapping all high-risk areas (such as pressure sores and scars) marked in the body risk area atlas according to the body contour data, precisely covering the corresponding anatomical locations on the user's 3D model.

[0121] The preferred constraint zone refers to the ideal spatial area on the surface of the occupant's personalized 3D body model where the seatbelt webbing should be tightly fitted and exert restraint force when planning the seatbelt path. This zone is determined based on biomechanical best practices and is located in areas with prominent bones, effective force transmission, and thin soft tissue (such as above the anterior superior iliac spine in the pelvis, or the sternum and clavicle region of the thorax). Its specific boundaries are calculated and calibrated on the user's 3D model based on body contour data.

[0122] S25: Generate a sitting comfort benchmark based on the forced avoidance area, the preferred constraint area, and the preset general comfort pressure threshold.

[0123] The universal comfort pressure threshold refers to a reference value for the range of pressure that various body parts can tolerate over a long period of time, applicable to healthy adults. The universal comfort pressure threshold is predetermined by those skilled in the art and will not be elaborated upon here.

[0124] The seat comfort benchmark is a "digital map" that defines the upper and lower limits of allowable pressure and is associated with each surface unit of the user's three-dimensional body model.

[0125] The rules for generating the sitting comfort benchmark are as follows: For all surface units within the mandatory avoidance area, their upper limit of comfort pressure is forcibly set to 0 kPa; for all surface units within the preferred constraint area, the aforementioned general comfort pressure threshold is directly used as their upper and lower limits; for other ordinary areas in the model that are neither mandatory avoidance nor preferred constraint, 80% of the general comfort pressure threshold is used as their upper limit of comfort pressure. The resulting sitting comfort benchmark serves as the direct quantitative basis for subsequent real-time evaluation and adjustment of sitting comfort.

[0126] It also includes the steps after the seatbelt is fastened:

[0127] S30: Collects user's body pressure distribution data and body micro-movement data.

[0128] Body pressure distribution data refers to digital information that is collected in real time by a flexible thin-film pressure sensor array integrated into the seat cushion and backrest of a child safety seat, characterizing the magnitude and two-dimensional position distribution of pressure on the contact surface between the occupant's body and the seat.

[0129] Micro-motion data refers to the small-amplitude (typically less than 10 mm) vibrations and displacements of the occupant's torso and pelvis under restraint, acquired in real time by a high-sensitivity MEMS accelerometer installed in the seatbelt webbing. In this embodiment, triaxial accelerometers are built into the shoulder straps and lap belt of the seatbelt, respectively, to collect acceleration signals at a sampling frequency of 100 Hz, reflecting minute tremors or involuntary swaying of the body.

[0130] S31: Calculate the rate of change of local pressure and the duration of pressure peak based on body pressure distribution data.

[0131] The rate of change of local pressure refers to the rate of average pressure change in a predefined region of interest (such as the left ischial tuberosity region) within body pressure distribution data, calculated over a unit of time (e.g., 1 second). The rate of change of local pressure is obtained by acquiring the average pressure value of that region at the current moment and the previous sampling moment, calculating the difference, and dividing by the sampling time interval. The specific calculation method is common knowledge in the field and will not be elaborated here.

[0132] The peak pressure duration refers to the cumulative time during which the real-time average pressure value of a specific area of ​​interest continuously exceeds a preset warning pressure threshold in body pressure distribution data. The warning pressure threshold is obtained by multiplying the upper limit of the comfortable pressure for the corresponding area in the sitting comfort benchmark by a safety factor (e.g., 1.2). The system records the duration of pressure exceeding the threshold using a timer. The specific implementation of the timer is common knowledge in the field and will not be elaborated here.

[0133] S32: Extracting micro-motion spectral features based on body micro-motion data.

[0134] Micro-motion spectral characteristics refer to the core parameters extracted from the frequency domain analysis of micro-motion data signals of the body within a time window (e.g., 5 seconds), which characterize the energy distribution of the signal's frequency components. By performing a Fast Fourier Transform (FFT) on the acceleration signal, the dominant frequency of the signal's energy concentration and the energy proportion of the signal within a preset typical frequency band (e.g., the 1-5Hz tremor band, the 5-15Hz muscle fiber tremor band) are extracted as micro-motion spectral characteristics. The Fast Fourier Transform is common knowledge in this field and will not be elaborated upon here.

[0135] S33: Combine pressure change rate, peak duration, and fretting spectrum characteristics to determine the level of local compression risk and the type of abnormal fretting.

[0136] Local pressure risk level refers to a quantitative level (e.g., "low", "medium", or "high") representing the ischemic risk to tissues in a given area, obtained by referring to a pre-defined pressure risk level reference table based on the calculated local pressure change rate and peak pressure duration. The pressure risk level reference table records the levels corresponding to different ranges of pressure change rate and peak pressure duration. This reference table is pre-set by those skilled in the art based on clinical research data and will not be elaborated upon here.

[0137] This control table was pre-set by those skilled in the art based on clinical research data, and will not be elaborated upon here.

[0138] The abnormal micro-motion type refers to matching the extracted micro-motion spectral features to one of several preset typical abnormal micro-motion categories using a pattern recognition algorithm. In this embodiment, the preset categories include "low-frequency spasm type," "high-frequency tremor type," and "irregular restlessness type." The abnormal micro-motion type is determined by inputting the real-time extracted spectral features into a pre-trained classification model (such as a support vector machine (SVM) model) for matching. The training method of the classification model is common knowledge in the art and will not be described in detail here.

[0139] S34: Generate comprehensive adjustment instructions based on the level of local compression risk and the type of abnormal micro-motion.

[0140] Integrated control commands refer to a structured data package of control commands that encodes tension adjustment requirements for pressure risks and coordinated control requirements for abnormal micro-movements.

[0141] The comprehensive adjustment command is generated by querying the preset risk adjustment comparison table and the micro-motion coordination comparison table, and combining them based on the query results.

[0142] The risk adjustment table records the target tension adjustment mode (such as "periodic slow release" and "single pulse release") and the target tension change range (such as "-10%" and "-20%) corresponding to different local compression risk levels.

[0143] The micro-motion synergy comparison table records the target adsorption area (such as "shoulder area" and "pelvic area") and target adsorption mode (such as "high frequency reverse damping" and "continuous reinforcement of adhesion") corresponding to different abnormal micro-motion types (such as "high frequency tremor type" and "low frequency spasm type").

[0144] The specific content of the aforementioned comparison tables was set by those skilled in the art based on extensive experimental data and simulation analysis, with the goal of achieving an optimal balance between safety and comfort, and will not be elaborated upon here. Based on the local pressure risk level and abnormal micro-motion type determined in S33, the system queries the two comparison tables respectively to obtain the corresponding target tension adjustment mode and amplitude, as well as the target adsorption area and mode. These two sets of information are then encapsulated in the same data package, generating a comprehensive adjustment command. The "target tension adjustment mode and amplitude" in this command is the direct source of the "tension adjustment parameters" in subsequent S40; the "target adsorption area and mode" in this command is the direct source of the "area control parameters" in subsequent S40.

[0145] S35: Controls the take-up retractor to perform corresponding dynamic tension adjustment based on the comprehensive adjustment command.

[0146] Dynamic tension adjustment refers to the take-up reel driving its built-in servo motor to perform corresponding actions based on the target tension adjustment parameters in the comprehensive adjustment command.

[0147] For example, if the instruction is "periodic slow release mode, amplitude -10%", the retractor will control the motor to perform a cycle of first slowly reversing to release a certain length of seat belt (reducing the tension by about 10%), holding it for a short time, and then slowly retracting it to the original position.

[0148] Reference Figure 2 It also includes tension adjustment methods:

[0149] S40: Determine tension adjustment parameters and zone control parameters based on comprehensive adjustment commands.

[0150] Tension adjustment parameters refer to the set of specific parameters extracted from the comprehensive adjustment command and used to directly control the operation of the take-up motor.

[0151] The tension adjustment parameters corresponding to the target tension adjustment mode and amplitude can be found by using the preset tension parameter comparison table. The table records different tension adjustment parameters corresponding to different target tension adjustment modes and amplitudes. The comparison content in the tension parameter comparison table is formed by those skilled in the art by recording the different tension adjustment parameters corresponding to different target tension adjustment modes and amplitudes in sequence, which will not be elaborated here.

[0152] Regional control parameters refer to the set of instruction parameters extracted from the comprehensive control instructions, used to control the coordinated operation of specific unit groups in the electrostatic adsorption array.

[0153] The control parameters corresponding to the target adsorption region and mode can be found by using the preset control parameter comparison table. The table records the different regional control parameters corresponding to different target adsorption regions and modes. The comparison content in the control parameter comparison table is formed by those skilled in the art by recording the different regional control parameters corresponding to different target adsorption regions and modes in sequence, which will not be elaborated here.

[0154] S41: Generate a sequence of commands for the speed and direction of the retractor motor based on the tension adjustment parameters.

[0155] The speed of the retractor motor refers to the rotational speed that the retractor motor needs to reach in order to achieve the tension change requirements specified in the tension adjustment parameters.

[0156] The winding motor speed corresponding to the tension adjustment parameter can be found by using the preset tension speed comparison table. The table records the different winding motor speeds corresponding to different tension adjustment parameters. The comparison content in the tension speed comparison table is formed by the experiment and recording of different winding motor speeds corresponding to different tension adjustment parameters by those skilled in the art, and will not be elaborated here.

[0157] A steering command sequence refers to a series of commands arranged in chronological order that control the rotation direction of the retractor motor (forward tensioning or reverse release).

[0158] The steering command sequence corresponding to the tension adjustment parameter can be found by using the preset tension steering reference table. The table records different steering command sequences corresponding to different tension adjustment parameters. The reference content in the tension steering reference table is formed by those skilled in the art by recording the different steering command sequences corresponding to different tension adjustment parameters in sequence, which will not be elaborated here.

[0159] S42: Generate a preset voltage application command sequence for each independent unit in the electrostatic adsorption array based on the area control parameters.

[0160] An electrostatic adsorption array refers to a mesh-like functional fabric sewn or laminated into the lining of seatbelt webbing, consisting of multiple independently addressable and actuated microelectrode pairs. The structure of the electrostatic adsorption array is designed by those skilled in the art according to functional requirements and will not be described in detail here.

[0161] The voltage application command sequence refers to the voltage waveform control commands generated in chronological order for each electrostatic adsorption unit specified in the area control parameters.

[0162] The voltage application command sequence corresponding to the regional control parameter can be found by using the preset regional voltage lookup table. The table records different voltage application command sequences corresponding to different regional control parameters. The reference content in the regional voltage lookup table is formed by those skilled in the art by recording the different voltage application command sequences corresponding to different regional control parameters in sequence, which will not be elaborated here.

[0163] S43: Based on the retractor motor speed and steering command sequence, control the retractor to perform multi-stage actions including tension release, hold and recovery phases. At the same time, based on the voltage application command sequence, control the electrostatic adsorption unit of the corresponding zone of the seat belt liner to apply an adsorption voltage of a specific intensity, and update the body pressure distribution data and body micro-motion data.

[0164] According to the steering command sequence, the tension release phase control retractor motor rotates in the reverse direction at a specified speed to release the seat belt webbing and reduce the tension; the holding phase control motor stops to maintain the current tension; and the recovery phase control motor rotates in the forward direction at a specified speed to retract the webbing and restore the tension to the target value.

[0165] Simultaneously, based on the timing and voltage parameters specified in the voltage application command sequence, the electrostatic adsorption unit electrodes of the target partition are controlled by the high-voltage drive circuit to generate a corresponding electrostatic field, thereby applying an adsorption force of a specific intensity.

[0166] Furthermore, after completing the aforementioned coordinated control actions, the system again collects the latest body pressure distribution data and body micro-movement data through the sensors in S30 to reflect the real-time state after adjustment.

[0167] S44: Evaluate the immediate effects of synergistic regulation based on updated body pressure distribution data and body micro-movement data.

[0168] The immediate effect of synergistic regulation refers to the quantitative indicator of the degree of improvement of this regulation by comparing the body pressure distribution data and body micro-movement data before (S30) and after (S43 after update). The specific comparative calculation method is common knowledge in this field and will not be elaborated here.

[0169] S45: When the immediate effect of the coordinated adjustment meets the preset adjustment completion effect, an adjustment completion confirmation signal is generated.

[0170] The adjustment completion effect refers to the threshold condition used to determine that the coordinated adjustment has achieved the expected goal. The adjustment completion effect is preset by those skilled in the art and will not be elaborated here.

[0171] When the immediate effect of the coordinated adjustment meets the adjustment completion requirement, it indicates that the adjustment is complete, and an adjustment completion confirmation signal can be generated; otherwise, the adjustment steps are repeated until the adjustment is complete.

[0172] Also includes:

[0173] S50: Collects real-time electrical characteristic data of the interface between each electrostatic adsorption unit and the user's clothing.

[0174] Real-time electrical characteristic data refers to the electrical parameters obtained by measuring each electrostatic adsorption unit through a measurement circuit. Specifically, this includes the equivalent capacitance and equivalent resistance values ​​measured when a low-voltage test signal is applied. These parameters are read directly through the measurement circuit, and the specific design of the measurement circuit is common knowledge in this field and will not be elaborated here.

[0175] S51: Determine the current equivalent adsorption strength and effective contact state based on real-time electrical property data.

[0176] The current equivalent adsorption strength refers to the actual adsorption force level calculated using a preset conversion formula based on the equivalent capacitance value in real-time electrical characteristic data. This conversion formula maps the capacitance value to an estimated adsorption force value. The specific parameters of the formula are determined by those skilled in the art through experimental calibration and will not be elaborated here.

[0177] Effective contact status refers to whether the unit and the clothing have formed effective contact, as determined by whether the equivalent resistance value in the real-time electrical characteristic data is lower than a preset good contact threshold. This threshold is set by those skilled in the art based on system characteristics and clothing material, and will not be elaborated upon here.

[0178] S52: Calculate the voltage amplitude compensation parameters based on the current equivalent adsorption intensity and effective contact state.

[0179] The voltage amplitude compensation parameter refers to the compensation coefficient that needs to be adjusted to compensate for adsorption force deviation or poor contact. This parameter is calculated through a compensation function, the input of which is the ratio of the current equivalent adsorption intensity to the target adsorption intensity, and a coefficient of the effective contact state (e.g., 1 for good contact, and a penalty coefficient greater than 1 for poor contact). The specific form of the compensation function is determined by those skilled in the art and will not be elaborated here.

[0180] S53: Generates the final drive voltage for each unit based on the voltage amplitude compensation parameters and the voltage application command sequence.

[0181] The final driving voltage refers to the voltage value actually applied to the electrodes of the electrostatic adsorption unit after compensation and adjustment. The amplitude of the final driving voltage is obtained by multiplying the base voltage amplitude specified in the voltage application command sequence by the corresponding voltage amplitude compensation parameter. Its timing is consistent with the base command.

[0182] S54: Control the corresponding electrostatic adsorption unit to perform voltage application operation based on the final driving voltage.

[0183] Based on the obtained final driving voltage, the corresponding electrostatic adsorption unit is controlled to perform voltage application operation, thereby generating a precise and controllable electrostatic adsorption force in a specific area of ​​the seat belt liner. This works in conjunction with the dynamic tension adjustment of the retractor to jointly achieve the adjustment goal of relieving local pressure and suppressing abnormal micro-movements.

[0184] Also includes:

[0185] S60: Collects dynamic load signals acting on the safety seat and seat belt system.

[0186] Dynamic load signal refers to a force signal that changes rapidly over time, collected by force sensors integrated into the seat belt anchor point and seat frame.

[0187] S61: Analyze the time-frequency characteristics and spatial distribution characteristics of load based on dynamic load signals.

[0188] Load time-frequency characteristics refer to the time-frequency distribution matrix obtained by performing wavelet transform on the dynamic load signal, which is used to characterize the energy distribution of the signal at different times and frequencies. Wavelet transform is common knowledge in this field and will not be elaborated here.

[0189] Spatial distribution characteristics refer to the main position and direction of action of a dynamic load in space, calculated using the triangulation principle based on the amplitude differences and time differences of arrival of signals from multiple sensors. The triangulation principle is common knowledge in this field and will not be elaborated upon here.

[0190] S62: Generate abnormal event patterns based on the time-frequency characteristics and spatial distribution characteristics of the load.

[0191] An exception event pattern is a pattern identifier that categorizes and describes current dynamic events.

[0192] By inputting the analyzed load time-frequency feature matrix and spatial coordinates into a pre-trained event classification neural network model, the model outputs the corresponding abnormal event pattern category. This neural network model employs an architecture combining convolutional neural networks and long short-term memory networks. Its input layer receives the time-frequency feature matrix as image features and spatial coordinates as auxiliary vectors. Deep features of the time-frequency image are extracted through multi-layer convolution and downsampling, fused with the spatial coordinate features, and then input into a fully connected layer for classification. Finally, it outputs a classification label corresponding to the current event, such as "muscle spasm," "external impact," or "severe cough." The structure and training method of the neural network model are defined by those skilled in the art and will not be elaborated here.

[0193] S63: If the abnormal event mode is a preset sudden abnormal event, the seat belt will be adjusted using the preset sudden handling method.

[0194] Sudden abnormal events refer to event types that are marked as requiring urgent handling in the output categories of the event classification model. The specific types are defined by those skilled in the art based on safety requirements, and will not be elaborated here.

[0195] Emergency response procedures refer to pre-programmed control procedures designed to handle various sudden and abnormal events. Specific emergency response procedures will be detailed in sections S70 to S75 and will not be elaborated upon here.

[0196] If the abnormal event mode is a sudden abnormal event, it means that the event requires emergency handling, and therefore the seat belt should be adjusted using the emergency handling method.

[0197] Emergency response methods include:

[0198] S70: Obtain event spatial location information and event intensity level information based on abnormal event patterns.

[0199] Event spatial location information refers to the coordinates of the main action location of dynamic loads directly extracted from the spatial distribution characteristics obtained from S61 analysis.

[0200] Event intensity level information refers to the quantized value obtained by calculating the peak amplitude of the dynamic load signal and mapping it to a preset intensity level (such as level 1 to 5). The mapping rules are set by those skilled in the art and will not be elaborated here.

[0201] S71: Determine the target control area based on the spatial location information of the event.

[0202] The target control area refers to the specific area on the interface between the safety seat and the occupant that requires activation of the intelligent material unit for control. By comparing the spatial coordinates of the event with the preset coordinate range of the seat surface area (set in advance by those skilled in the art, and not elaborated here), the smallest area unit containing that coordinate point is determined as the target control area.

[0203] S72: Generate target property change strategies based on event intensity level information.

[0204] The target property alteration strategy refers to the specific instructions set to change the physical state of a smart material unit in response to events of specific intensity.

[0205] The target property alteration strategy is generated by querying a pre-defined graded property lookup table. This lookup table records different target physical state parameters corresponding to different event intensity levels, including target stiffness values, target viscosity values, and target response times required for state transitions. This lookup table was established by those skilled in the art based on experimental data of the rheological properties of smart materials under different driving conditions, combined with safety and comfort requirements. The specific experimental data and lookup table content are set by those skilled in the art and will not be elaborated here.

[0206] S73: Determine the smart material unit based on the target control region.

[0207] A smart material unit refers to a functional unit integrated into the contact interface of a child safety seat that can change its physical properties in response to external signals. This unit encapsulates controllable smart materials, such as electrorheological fluids, whose rheological properties (such as viscosity and shear modulus) can change rapidly and reversibly when an external electric field is applied.

[0208] By querying a pre-defined coordinate unit database, the physical numbers of all smart material units located within the target control area can be retrieved. This database, based on a gridded coordinate system of the seat surface, records a list of smart material unit numbers corresponding to each grid unit. This database is pre-constructed by those skilled in the art based on the specific design of the seat and the unit integration location, and will not be elaborated upon here. The system queries and outputs the corresponding list of smart material unit numbers from this database based on the target control area coordinates determined in step S71.

[0209] S74: Generate current control parameters by changing the strategy based on the target physical properties.

[0210] Current control parameters refer to the current control command parameters that need to be applied to drive the smart material unit to achieve the target change in physical properties.

[0211] The current control parameters are obtained by querying a pre-defined physical property-current mapping model. This mapping model is based on the field strength-rheological property calibration curves of the smart material and is used to convert the target stiffness, target viscosity, and response time parameters specified in the target property modification strategy into the current amplitude, waveform, and duration required to drive the electromagnetic coil inside the smart material unit. The specific form of the calibration curve is determined by those skilled in the art based on the physical properties of the selected smart material and will not be elaborated upon here.

[0212] S75: Based on current control parameters, smart material units are driven to change their rheological state to provide dynamic constraint support and energy dissipation.

[0213] Based on the generated current control parameters, the system applies a specified current to the electromagnetic coil inside the target smart material unit via a controllable current source. The magnetic field generated in the coil causes the magnetorheological fluid inside the unit to transform from a low-viscosity Newtonian fluid to a high-viscosity, high-shear-modulus Bingham-like fluid within milliseconds, thereby achieving a rapid and controllable change in its rheological state. This state transition allows the smart material unit to locally form a momentary "solid" or "high-damping" support point in the target control area. When abnormal dynamic loads (such as muscle spasm impacts) are transmitted to this area, the locally hardened smart material unit can directly absorb and dissipate the impact energy through solid contact, and limit the abnormal displacement of body tissue with additional local constraint forces, thereby providing dynamic constraint support and energy dissipation, effectively suppressing the risk of injury or discomfort to occupants caused by sudden abnormal movements.

[0214] Also includes:

[0215] S80: Acquires the adjusted dynamic load signal.

[0216] The adjusted dynamic load signal refers to the dynamic load signal collected again from the force sensor after the execution of step S75.

[0217] S81: Determine the event decay state based on the adjusted dynamic load signal.

[0218] Event decay status refers to determining whether the original abnormal event has decayed and the degree of decay by calculating the ratio of the peak energy of the adjusted dynamic load signal to the peak energy of the original signal collected in step S60.

[0219] S82: Determine whether the preset safe recovery threshold has been reached based on the event decay status.

[0220] The safe recovery threshold is a threshold value used to determine whether an event has decayed to a safe level. The safe recovery threshold is preset by those skilled in the art and will not be elaborated upon here.

[0221] When the energy ratio calculated by S81 is lower than the safe recovery threshold, it is determined that the safe recovery condition has been met.

[0222] S83: When the safety recovery threshold is reached, the control smart material unit is detrimentalized and restored to the basic flexible state.

[0223] When the safety recovery threshold is reached, it indicates that the safety recovery conditions have been met. The intelligent material unit can be detrimentalized and restored to its basic flexible state.

[0224] Also includes:

[0225] S90: Collects user's physiological rhythm signals and posture maintenance duration data.

[0226] Physiological rhythm signals refer to signals that reflect the periodic changes in a user's vital signs, collected by a non-contact millimeter-wave bioradar sensor. Specifically, these include respiratory rate signals calculated by detecting micro-movements in the chest and abdomen, and heart rate variability signals extracted by analyzing the micro-Doppler effect on the skin surface.

[0227] Posture maintenance duration data refers to the duration during which a user remains relatively still in their current sitting posture. By analyzing the movement trajectory of the pressure center point in the body pressure distribution data (S30), when the movement amplitude of this trajectory is consistently less than a preset threshold (e.g., 2 cm) within a preset duration (e.g., 5 seconds), it is determined to be posture maintenance, and the duration of this state is accumulated by a timer inside the system. The specific determination threshold is set by those skilled in the art and will not be elaborated here.

[0228] S91: Establish a fatigue state prediction model based on physiological rhythm signals and posture maintenance duration data.

[0229] A fatigue prediction model is a mathematical model that can predict the probability of user fatigue based on real-time physiological signals and posture maintenance duration. This model is trained by performing regression analysis on historical data (including physiological signals, posture duration, and corresponding user subjective fatigue scores).

[0230] S92: Based on the fatigue state prediction model, determine whether the preset fatigue risk period is about to begin.

[0231] The fatigue risk period refers to the time interval when the real-time fatigue probability value output by the fatigue state prediction model continuously exceeds the preset fatigue probability threshold (such as 0.7).

[0232] A fatigue state prediction model is used to determine whether a user is about to enter a preset fatigue risk period. Specifically, real-time physiological rhythm signals (such as respiratory rate and heart rate variability) and posture maintenance duration data collected by the S90 are input into the established fatigue state prediction model. The model outputs a real-time fatigue probability value ranging from 0 to 1. The system continuously monitors this probability value. If it is higher than a preset fatigue probability threshold for a preset number of consecutive sampling periods (e.g., 3 consecutive periods, 30 seconds each), the user is determined to be about to enter a fatigue risk period. The fatigue probability threshold and the number of consecutive judgment periods are set by those skilled in the art based on model performance and safety redundancy requirements, and will not be elaborated here.

[0233] S93: When it is determined that the fatigue risk period has been entered, the control retractor and electric adjustment mechanism perform preset small-amplitude cyclic movements to relieve the static load on the muscles.

[0234] Small-amplitude cyclical movements refer to periodic micro-motion modes used to alleviate fatigue. For example, controlling the retractor to fluctuate tension by ±1cm at a frequency of 0.1Hz, and coordinating with the lumbar support to perform slow lifting and lowering with an amplitude of 3mm. Specific motion parameters are set by those skilled in the art and will not be elaborated here.

[0235] When it is determined that the fatigue risk period has begun, the retractor and electric adjustment mechanism should be controlled to perform small-amplitude cyclic movements to relieve the static load on the muscles.

[0236] Based on the same inventive concept, embodiments of the present invention provide a car seat belt adjustment system adapted to a child safety seat, comprising:

[0237] The data acquisition module is used to collect user identification, real-time body posture data, real-time three-dimensional posture information, body pressure distribution information, user current status, body pressure distribution data, body micro-motion data, real-time electrical characteristic data, dynamic load signal, adjusted dynamic load signal, physiological rhythm signal, and posture maintenance duration data.

[0238] A memory for storing a program that implements a method for adjusting a car seatbelt to fit a child safety seat;

[0239] The processor is used to load and execute programs stored in memory.

[0240] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0241] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for adjusting a car seat belt to fit a child safety seat, characterized in that, include: Collect user identification and real-time body posture data; Determine personalized requirements parameters based on user identity identifiers; Based on individualized requirements, determine the seat belt restraint biomechanical model and seating comfort benchmark; Collect real-time 3D posture information and body pressure distribution information of users; Based on the seat belt restraint biomechanical model and sitting comfort benchmark, real-time three-dimensional posture information and body pressure distribution information are analyzed to generate sitting posture adaptive optimization instructions. The system controls the preset electric adjustment mechanism of the child safety seat to execute posture adaptive optimization commands and collects the user's current status. The system generates a seatbelt wrapping path based on the user's current status, the seatbelt constraint biomechanical model, and preset fixed geometric constraints. Based on the seat belt wrapping path, a preset auxiliary operating device is driven to complete the automatic threading and fastening of the seat belt; It also includes methods for determining seatbelt restraint biomechanical models and seating comfort benchmarks: Based on personalized demand parameters, determine the occupant's disability type identification and body contour data; Based on occupant disability type identification, the priority of restraint targets and the physical risk area map are determined; Generate personalized weighting coefficients based on the priority of constraint objectives; Constructing a seatbelt constraint biomechanical model based on personalized weighting coefficients; Combining body contour feature data with body risk area maps to generate mandatory avoidance areas and preferred constraint areas; A seating comfort benchmark is generated based on the mandatory avoidance area, the preferred constraint area, and the preset general comfort pressure threshold.

2. The method for adjusting a car seat belt to fit a child safety seat according to claim 1, characterized in that, It also includes the steps after the seatbelt is fastened: Collect users' body pressure distribution data and body micro-movement data; Based on body pressure distribution data, calculate the rate of change of local pressure and the duration of pressure peak; Extracting micro-motion spectral features based on body micro-motion data; The risk level of local compression and the type of abnormal fretting are determined by combining the pressure change rate, peak duration, and fretting spectrum characteristics. A comprehensive adjustment command is generated based on the local compression risk level and the type of abnormal micro-motion. Based on the comprehensive adjustment command, the take-up unit is controlled to perform corresponding dynamic tension adjustment.

3. The method for adjusting a car seat belt to fit a child safety seat according to claim 2, characterized in that, It also includes tension adjustment methods: Tension adjustment parameters and zone control parameters are determined based on comprehensive adjustment commands; Based on the tension adjustment parameters, a sequence of commands for the take-up motor speed and direction is generated; Based on the area control parameters, a preset voltage application command sequence is generated for each independent unit in the electrostatic adsorption array; The retractor is controlled to perform multi-stage actions including tension release, hold and recovery phases based on the retractor motor speed and steering command sequence. At the same time, the electrostatic adsorption unit of the corresponding section of the seat belt liner is controlled to apply adsorption voltage based on the voltage application command sequence, and the body pressure distribution data and body micro-motion data are updated. The immediate effects of synergistic regulation were assessed based on updated body pressure distribution data and body micro-movement data. When the immediate effect of the coordinated adjustment meets the preset adjustment completion effect, an adjustment completion confirmation signal is generated.

4. The method for adjusting a car seat belt to fit a child safety seat according to claim 3, characterized in that, Also includes: Collect real-time electrical characteristic data of the interface between each electrostatic adsorption unit and the user's clothing; Determine the current equivalent adsorption strength and effective contact state based on real-time electrical property data; The voltage amplitude compensation parameters are calculated based on the current equivalent adsorption intensity and effective contact state. The final driving voltage of each unit is generated based on the voltage amplitude compensation parameters and the voltage application command sequence. The voltage application operation of the corresponding electrostatic adsorption unit is controlled based on the final driving voltage.

5. The method for adjusting a car seat belt to fit a child safety seat according to claim 1, characterized in that, Also includes: Collect dynamic load signals acting on the safety seat and seat belt system; Based on dynamic load signals, analyze the time-frequency characteristics and spatial distribution characteristics of the load; Anomaly patterns are generated based on the time-frequency and spatial distribution characteristics of the load. If the abnormal event mode is a preset sudden abnormal event, the seat belt will be adjusted according to the preset sudden handling method.

6. The method for adjusting a car seat belt to fit a child safety seat according to claim 5, characterized in that, The emergency response methods include: Based on abnormal event patterns, spatial location information and event intensity level information of events are obtained; The target control area is determined based on the spatial location information of the event; Target property alteration strategies are generated using event intensity level information; The smart material unit is determined based on the target control region; Current control parameters are generated by changing the strategy based on the target physical properties; Current-controlled parameters are used to drive smart material units to change their rheological state, providing dynamic constraint support and energy dissipation.

7. A method for adjusting a car seat belt to fit a child safety seat according to claim 6, characterized in that, Also includes: Acquire the adjusted dynamic load signal; Determine the event decay state based on the adjusted dynamic load signal; The preset safe recovery threshold is determined based on the event decay status. When the safety recovery threshold is reached, the smart material unit is detrimentalized and returns to its basic flexible state.

8. A method for adjusting a car seat belt to fit a child safety seat according to claim 1, characterized in that, Also includes: Collect users' physiological rhythm signals and posture maintenance duration data; A fatigue state prediction model was established based on physiological rhythm signals and posture maintenance duration data. Based on the fatigue state prediction model, it is determined whether the preset fatigue risk period is about to begin; When the fatigue risk period is detected, the control retractor and electric adjustment mechanism perform preset small-amplitude cyclic movements to relieve the static load on the muscles.

9. A car seatbelt adjustment system adapted to a child safety seat, characterized in that, include: The data acquisition module is used to collect user identification, real-time body posture data, real-time 3D posture information, body pressure distribution information, and the user's current status. A memory for storing a program that implements a method for adjusting a car seat belt to fit a safety seat as described in any one of claims 1 to 8; The processor is used to load and execute programs stored in memory.

Images