Anti-motion sickness control method and system based on multi-modal perception and vehicle

Abstract

This invention discloses a motion sickness control method, system, and vehicle based on multimodal perception, belonging to the field of vehicle technology. The control method includes the following steps: collecting vehicle body state data and multimodal state data of each person in the vehicle; fusing the vehicle body state data and the multimodal state data of each person using a fusion algorithm to obtain fused data for each person; outputting the motion sickness risk level of each person using a deep learning model based on the fused data of each person; calculating the priority level of each person based on their motion sickness risk level and individual characteristic information; and adjusting vehicle parameters based on the motion sickness risk level of the person with the highest priority level. By collecting vehicle body state data and multimodal state data of each person in the vehicle, the vehicle's driving status and the real-time physiological state of the driver and passengers can be obtained, and vehicle parameters can be adjusted accordingly to prevent motion sickness, facilitating precise and comfortable effective prevention and control, and improving the driving and riding experience.

Description

Technical Field

[0001] This invention relates to the field of vehicle technology, and in particular to a motion sickness control method, system, and vehicle based on multimodal perception. Background Technology

[0002] Motion sickness (commonly known as "car sickness") is a key bottleneck in improving the driving and riding experience. With the development of vehicle technology, anti-motion sickness technology is being increasingly widely applied to vehicles. Current anti-motion sickness technology only monitors vehicle posture parameters and analyzes the risk of motion sickness among drivers and passengers based on the vehicle's driving status to determine whether intervention is necessary. However, such anti-motion sickness technology ignores individual differences, real-time physiological states, and behavioral characteristics of drivers and passengers. This results in a lack of targeted prevention and control measures, limited effectiveness, and difficulty in achieving precise, comfortable, and effective prevention and control, which is detrimental to improving the driving and riding experience. Summary of the Invention

[0003] The present invention aims to solve the technical problems existing in the above-mentioned related technologies, and proposes a motion sickness control method, system and vehicle based on multimodal perception, which can accurately and comfortably prevent and control motion sickness.

[0004] According to a first aspect of the present invention, a motion sickness prevention control method based on multimodal perception includes the following steps: Collect vehicle body status data and multimodal status data of each person in the vehicle. Then, use a fusion algorithm to fuse the vehicle body status data and the multimodal status data of each person to obtain fused data of each person. Based on the fused data from each individual, a deep learning model is used to output the motion sickness risk level for each individual. Priority levels for each individual are calculated based on their motion sickness risk level and individual characteristics. Based on the motion sickness risk level of the person with the highest priority, adjust vehicle parameters to prevent motion sickness.

[0005] The motion sickness control method based on multimodal perception according to the first aspect of the present invention has at least the following technical effects: by collecting vehicle body state data and multimodal state data of each person in the vehicle, the vehicle's driving state and the real-time physiological state of the driver and passengers can be obtained, and vehicle parameters can be adjusted accordingly to prevent motion sickness, facilitating precise and comfortable effective prevention and control, and improving the driving experience. Furthermore, for multi-occupant scenarios, the present invention adds a priority ranking algorithm, prioritizing suspension adjustment based on the motion sickness risk level of the highest priority passenger, which better meets actual needs.

[0006] According to some embodiments of the present invention, the step of collecting multimodal state data of each person on the vehicle includes: The data collected included skin conductivity, heart rate variability, respiratory rate, pupil diameter change rate, head shaking frequency and amplitude, and sitting pressure distribution of each person on the vehicle.

[0007] According to some embodiments of the present invention, the step of calculating the priority level of each person based on their motion sickness risk level and individual characteristic information includes: The age and motion sickness sensitivity of each person are obtained from the human-computer interaction module as individual characteristic information.

[0008] According to some embodiments of the present invention, the control method further includes the following steps: Monitor the motion sickness risk level of each person. If the motion sickness risk level of any person is greater than or equal to a preset level threshold and the duration is greater than a preset duration threshold, adjust the suspension parameters and cabin parameters to prevent motion sickness.

[0009] According to some embodiments of the present invention, the step of adjusting vehicle parameters includes: Adjust the suspension damping and / or stiffness parameters according to the preset risk level-suspension parameter mapping relationship.

[0010] According to some embodiments of the present invention, the control method further includes the following steps: The comfort feedback data is obtained from the human-computer interaction module, and the fusion algorithm and / or the risk level-suspension parameter mapping relationship are optimized based on the comfort feedback data.

[0011] According to a second aspect of the present invention, the motion sickness prevention control system includes a multimodal perception module, a data fusion processing module, a motion sickness risk assessment module, a suspension control module, an occupant interaction module, and a communication module. The modules work together to implement the above-mentioned motion sickness prevention control method based on multimodal perception.

[0012] According to the second aspect of the present invention, the motion sickness control system has at least the following technical effects: by applying the above-mentioned motion sickness control method based on multimodal perception, the vehicle driving status and the real-time physiological status of the driver and passengers are obtained, and the vehicle parameters are adjusted accordingly to prevent motion sickness, which facilitates precise, comfortable and effective prevention and control, and helps to improve the driving experience.

[0013] According to some embodiments of the present invention, the multimodal sensing module includes flexible bioelectrodes, cockpit millimeter-wave radar, infrared camera, seat pressure sensor array, and vehicle body status sensor.

[0014] According to some embodiments of the present invention, the communication module adopts a hybrid communication architecture of Ethernet and CAN bus.

[0015] According to a third aspect of the present invention, a vehicle includes the aforementioned motion sickness control system.

[0016] According to the third aspect of the present invention, the vehicle has at least the following technical effects: by setting the above-mentioned anti-motion sickness control system, the vehicle's driving status and the real-time physiological status of the driver and passengers can be obtained, and the vehicle parameters can be adjusted accordingly to prevent motion sickness, which facilitates precise, comfortable and effective prevention and control, and helps to improve the driving experience.

[0017] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0018] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart of a motion sickness control method based on multimodal perception according to an embodiment of the present invention. Detailed Implementation

[0019] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0020] In the description of this invention, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, "several" means one or more, "multiple" means two or more, "greater than," "less than," "exceeding," etc., are understood to exclude the stated number, while "above," "below," "within," etc., are understood to include the stated number. If "first" or "second" is used, it is only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0021] In the description of this invention, unless otherwise explicitly defined, terms such as "set up," "install," and "connect" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this invention in conjunction with the specific content of the technical solution.

[0022] The following is for reference. Figure 1 This invention describes a motion sickness control method, system, and vehicle based on multimodal perception, according to embodiments of the present invention.

[0023] The motion sickness prevention control method based on multimodal perception according to a first aspect of the present invention includes the following steps: Step S100: Collect vehicle body status data and multimodal status data of each person in the vehicle. Then, perform fusion processing on the vehicle body status data and the multimodal status data of each person using a fusion algorithm to obtain fused data of each person.

[0024] Step S200: Based on the fused data of each person, output the motion sickness risk level of each person through a deep learning model; Step S300: Calculate the priority level of each person based on their motion sickness risk level and individual characteristics. Step S400: Adjust vehicle parameters according to the motion sickness risk level of the person with the highest priority to prevent motion sickness.

[0025] By collecting vehicle status data and multimodal status data of passengers, the vehicle's driving status and the real-time physiological status of the driver and passengers can be obtained. Based on this, vehicle parameters can be adjusted to prevent motion sickness, which facilitates precise and comfortable prevention and control, and improves the driving experience. Moreover, multimodal status data reduces the interference of single sensors, improves the accuracy of motion sickness risk identification, and reduces the error of risk level classification.

[0026] In step S200, it is understood that existing technologies lack a dedicated motion sickness risk quantification model, only providing feedback adjustments after sudden changes in vehicle posture. This fails to identify early signs of motion sickness (such as increased breathing rate and frequent head movements), easily missing the optimal control window and exacerbating discomfort. Therefore, this invention establishes a motion sickness risk level model based on the real-time state of the occupants, dynamically matching suspension control parameters and updating the risk level every 5 minutes based on the latest perceived data, ensuring real-time assessment and achieving adaptive control that is "personalized and condition-dependent." By identifying early signs of motion sickness and dynamically adjusting vehicle parameters, the vehicle's pitch and roll angles are controlled within ±1.5°, reducing the 2-8Hz low-frequency vibration transmission rate and improving the subjective comfort of the occupants.

[0027] The motion sickness risk level model was established as follows: a deep learning-based "occupant status-motion sickness risk" mapping model was constructed. The model input was a fused 12-dimensional feature vector, and the output was a motion sickness risk level of 0-5 (Level 0: no risk; Level 1-2: low risk; Level 3-4: medium risk; Level 5: high risk). The model was trained using a CNN-LSTM hybrid network. The training dataset contained multimodal data from 1000 occupants of different ages (3-60 years old) and their corresponding subjective motion sickness ratings. The model training accuracy was ≥95%, and the generalization error was ≤5%. Based on the occupant's subjective comfort rating (1-10 points), a mapping relationship between risk level and comfort was established: Level 0 (9-10 points), Level 1 (8-9 points), Level 2 (7-8 points), Level 3 (5-7 points), Level 4 (3-5 points), and Level 5 (1-3 points).

[0028] In some embodiments of the present invention, the step of collecting multimodal state data of each person on the vehicle in step S100 above includes: Step S110 involves collecting data on skin conductivity, heart rate variability, respiratory rate, pupil diameter change rate, head shaking frequency and amplitude, and seated pressure distribution for each passenger on the vehicle. It is understood that real-time physiological states (fatigue and anxiety lower the motion sickness threshold) and behavioral characteristics affect the risk of motion sickness. Existing technologies lack precise capture of passenger states, resulting in a lack of targeted control strategies. This embodiment constructs a multimodal passenger state perception system using multimodal state data, integrating physiological biological signals, visual features, and posture data to achieve precise quantitative assessment of motion sickness risk.

[0029] In step S100, the process of fusing the vehicle body state data and the multimodal state data of each person using a fusion algorithm employs a three-level processing mechanism of "noise reduction-synchronization-feature fusion". Noise reduction processing: Wavelet transform algorithm is used to remove power frequency interference for physiological signals, median filtering algorithm is used to remove light interference for visual signals, and Kalman filtering algorithm is used to smooth noise for vehicle body status data.

[0030] Time synchronization: Based on the vehicle clock signal, the data collected by different sensors are synchronized to the millisecond level to avoid evaluation errors caused by data timing deviations.

[0031] Feature fusion: A multimodal fusion algorithm with attention mechanism is adopted to perform weighted fusion of physiological signal features (mean skin conductivity, standard deviation of heart rate variability, respiratory rate), visual features (pupil diameter change rate, head sway amplitude), posture features (pressure distribution entropy value) and vehicle body state features (acceleration change rate). The weights are dynamically allocated based on the correlation coefficient between each feature and motion sickness risk (e.g., physiological signal weight 0.4, visual feature weight 0.3, posture feature weight 0.1, vehicle body state feature weight 0.2).

[0032] In some embodiments of the present invention, step S100 further includes: Step S120: Simultaneously monitor the operating status of each sensing device. If a sensing device malfunction is detected, switch to redundant sensing. Real-time monitoring of device status and automatic switching to redundancy schemes in case of malfunction ensures system operational stability.

[0033] In some embodiments of the present invention, step S300 specifically includes: Step S310: Obtain the age and motion sickness sensitivity of each person from the human-computer interaction module as individual characteristic information.

[0034] In steps S300 and S310, it is understood that even if the real-time states and multimodal state data of all occupants are consistent, the severity of motion sickness may vary due to differences in age and motion sickness sensitivity. Therefore, this invention adds a priority ranking algorithm for multi-occupant scenarios, prioritizing suspension adjustment based on the motion sickness risk level of the highest priority occupant, which better meets practical needs. Individual characteristic information such as age and motion sickness sensitivity of each person is actively input by the driver and passengers through the human-vehicle interaction module. Based on the occupant's age (age coefficient set to 1.5 for the elderly and children, and 1.0 for young adults), motion sickness sensitivity (sensitivity coefficients of 0.1-1.0 for levels 1-10), and real-time risk level, the priority level of each occupant is calculated (priority level = age coefficient × sensitivity coefficient × motion sickness risk level), and the priority ranking result is output. It supports adaptive adjustment for occupants of different ages and with different motion sickness sensitivities, enhancing personalized adaptability. Adaptability scenarios cover urban commuting, long-distance driving, family travel, etc., expanding the user adaptability to the entire population. In addition, a weighted average algorithm can be used to integrate the parameter requirements of other occupants to balance the comfort of all occupants.

[0035] In some embodiments of the present invention, the control method further includes the following steps: Step S500: Monitor the motion sickness risk level of each person. If the motion sickness risk level of any person is greater than or equal to a preset level threshold and the duration is greater than a preset duration threshold, adjust the suspension parameters and cabin parameters to prevent motion sickness. The preset level threshold can be set to level 5, and the preset duration threshold can be set to 3 seconds. This application sets up targeted handling for possible abnormal emergency situations: If a passenger's motion sickness risk level is detected to be greater than or equal to level 5 (high risk) for 3 seconds, an emergency linkage mechanism is triggered: the suspension control module switches to "emergency comfort mode", adjusting the suspension damping to the upper limit of the corresponding level and the stiffness to the lower limit; the cabin system is linked to start emergency measures, the seat activates a gentle lumbar massage (frequency 50 times / minute), the car audio plays soothing music (frequency 80-120Hz), and the central control screen pushes motion sickness prevention tips (such as "It is recommended to maintain a sitting posture and look ahead"); a visual + voice prompt is sent to the driver through the instrument panel ("The risk of motion sickness for rear passengers is too high, it is recommended to slow down or stop and rest nearby").

[0036] In some embodiments of the present invention, the step of adjusting vehicle parameters in step S400 above includes: Step S410: Adjust the suspension damping and / or stiffness parameters according to the preset risk level-suspension parameter mapping relationship. An adaptive control strategy based on the "risk level-parameter mapping" is employed, outputting suspension damping / stiffness adjustment commands based on the model predictive control (MPC) algorithm. The controlled object is either an active air suspension or a CDC continuously damped suspension. The baseline mapping representation of risk level-suspension parameters is established as follows:

[0037] Step S420: Dynamically adjust the suspension damping and / or stiffness parameters based on vehicle body status data (e.g., increase the corresponding suspension damping by 10% when the pitch angle exceeds 1°). This ensures a high degree of adaptation between the suspension control parameters and the occupant's condition, effectively reducing the incidence of motion sickness.

[0038] In some embodiments of the present invention, the control method further includes the following steps: Step S600: Obtain comfort feedback data from the human-computer interaction module, and optimize the fusion algorithm and / or the risk level-suspension parameter mapping relationship based on the comfort feedback data. The fusion algorithm is optimized by adjusting feature weights; for example, if a child reports "discomfort relief," the system adjusts the weight of the child's corresponding visual feature from 0.3 to 0.4. Manual calibration of sensitivity and feedback comfort is supported, forming a closed-loop optimization mechanism of "perception-evaluation-control-feedback," solving the problem of the lack of an interactive closed loop in existing technologies.

[0039] The motion sickness prevention control system of the second aspect of the present invention includes a multimodal perception module, a data fusion processing module, a motion sickness risk assessment module, a suspension control module, an occupant interaction module, and a communication module. These modules work together to implement the aforementioned multimodal perception-based motion sickness prevention control method. By applying the aforementioned multimodal perception-based motion sickness prevention control method, the vehicle's driving status and the real-time physiological status of the driver and passengers are obtained, and vehicle parameters are adjusted accordingly to prevent motion sickness. This facilitates precise, comfortable, and effective prevention, improving the driving and riding experience. It can be deeply integrated with in-vehicle intelligent cockpits and autonomous driving systems, providing technical support for subsequent personalized intelligent travel solutions. It can be applied to electric vehicles equipped with active / semi-active suspensions, especially suitable for family travel (including the elderly and children), long-distance driving, and urban commuting in congested areas where motion sickness is prone to occur. It can be directly integrated into existing intelligent cockpit and drive-by-wire chassis systems without significant modifications to the vehicle body structure, exhibiting good compatibility and broad industrial application prospects. Among them: The multimodal perception module is used to collect physiological and bio-signals, visual features, posture data, and vehicle status data of one or more occupants. It includes a physiological signal acquisition unit, a visual feature acquisition unit, a posture acquisition unit, a vehicle status acquisition unit, and a multi-occupant identification unit. The physiological signal acquisition unit uses flexible bioelectrodes integrated into the seat back to collect occupant skin conductivity (reflecting anxiety / tension levels) and heart rate variability (reflecting physiological comfort), with a sampling frequency of 100Hz; it also uses cabin millimeter-wave radar to monitor occupant breathing rate non-contactly, avoiding discomfort from physical contact. The visual feature acquisition unit uses a cabin infrared camera (supporting low-light environment recognition) to collect occupant pupil diameter, eye closure frequency, and head movement frequency and amplitude, with a sampling frequency of 30fps. The posture acquisition unit uses a seat pressure sensor array to collect occupant sitting posture data (such as lumbar support pressure and hip pressure distribution) to determine if the occupant is in a posture prone to motion sickness (such as leaning forward or tilting). Vehicle status acquisition unit: A vehicle status sensor integrating a 3-axis accelerometer, gyroscope, and suspension travel sensor to collect longitudinal / lateral / vertical acceleration, pitch angle, roll angle, and suspension travel data, providing a basis for control reference; Multi-occupant identification unit: Through the position correlation of cockpit millimeter-wave radar, infrared camera, and seat pressure sensor array, it distinguishes occupants in different seats (driver, front passenger, and rear seats), realizing independent acquisition and identification of multi-occupant status.

[0040] The data fusion processing module is used to reduce noise, synchronize and fuse multi-source sensing data, receive fault diagnosis signals and switch redundancy schemes.

[0041] The motion sickness risk assessment module is used to train a deep learning model based on fused data and output the motion sickness risk level and priority for one or more passengers.

[0042] The suspension control module is used to dynamically adjust the suspension damping / stiffness parameters according to the risk level and priority, and to trigger emergency measures in case of abnormalities.

[0043] The occupant interaction module supports sensitivity calibration, comfort feedback, and emergency information display. It includes a touch interaction unit, a voice interaction unit, and a status display unit, supporting individual characteristic calibration and comfort feedback. Individual characteristic calibration: Occupants can manually input their motion sickness sensitivity (level 1-10) and age via voice or the central control screen. Comfort feedback: Occupants can provide feedback on their current comfort level ("comfortable," "moderate," "uncomfortable") via touch or voice. The system adjusts the feature weights of the multimodal fusion algorithm and the suspension parameter mapping table based on the feedback, with a self-learning cycle of 10 trips. Status display unit: The central control screen displays the current motion sickness risk level, suspension control status, and a summary of perception data in real time, enhancing the user's right to know.

[0044] The communication module enables low-latency data transmission between modules. It employs a hybrid communication architecture combining Ethernet and CAN bus, achieving data transmission latency ≤20ms and a 50% improvement in control response speed compared to existing technologies. Stable performance is maintained even under extreme conditions (such as strong light and bumps). Ethernet enables high-speed transmission of multimodal status data with a transmission rate ≥1Gbps; the CAN bus transmits suspension control commands with a communication latency ≤10ms, ensuring stable coordination among modules. The hardware fault self-diagnosis module monitors the working status of each sensing device (bioelectrode, millimeter-wave radar, infrared camera, etc.) in real time via the CAN bus (signal transmission interruption or data anomaly rate ≥15% is considered a fault). When a device fault is detected, a redundancy scheme is automatically triggered: if the bioelectrode is faulty, an alternative assessment model is constructed using visual features (pupil diameter change rate, head sway amplitude) and posture features (pressure distribution entropy value) to indirectly output the risk level; if the millimeter-wave radar is faulty, the frequency of chest rise and fall of the occupant is identified through infrared camera images, replacing respiratory rate monitoring. Fault information is fed back to the central control screen in real time, displaying a message: "A sensing device is faulty; redundancy scheme activated; further maintenance recommended." The design incorporates a multimodal data fusion algorithm and a low-latency communication interface to improve the reliability of sensing data and the response speed of control commands, thereby addressing the shortcomings of existing technologies in multi-device collaboration.

[0045] In other embodiments of the present invention, the control system may further include a hardware fault self-diagnosis module: monitoring the operating status of sensing devices and each module, and automatically switching redundant sensing schemes after identifying faults.

[0046] In other embodiments of the invention, the cockpit millimeter-wave radar can be replaced by a high-resolution camera (monitoring respiratory rate by recognizing occupant chest movements in images); the flexible bioelectrodes can be replaced by wearable smart devices (such as smartwatches) that transmit physiological signals via Bluetooth; the infrared camera can be replaced by a TOF camera to improve attitude recognition accuracy. The deep learning model can be replaced by a support vector machine (SVM) or random forest algorithm, suitable for vehicle controllers with lower computing power; the data fusion algorithm can be replaced by Bayesian estimation to reduce algorithm complexity. Suspension damping adjustment can be replaced by suspension stiffness adjustment (such as using magnetorheological springs); single-variable control can be replaced by multi-variable collaborative control (simultaneously adjusting suspension damping, stiffness, and seat posture).

[0047] The above alternative solutions can all achieve the core objective of this invention (adaptive suspension adjustment for motion sickness prevention based on occupant status), but they may differ in terms of perception accuracy, response speed, and adaptability, and all fall within the protection scope of this invention.

[0048] The following describes the motion sickness control method, system, and vehicle based on multimodal perception according to a specific embodiment of the present invention.

[0049] The aforementioned control system is implemented using a pure electric mid-size sedan as the application platform. The specific configuration and workflow are as follows: Hardware configuration: Multimodal perception module: flexible bioelectrodes (integrated into the backrests of the driver and passenger seats), 60GHz cockpit millimeter-wave radar (installed in the center of the roof interior panel), infrared camera (installed in the steering column), seat pressure sensor array (driver and passenger seats), 3-axis accelerometer (at the center of gravity of the vehicle body), gyroscope (in the longitudinal beams of the vehicle body), and suspension travel sensor (4-wheel suspension).

[0050] Control modules: data fusion processor, motion sickness risk assessment chip (integrated CNN-LSTM algorithm), suspension controller (supports air suspension damping / stiffness adjustment).

[0051] Interactive modules: central control touch screen, voice assistant (supports voice commands for "motion sickness mode on / off" and "comfort feedback").

[0052] Communication module: vehicle-mounted Ethernet switch, CAN FD bus controller.

[0053] Workflow: Startup phase: After the vehicle is powered on, the multimodal perception module and hardware fault self-diagnosis module perform self-checks to confirm that all devices are normal; the occupants input their own information through the central control screen (driver 30 years old, sensitivity level 5; front passenger 60 years old, sensitivity level 8; rear seat 6-year-old child, sensitivity level 9), and the system initializes.

[0054] Perception and Fusion Phase: During vehicle operation, the multi-occupant identification unit distinguishes occupants in each seat. Bioelectrodes collect data on the skin conductivity of the driver, front passenger, and child, with heart rate variability of 20ms, 35ms, and 40ms, respectively. Millimeter-wave radar monitors the front passenger's respiratory rate at 25 breaths / minute and the child's at 28 breaths / minute. Infrared cameras collect data on the child's pupil diameter change rate at 25% and head movement frequency at 5 times / minute, and the front passenger's pupil change rate at 18%. Pressure sensors collect data on the child's posture deviation (uneven pressure on the lower back). Vehicle status sensors collect data on longitudinal acceleration at 0.2g and pitch angle at 1.2°. The data fusion processing module reduces noise and synchronizes the above data, outputting a 12-dimensional fused feature vector for each occupant.

[0055] Risk assessment phase: The motion sickness risk assessment module outputs the risk level and priority of each passenger based on the fused feature vector: child level 4 (overall priority = 1.5 × 0.9 × 4 = 5.4), front passenger level 3 (overall priority = 1.5 × 0.8 × 3 = 3.6), driver level 2 (overall priority = 1.0 × 0.5 × 2 = 1.0), with the priority order being child > front passenger > driver.

[0056] Control Phase: The suspension control module retrieves the baseline parameters based on the highest priority (child, level 4), combines them with the real-time pitch angle correction parameters, and outputs adjustment commands to the air suspension. The adjustment time is 18ms. After driving for 5 minutes, the child risk level rises to level 5 and lasts for 3 seconds. The system triggers emergency measures: the rear seats start massaging and play soothing music, the central control screen pushes tips, and the instrument panel prompts the driver to slow down.

[0057] Feedback optimization phase: The front passenger reported "moderate" and the child reported "discomfort relieved." The system adjusted the weight of the child's corresponding visual features from 0.35 to 0.4. At this time, the hardware fault self-diagnosis module detected abnormal signals from the driver's bioelectrode (abnormality rate 20%) and automatically activated the redundancy scheme, using the driver's head movement frequency and posture data to replace physiological signals to assess risk. The vehicle according to a third aspect embodiment of the present invention includes the aforementioned anti-motion sickness control system. By setting up the aforementioned anti-motion sickness control system, the vehicle's driving status and the real-time physiological status of the driver and passengers can be obtained, and vehicle parameters can be adjusted accordingly to prevent motion sickness, facilitating precise, comfortable, and effective prevention and control, and improving the driving and riding experience.

[0058] The preferred embodiments of the present invention have been described in detail above, but the present invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention, and these equivalent modifications or substitutions are all included within the scope defined by the claims of this application.

Claims

1. A motion sickness control method based on multimodal perception, characterized in that, Includes the following steps: Collect vehicle body status data and multimodal status data of each person in the vehicle. Then, use a fusion algorithm to fuse the vehicle body status data and the multimodal status data of each person to obtain fused data of each person. Based on the fused data from each individual, a deep learning model is used to output the motion sickness risk level for each individual. Priority levels for each individual are calculated based on their motion sickness risk level and individual characteristics. Based on the motion sickness risk level of the person with the highest priority, adjust vehicle parameters to prevent motion sickness.

2. The motion sickness control method based on multimodal perception according to claim 1, characterized in that: The steps for collecting multimodal state data of each person on the vehicle include: The data collected included skin conductivity, heart rate variability, respiratory rate, pupil diameter change rate, head shaking frequency and amplitude, and sitting pressure distribution of each person on the vehicle.

3. The motion sickness control method based on multimodal perception according to claim 1, characterized in that: The step of calculating the priority level of each person based on their motion sickness risk level and individual characteristics includes: The age and motion sickness sensitivity of each person are obtained from the human-computer interaction module as individual characteristic information.

4. The motion sickness control method based on multimodal perception according to claim 1, characterized in that: It also includes the following steps: Monitor the motion sickness risk level of each person. If the motion sickness risk level of any person is greater than or equal to a preset level threshold and the duration is greater than a preset duration threshold, adjust the suspension parameters and cabin parameters to prevent motion sickness.

5. The motion sickness control method based on multimodal perception according to claim 1, characterized in that: The steps for adjusting vehicle parameters include: Adjust the suspension damping and / or stiffness parameters according to the preset risk level-suspension parameter mapping relationship.

6. The motion sickness control method based on multimodal perception according to claim 5, characterized in that: It also includes the following steps: The comfort feedback data is obtained from the human-computer interaction module, and the fusion algorithm and / or the risk level-suspension parameter mapping relationship are optimized based on the comfort feedback data.

7. A motion sickness prevention control system, characterized in that: It includes a multimodal perception module, a data fusion processing module, a motion sickness risk assessment module, a suspension control module, an occupant interaction module, and a communication module. These modules work together to implement the motion sickness prevention control method based on multimodal perception as described in any one of claims 1 to 6.

8. The motion sickness control system according to claim 7, characterized in that: The multimodal sensing module includes flexible bioelectrodes, cabin millimeter-wave radar, infrared cameras, seat pressure sensor arrays, and vehicle body status sensors.

9. The motion sickness control system according to claim 7, characterized in that: The communication module adopts a hybrid communication architecture of Ethernet and CAN bus.

10. A vehicle, characterized in that: Including the motion sickness control system as described in claim 7.

Images