A vestibular anti-slosh control strategy-based 3D system for preventing dizziness

Abstract

The application discloses a vestibular vibration anti-dizzy 3D system based on an Anti-slosh control strategy, comprising an inertial measurement unit, a control module and a vibration stimulation module; the inertial measurement unit collects the instantaneous angular velocity of head and neck pitching, rolling and yawing, and extracts the dominant oscillation frequency; the control module adopts a zero vibration derivative input shaper to suppress residual oscillation; and the shape motion signal is converted into a stimulation parameter, and the vibration parameter is adjusted through a smoothing factor to ensure that the stimulation is synchronized with the head and neck movement; the vibration stimulation module transmits the vibration parameter to a vibration motor fixed at the mastoid part, mechanically transmits the vibration to the inner ear through bone conduction, realizes personalized vestibular stimulation to offset the sensory conflict caused by the non-expected head and neck movement, and realizes closed-loop adaptive vestibular stimulation. In the objective evaluation index, the motion sickness index based on the cumulative head and neck movement is reduced by an average of 50.69%, and the cumulative speed of the CS risk is reduced by 8.69%.

Description

Technical Field

[0001] This invention relates to human interaction comfort optimization technology in the fields of virtual reality (VR), augmented reality (AR) and mixed reality (MR), specifically to a vestibular vibration anti-sickness 3D system based on an anti-slosh control strategy, which is suitable for alleviating cybersickness (CS) in immersive 3D environments and improving the human-computer interaction experience. Background Technology

[0002] Immersive 3D technologies such as virtual reality and augmented reality have been widely applied in entertainment, medical rehabilitation, military training, and industrial design, achieving application effects that surpass traditional methods by providing a sense of spatial presence and concrete interaction. However, motion sickness is a key bottleneck restricting its large-scale adoption. 30%-70% of users experience symptoms such as nausea, eye strain, and postural instability, leading to a 40% decrease in task completion rates in enterprise scenarios and 58% of consumer VR usage time being less than 30 minutes. The pathogenesis of motion sickness mainly stems from three theories: sensory conflict, postural instability, and neural mismatch. The core contradiction lies in the mismatch between visual and vestibular signals and the convergence-accommodation conflict. Existing mitigation solutions mainly focus on hardware optimization (such as field of view limitation and latency compensation), software improvement (such as motion platform adaptation), and machine learning prediction, but these suffer from high implementation costs, closed-loop response delays, and poor individual adaptability.

[0003] Input shaping is a widely used feedforward control method for vibration control of flexible systems. This technique generates a shaped command signal to drive the system by convolving the desired reference input with a specifically designed pulse sequence. The residual oscillations are canceled out by the superposition effect of multiple delayed pulses, achieving zero residual vibration after a finite time. In the development of input shaping technology, the ZeroVibration (ZV) shaper achieves vibration suppression through the timing design of two pulses, but it is sensitive to changes in system parameters. To improve robustness, Singer and Seering proposed the ZeroVibration Derivative (ZVD) shaper, which significantly enhances tolerance to frequency identification errors by setting the derivative of the percentage of residual oscillations with respect to frequency to zero at the system's resonant frequency. The ZVD shaper consists of three pulses, whose timing and amplitude are determined based on the system's natural frequency and damping ratio, and can maintain a low level of residual vibration even with modeling errors.

[0004] In recent years, input shaping technology has been successfully applied to fields such as flexible structures for spacecraft, anti-sway systems for cranes, and precision manufacturing equipment. However, in the field of virtual reality motion sickness prevention, existing technologies have not yet applied the principles of input shaping to vestibular stimulation control. Human head and neck movements can be viewed as a flexible system with specific oscillatory characteristics. The human skull can be likened to a container, and the soft tissues and blood in the brain to the fluid within the container. Under visually induced motion stimulation, unintended oscillations are generated, which then decay due to viscosity. However, existing bone conduction vibration systems use open-loop control with fixed vibration frequency and amplitude, which cannot adapt to individual physiological differences and scene changes, limiting their anti-motion sickness effect. A closed-loop adaptive optimization scheme is urgently needed. Summary of the Invention

[0005] This invention aims to provide a vestibular vibration anti-motion sickness 3D system based on an anti-slosh control strategy. By dynamically capturing head and neck movement information, it generates personalized vestibular vibration stimulation to balance the conflict between visual and vestibular signals. It also introduces input shaping technology into vestibular vibration control, using a ZVD shaper to generate an inhibitory signal that is opposite to the head and neck oscillation, thereby alleviating motion sickness in 3D environments from the root cause and improving the comfort and duration of the user's immersive experience.

[0006] This invention discloses a vestibular vibration anti-dizziness 3D system based on an Anti-slosh control strategy, which includes an inertial measurement unit, a control module, and a vibration stimulation module.

[0007] The inertial measurement unit collects the instantaneous angular velocities of the head and neck in the pitch, roll, and yaw directions in real time, and extracts the dominant oscillation frequency based on a weighted frequency estimation algorithm;

[0008] The control module uses a zero-vibration derivative input shaper to suppress residual oscillations; and converts the shape motion signal into stimulation parameters while compensating for system transmission delay. Vibration parameters are dynamically adjusted through a smoothing factor to ensure that stimulation is synchronized with head and neck movements.

[0009] The vibration stimulation module transmits the processed vibration parameters to a vibration motor fixed to the mastoid process, which then transmits mechanical vibrations to the inner ear via bone conduction. This achieves personalized vestibular stimulation to counteract sensory conflicts caused by unintended head and neck movements, thus realizing closed-loop adaptive vestibular stimulation.

[0010] Furthermore, the dominant oscillation frequency is:

[0011] ;

[0012] As the dominant oscillation frequency, The pitch angular velocity, For roll angular velocity, This is the yaw rate.

[0013] Furthermore, the instruction after suppressing residual oscillations Generated by delayed pulse weighted convolution:

[0014] ;

[0015] in, Indicates the reference motion signal, For type i gain, For type i delay time, based on the system's oscillation parameters Determine the gain and delay time:

[0016] ;

[0017] For the oscillation parameters of the system, , These are the measurement parameters for magnetic resonance elastography.

[0018] Furthermore, the stimulation frequency is expressed as:

[0019] ;

[0020] For stimulation frequency, For the previous moment ( The calculated stimulus frequency values, As a smoothing factor, The dominant oscillation frequency at the current moment, This is due to system transmission delay;

[0021] The stimulus phase is represented as:

[0022] ;

[0023] For the stimulus phase, For phase wrap function, The current phase of head and neck movement. This is the current dominant oscillation frequency. This is due to system transmission delay;

[0024] The stimulus amplitude is expressed as:

[0025] ;

[0026] in, To stimulate amplitude, For the normalization term of the reshaped instructions, This is the command after shaping with zero vibration derivative. The function identifier limits the amplitude to a safe range. ], This is the minimum amplitude. This represents the maximum amplitude.

[0027] Furthermore, the inertial measurement unit is integrated into the head-mounted display device.

[0028] Furthermore, the vibration stimulation module includes at least two bone conduction vibration motors, which are symmetrically fixed to the mastoid regions on both sides of the head-mounted display device.

[0029] The beneficial effects achieved by this invention are:

[0030] Compared to existing studies on vestibular bone conduction anti-spondylosis stimulation, which often employs an open-loop stimulation mode with fixed frequency and amplitude, this invention treats the cranial system as a container, and the blood and soft tissues within the brain that can undergo relative movement as the liquid inside the container. Based on the dynamic characteristics of head and neck vibration response under external stimulation, and drawing on the input shaping concept of suppressing liquid sloshing in Anti-slosh control, the invention adaptively generates the stimulation frequency, phase, and amplitude parameters of vestibular bone conduction vibration according to the real-time detected head and neck movement state, thereby avoiding the problem of the traditional fixed stimulation mode being disconnected from the actual movement process.

[0031] Based on comparative experiments, the following results were achieved compared to existing technologies:

[0032] Effectively suppresses head and neck movement: Among objective assessment indicators, the motion sickness index (CSI) based on cumulative head and neck movement decreased by an average of 50.69%, and the cumulative rate of CS risk decreased by 8.69%.

[0033] Increased body balance: The average area of ​​the 90% confidence ellipse calculated based on the center of pressure (COP) of the foot decreased by 9.24%, effectively improving postural stability.

[0034] Significant relief of subjective discomfort: In the Kennedy SSQ (Subjective Symptom Questionnaire) assessment, disorientation was reduced by an average of 11.08%, and nausea was reduced by an average of 3.64%.

[0035] Highly adaptable: The closed-loop control system can capture and compensate for individual differences and scene changes in real time, overcoming the limitation of poor adaptability of the traditional open-loop BCV scheme. Attached Figure Description

[0036] Figure 1 System control flowchart;

[0037] Figure 2 The system's hardware connection diagram and application environment diagram are shown.

[0038] 1-Inertial measurement unit, 2-Vibration stimulation module, 3-Control module. Detailed Implementation

[0039] The present invention will be further described below with reference to specific embodiments, and the advantages and features of the present invention will become clearer as a result. However, these embodiments are merely exemplary and do not constitute any limitation on the scope of the present invention. Those skilled in the art should understand that modifications or substitutions can be made to the details and form of the technical solutions of the present invention without departing from the spirit and scope of the present invention, but all such modifications and substitutions fall within the protection scope of the present invention.

[0040] like Figure 1 As shown, the present invention provides a vestibular vibration anti-nausea 3D system based on the Anti-slosh control strategy. The vestibular vibration anti-nausea 3D system based on the Anti-slosh control strategy includes an inertial measurement unit 1, a control module 3, a vibration stimulation module 2, and a detection module.

[0041] An inertial measurement unit (IMU)1 is integrated inside a head-mounted display (HMD) (such as the MPU6050) to collect instantaneous angular velocity data of the user's head and neck in the pitch, roll, and yaw directions in real time.

[0042] The inertial measurement unit (IMU) 1, integrated into the head-mounted display (HMD), is used to acquire instantaneous angular velocities in the pitch, roll, and yaw directions of the head and neck in real time. , , ).

[0043] The dominant oscillation frequency is extracted based on the weighted frequency estimation algorithm. As an indicator for quantifying the degree of head and neck slosh, its formula is:

[0044] ;

[0045] The control module (Micro-Controller) 3 uses a high-performance microcontroller (such as an Arduino UNO) and incorporates an Anti-slosh closed-loop adaptive control algorithm. This module is responsible for data processing, stimulus parameter calculation, and PWM signal output.

[0046] Control module 3 uses a zero-vibration derivative (ZVD) input shaper to suppress residual oscillations and avoid secondary mode excitation.

[0047] The ZVD shaper generates a shaped command u(t) by convolving a reference motion signal r(t) with three delayed pulses. This command is designed to produce an equivalent suppression signal that is opposite to the head and neck oscillations, thus canceling out residual oscillations. (Shaped command) It is generated by delayed pulse weighted convolution, and its formula is:

[0048] ;

[0049] in Represents the reference motion signal, based on the system's oscillation parameters. Determine the gain ( ) and delay time ( ),here The value of 0.23 is derived from magnetic resonance elastography measurements of healthy brain tissue from different age groups. The formula for gain A is:

[0050] ;

[0051] Delay time , , , The natural frequency of head and neck movements was determined through preliminary experiments to ensure that the orthopedic commands effectively suppressed residual head and neck oscillations; the measured... Then you can Converted into shaping instructions by the shaper This plastic surgery instruction The aim is to generate an inverse and equivalent suppression signal to head and neck oscillations, ensuring that the orthopedic command can effectively suppress residual head and neck oscillations.

[0052] Control module 3 further calculates the anti-slosh stimulation parameters:

[0053] The shape motion signal is converted into stimulation parameters (stimulation frequency, stimulation phase, stimulation amplitude), while simultaneously compensating for system transmission delay. ), through smoothing factor ( The vibration parameters are dynamically adjusted to ensure synchronization between the stimulus and head and neck movements. The functions of the three stimulus parameters are as follows:

[0054] Stimulation frequency: synchronized with the dominant oscillation frequency of the head and neck to achieve resonance inhibition;

[0055] Stimulation phase: The head and neck movements are counteracted by anti-phase superposition (phase difference π);

[0056] Stimulus amplitude: proportional to the amplitude of the shaping command, controlling the intensity of anti-slosh inhibition.

[0057] The formula for the stimulation parameters is:

[0058] Stimulation frequency: ;

[0059] Stimulus phase: ;

[0060] Stimulation amplitude: ,in The normalization term is defined as follows: The maximum value within the sliding time window, while the clip function limits the amplitude to a safe range. ].

[0061] The control module 3 transmits the processed vibration parameters (frequency, phase, amplitude) to the vibration stimulation module 2. The vibration stimulation module 2 transmits mechanical vibration to the inner ear through bone conduction via a vibration motor fixed to the mastoid region, thereby achieving personalized vestibular stimulation and simultaneously counteracting sensory conflicts caused by undesirable head and neck movements, thus realizing closed-loop adaptive vestibular stimulation.

[0062] The vibration stimulation module (BCV)2 contains at least two bone conduction vibration motors (such as Telesky 1027) that are symmetrically fixed to the mastoid regions on both sides of the HMD to output personalized mechanical vibrations, the parameters of which include stimulation frequency, stimulation phase and stimulation amplitude.

[0063] The detection module is used to assist in verifying the anti-dizziness effect and may include a pressure platform for measuring postural stability and monitoring equipment for collecting physiological indicators such as heart rate, blood pressure and blood oxygen saturation.

[0064] In one embodiment, this implementation method selects a healthy male subject weighing 60kg and aged 24 years to experience a VR roller coaster scene as a specific example, and elaborates on the entire process of system deployment, operation and effect verification, so that the technical solution is reproducible.

[0065] like Figure 2 As shown, the MPU6050 is fixed to the corresponding position of the occipital bone inside the head display using a patch process to ensure that the IMU coordinate system coincides with the head motion coordinate system—the X-axis is along the front-back direction of the head (forward is positive), the Y-axis is along the left-right direction of the head (right is positive), and the Z-axis is along the vertical direction of the head (up is positive), and the IMU zero drift error is calibrated (static angular velocity drift ≤ 0.5° / s).

[0066] Two Telesky1027 motors are symmetrically fixed to the inside of the earcups on both sides of the head-mounted display using a medical-grade silicone bracket. The center of the motor is aligned with the center of the mastoid process (deviation ≤2mm). The bracket stiffness is set to Shore A 30A (to balance stability and wearing comfort) to prevent motor vibration from being transmitted to the head-mounted display body and causing additional interference.

[0067] The Arduino UNO connects to the head-mounted display power supply module via Micro-USB (to obtain a stable 5V voltage), communicates with the IMU via the I2C bus (SDA pin A4, SCL pin A5), connects to the vibration motor drive circuit via digital pin 9 (PWM output), and receives data from the detection device via the Bluetooth module (HC-05). All wiring uses shielded wires to reduce electromagnetic interference.

[0068] After installation, system-level calibration is performed—start the VR scene and have the subject perform standard head movements (pitch ±30°, roll ±20°, yaw ±45°), compare the data collected by the IMU with the data from a high-precision motion capture system (such as OptiTrack), and ensure that the measurement error of the motion parameters is ≤5%; adjust the drive voltage of the vibration motor so that the amplitude and the duty cycle of the PWM signal are linearly related (duty cycle of 5%-95% corresponds to amplitude of 0.1-1.0g).

[0069] After completing the preliminary work, the execution of the program is controlled, which includes the following steps:

[0070] Step 1: Motion detection is performed. The IMU collects the angular velocity of the subject's head and neck, measuring the angular velocities in the pitch, roll, and yaw directions. , , Then, the dominant oscillation frequency is calculated using the angular velocities in these three directions, and used as the input for the Anti-slosh algorithm.

[0071] ;

[0072] Step two: Generate shaping instructions using the ZVD algorithm to suppress residual head and neck oscillations.

[0073] In the example, the damping ratio 0.23, Calculate the oscillation parameters according to the theory of second-order vibration systems. = Based on the K value, the gain and delay of the ZVD input shaper are further derived:

[0074] ;

[0075] ;

[0076] ;

[0077] Delay time , , , The natural frequency of head and neck movement was determined through preliminary experiments to ensure that the shaping command could effectively suppress residual oscillations in the head and neck.

[0078] ;

[0079] Measured through preliminary experiments Then you can Converted into shaping instructions by the shaper This plastic surgery instruction The aim is to generate an inverse and equivalent suppression signal to head and neck oscillations, ensuring that the orthopedic command can effectively suppress residual head and neck oscillations.

[0080] Step 3: Calculate the frequency of the final vibration stimulus by combining system delay and motion data. Phase ,amplitude

[0081] Known system delay Smoothing time constant Then the smoothing factor .

[0082] The stimulation frequency can be obtained by combining the dominant oscillation frequency measured in step one. ;

[0083] After obtaining the stimulus frequency, input the stimulus phase. It can obtain the phase of head and neck movements.

[0084] Combined with the shaping instructions obtained in step two To calculate the amplitude of the stimulus ,

[0085] Step four, calculate the amplitude Convert to PWM duty cycle to calculate the stimulus parameters. The drive motor outputs mechanical vibration to achieve anti-sway control of the head and neck.

[0086] Then, a new round of motion detection is performed to achieve real-time, closed-loop control of the program.

[0087] To ensure the reliability and effectiveness of the proposed solution under real-world complex working conditions, a multi-dimensional real-vehicle comparative verification experiment was conducted. By collecting physiological feedback and subjective feelings from subjects in typical dynamic scenarios, the system's ability to suppress motion sickness and its effect on improving human posture stability were quantitatively evaluated. The following is a detailed analysis of the core experimental data:

[0088] 1. This application, referencing the existing RevisedMSI calculation model, focuses on the individual's head and neck dynamic response under actual motion stimuli, comprehensively considering the cumulative effect of head movement in pitch, roll, and yaw directions, and constructs a motion sickness quantitative calculation method oriented towards individual differences. In this application, the experiment no longer directly collects the dynamic response of the vehicle or seat, but instead acquires the individual's head and neck dynamic parameters under motion stimuli in real time through sensors, including linear acceleration and angular velocity in the pitch, roll, and yaw directions. The acceleration is dynamically weighted using the real-time angular motion characteristics of the head, making the quantification process closer to the individual's actual neural control response under motion stimuli. Simultaneously, an individual motion sickness sensitivity coefficient is introduced to construct a motion sickness quantitative calculation method oriented towards individual differences.

[0089] The cumulative head and neck motion sickness index (CSI) was finally measured. Compared with the open-loop stimulation mode in the prior art, the stimulation system provided by the present invention reduced the CSI based on cumulative head and neck motion by an average of 50.69% and reduced the accumulation rate of motion sickness (CS) risk by 8.69%. This index can more realistically represent the actual response of external stimuli after personalized human posture adjustment, and provides a dynamic basis for the regression of group motion sickness risk.

[0090] 2. This application also conducts a comparative experiment with existing technical solutions based on COP parameters. Based on the 90% confidence ellipse area calculated by the Center of Pressure (COP), the displacement trajectories of the subjects in the forward-backward (AP) and left-right (ML) directions are recorded in different VR or sports scenarios. The larger the ellipse area, the greater the amplitude of the subject's posture swing under visual / motor stimulation, and the higher the regulatory load required to maintain balance.

[0091] Experimental results show that, through the intervention of this system, the area of ​​the 90% confidence ellipse of COP was reduced by an average of 9.24%. The significant reduction in area indicates that the amplitude of the subject's body sway was effectively controlled, reducing the compensatory posture adjustment pressure required to maintain balance. The experiment confirms that high dynamic and high-conflict visual scenes will increase body sway, and this system can effectively alleviate such disturbances.

[0092] 3. This application also conducts a comparative experiment based on the Subjective Discomfort Symptom Questionnaire (SSQ) and existing technical solutions. The Kennedy Simulator Symptom Questionnaire (SSQ), currently the most authoritative questionnaire in the field of virtual reality and motion simulation, is used. Subjects rate 16 physiological symptoms (such as general discomfort, headache, eye strain, nausea, etc.) on a 4-level scale from 0 to 3. The scores are then weighted into three core dimensions: nausea, eye movement stress, and disorientation.

[0093] Subjective symptom assessment results showed improvement in all indicators for the subjects, with an average reduction of 11.08% in disorientation and 3.64% in nausea. The decrease in subjective scores was highly consistent with the improvement in objective dynamic index (CSI) and postural stability index (COP), demonstrating the comprehensive effectiveness of the system in relieving motion sickness.

[0094] The above are merely preferred embodiments of the present invention and do not constitute any limitation on the scope of protection of the present invention; all technical solutions formed by equivalent transformations or equivalent substitutions fall within the scope of protection of the present invention; the parts of the present invention not described in detail are well-known technologies to those skilled in the art.

Claims

1. A vestibular vibration anti-sickness 3D system based on an anti-slosh control strategy, characterized in that, The vestibular vibration anti-dizziness 3D system based on the Anti-slosh control strategy includes an inertial measurement unit (1), a control module (3), and a vibration stimulation module (2). The inertial measurement unit (1) collects the instantaneous angular velocities of the head and neck in the pitch, roll and yaw directions in real time, and extracts the dominant oscillation frequency based on the weighted frequency estimation algorithm; ； As the dominant oscillation frequency, The pitch angular velocity, For roll angular velocity, Yaw angular velocity; The control module (3) uses a zero-vibration derivative input shaper to suppress residual oscillations; and transmits the command after suppressing residual oscillations. The signals collected by the inertial measurement unit (1) are converted into stimulation parameters, including stimulation frequency, stimulation phase, and stimulation amplitude; and at the same time, the system transmission delay is compensated, and the vibration parameters are dynamically adjusted by the smoothing factor to ensure that the stimulation is synchronized with the head and neck movement; The vibration stimulation module (2) transmits the processed vibration parameters to the vibration motor fixed at the mastoid region, and transmits mechanical vibration to the inner ear through bone conduction, thereby realizing personalized vestibular stimulation to counteract the sensory conflict caused by undesired head and neck movements and realize closed-loop adaptive vestibular stimulation. Instructions after suppressing residual oscillations Generated by delayed pulse weighted convolution: ； in, Indicates the reference motion signal, For type i gain, For type i delay time, based on the system's oscillation parameters Determine the gain and delay time: ； For the oscillation parameters of the system, , These are the parameters measured in magnetic resonance elastography. Stimulation frequency is expressed as: ； For stimulation frequency, For the previous moment ( The calculated stimulus frequency values, As a smoothing factor, The dominant oscillation frequency at the current moment, This is due to system transmission delay; The stimulus phase is represented as: ； For the stimulus phase, For phase wrap function, The current phase of head and neck movement. This is the current dominant oscillation frequency. This is due to system transmission delay; The stimulus amplitude is expressed as: ； in, To stimulate amplitude, For the normalization term of the reshaped instructions, This is the command after shaping with zero vibration derivative. The function identifier limits the amplitude to a safe range. ], This is the minimum amplitude. This represents the maximum amplitude.

2. The vestibular vibration anti-dizziness 3D system based on the Anti-slosh control strategy according to claim 1, characterized in that, The inertial measurement unit (1) is integrated into the head-mounted display device.

3. The vestibular vibration anti-dizziness 3D system based on the Anti-slosh control strategy according to claim 2, characterized in that, The vibration stimulation module (2) includes at least two bone conduction vibration motors, which are symmetrically fixed to the mastoid regions on both sides of the head-mounted display device.

Images