An adaptive coma wake-up vehicle system and control method after a vehicle accident
The adaptive coma wake-up system after vehicle accidents utilizes millimeter-wave radar and a flexible pressure sensor array combined with environmental sensors to fuse multi-source data, achieving accurate coma level judgment and graded wake-up. This solves the problems of false coma judgment and low wake-up efficiency in existing systems, improving safety and effectiveness after vehicle accidents.
Patent Information
- Application Number
- CN202511875700.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-12
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2045-12-12
AI Technical Summary
Existing vehicle systems cannot accurately determine the unconscious state of occupants after an accident, have low wake-up efficiency, and lack adaptive and feedback mechanisms, leading to misjudgments or missed judgments, and failing to provide effective intervention within the golden rescue time.
By combining millimeter-wave radar modules and flexible pressure sensor arrays with environmental sensors, an adaptive wake-up strategy is implemented through a multi-mode execution unit. The central controller performs multi-source data fusion and feature extraction to achieve coma level judgment and graded wake-up.
It improves the accuracy and anti-interference ability of coma recognition, dynamically adjusts the intensity of stimulation, adapts to the needs of complex scenarios, reduces the risk of misjudgment, and improves the effectiveness and safety of awakening.
Smart Images

Figure CN121469586B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of vehicle safety assistance technology, and in particular to an adaptive coma wake-up vehicle system and control method after a vehicle accident. Background Technology
[0002] With the continuous increase in car ownership, the incidence of vehicle accidents remains high. In serious traffic accidents, occupants often fall into varying degrees of unconsciousness due to severe impact or brain injury. Traditional safety systems focus on passive protection but lack real-time monitoring and effective intervention for post-injury physiological states. If the injured person cannot be found and awakened or called for help within the golden rescue time, the probability of survival will be significantly reduced.
[0003] Existing vehicle safety monitoring technologies face the following shortcomings: First, the limitations of single-sensor operation: most systems use a single sensor, resulting in a limited sensing modality that is easily affected by environmental or posture interference, making it impossible to comprehensively and accurately determine the true coma state of the occupant. Second, static judgment mechanisms: the determination of coma status often relies on fixed physiological thresholds, ignoring significant individual physiological differences and easily leading to misjudgments or missed judgments. Third, lack of adaptation and feedback: existing intervention measures are usually pre-set, lacking tiered interventions for different coma levels. Furthermore, the intervention process lacks a closed-loop feedback mechanism, failing to dynamically adjust the stimulation intensity based on the occupant's real-time physiological responses, affecting awakening efficiency and safety.
[0004] Therefore, there is an urgent need to develop a robust, accurate, and adaptive post-accident coma awakening system that can accurately determine the coma status of occupants and implement targeted adaptive awakening strategies to ensure precise and efficient safety intervention for occupants within the golden rescue time. Summary of the Invention
[0005] In view of this, the present invention provides an adaptive coma wake-up vehicle system and control method after a vehicle accident, in order to solve the technical problems in the prior art where the vehicle system has low efficiency in waking up occupants, inaccurate judgment, and is unable to implement safe and effective auxiliary rescue after a vehicle accident.
[0006] To achieve the above-mentioned technical objectives, the present invention adopts the following technical solution:
[0007] On one hand, the present invention provides an adaptive coma wake-up vehicle system after a vehicle accident, including a central controller, and a millimeter-wave radar module, a flexible pressure sensing array, an environmental sensor group and a multi-mode execution unit connected to the central controller;
[0008] The millimeter-wave radar module is installed inside the vehicle seat back to collect the occupant's chest displacement and obtain respiratory signals in a non-contact manner.
[0009] A flexible pressure sensor array is installed on the underside of the seat cushion and backrest cover to collect micro-vibration signals of the occupant's body surface.
[0010] An environmental sensor array, including interior and exterior temperature and humidity sensors and a noise detector, is used to collect environmental parameters;
[0011] The central controller is used to extract respiratory rate from respiratory signals and micro-vibration energy value from micro-vibration signals on the body surface. Based on respiratory rate and micro-vibration energy value, it determines the degree of coma of the occupant and controls the multi-mode execution unit to execute the wake-up strategy according to the temperature and humidity inside and outside the vehicle, noise information and the degree of coma.
[0012] The multi-mode execution unit includes at least a pneumatic vibration generator and a temperature stimulation module, used to implement graded wake-up of the occupants according to the instructions of the central controller.
[0013] Furthermore, the sampling frequency of the millimeter-wave radar module is not lower than 20Hz.
[0014] Furthermore, the flexible pressure sensing array contains at least 32 sensing units, with a spatial resolution of no more than 2 cm and a sampling frequency of no less than 100 Hz.
[0015] Furthermore, the multi-mode execution unit also includes: a light stimulation unit and an acoustic stimulation unit;
[0016] The pneumatic vibration generator, temperature stimulation module, light stimulation module, and sound stimulation module can be activated individually or in combination. According to the instructions of the central controller, the stimulation position, action area, and parameters are adjusted according to the occupant's body shape adaptive adjustment mechanism to adapt to different body shapes, weights, and sitting postures of the occupants.
[0017] On the other hand, the present invention also discloses a method for controlling coma and waking up after a vehicle accident, applied to the system described above, comprising:
[0018] After a vehicle accident occurs, the central controller activates the millimeter-wave radar module and pressure sensor array to collect the occupants' breathing signals and micro-vibration signals on their body surface.
[0019] The central controller extracts respiratory frequency from respiratory signals and micro-vibration energy values from body surface micro-vibration signals based on a preset feature extraction method.
[0020] The occupant's coma status and coma level are determined based on respiratory rate and micro-vibration energy values;
[0021] Based on the determined level of coma and environmental data collected by environmental sensors, the multi-mode execution unit is controlled to implement the corresponding graded wake-up strategy.
[0022] Furthermore, the preset feature extraction method includes:
[0023] Respiratory rate is obtained by extracting respiratory cycle information through time-domain peak detection or short-time Fourier transform;
[0024] High-frequency component filtering and statistical analysis were performed on the micro-vibration signal on the body surface to extract the intensity, frequency and spatial distribution characteristics of the micro-vibration and obtain the micro-vibration energy value.
[0025] Furthermore, the method of determining the occupant's coma status and coma level based on respiratory rate and micro-vibration energy value includes:
[0026] Millimeter-wave radar is used as the primary data source, and a flexible pressure sensing array is used as the secondary data source.
[0027] The basic confidence level of the first data source is calculated based on the respiratory rate, and the basic confidence level of the second data source is calculated based on the micro-vibration energy.
[0028] Using Dempster's composition rules, the two data sources are merged based on the basic confidence assignments of the first and second data sources to obtain the merged basic confidence assignments;
[0029] The fusion belief degree is determined based on the basic credibility allocation after fusion. The state with the highest fusion belief degree and greater than the decision threshold is used as the basis for judgment to determine the coma state of the occupant.
[0030] Furthermore, the method of determining the occupant's level of unconsciousness based on respiratory rate and micro-vibration energy value also includes:
[0031] When 8 ≤ f < 12 beats / minute and 0.01 ≤ E < 0.05 g² / Hz, it is considered a mild coma.
[0032] When 4 ≤ f < 8 times / minute and 0.005 ≤ E < 0.01 g² / Hz are met, the condition is considered moderate coma.
[0033] When f < 4 times / minute and E < 0.005 g² / Hz, it is determined to be a deep coma;
[0034] Where f represents the breathing frequency and E represents the micro-vibration energy value.
[0035] Furthermore, the tiered wake-up strategy includes:
[0036] When a patient is determined to be in a mild coma, the pneumatic vibration generator is activated, and the inflation time is set to the first duration.
[0037] When the patient is determined to be in a moderate coma, activate the pneumatic vibration generator and set the inflation time to the second duration.
[0038] When a patient is determined to be in a deep coma, the pneumatic vibration generator is activated and the inflation time is set to the third duration. At the same time, the temperature stimulation module alternately sprays cold water and blows hot air. The temperature of the cold water is set according to the outside temperature of the vehicle, and the temperature of the hot air is set according to the humidity inside the vehicle.
[0039] Furthermore, the method for setting the temperature of the cold water includes:
[0040] When the humidity inside the vehicle is ≤80%, determine whether the outside temperature is below 5℃; when the outside temperature is <5℃, set the temperature of the cold water to 15-20℃; when the outside temperature is ≥5℃, set the temperature of the cold water to 8-15℃.
[0041] When the humidity inside the vehicle exceeds 80%, the cold water spray function will be disabled, and only the hot air and vibration combination mode will be enabled.
[0042] Compared with existing technologies, the adaptive coma wake-up vehicle system and method proposed in this invention have the following advantages:
[0043] (1) This scheme simultaneously uses millimeter-wave radar module to obtain chest cavity respiratory displacement signal and flexible pressure sensor array to obtain body surface micro-vibration signal, forming a multi-channel physiological monitoring system, which significantly improves the detection accuracy of key coma features such as weakened breathing and reduced body micro-reflection, and effectively reduces the risk of misjudgment caused by single sensor being susceptible to posture, obstruction or excessive vibration interference.
[0044] (2) The central controller uses a preset algorithm to extract chest movement features such as respiratory rate, amplitude, energy, and waveform morphology, and simultaneously calculates micro-vibration energy values. This allows for the inference of the occupant's level of coma from multidimensional physiological information, enabling the classification of mild, moderate, and deep coma. Compared to traditional methods that rely solely on a single physiological parameter or fixed threshold, this system can more accurately capture abnormal changes such as respiratory depression and loss of microreflexes, providing a more refined state assessment and a reliable basis for subsequent awakening strategy selection.
[0045] (3) By using temperature, humidity, and noise information collected by environmental sensors, along with the occupant's coma level, as control variables, the aerodynamic vibration stimulation, temperature stimulation, and other awakening methods of the multi-mode actuator can be dynamically adjusted. This allows for enhanced mechanical stimulation in noisy environments and reduced temperature stimulation in low-temperature environments. Furthermore, it automatically selects mild, moderate, or strong stimulation strategies based on the coma level. This adaptive, multi-mode graded awakening mechanism improves the effectiveness of awakening while also avoiding secondary injuries caused by overstimulation, making it suitable for complex scenarios following vehicle accidents. Attached Figure Description
[0046] Figure 1 A schematic diagram of the adaptive coma wake-up vehicle system after a vehicle accident provided by the present invention;
[0047] Figure 2 This is a flowchart illustrating the method for controlling unconsciousness and waking up after a vehicle accident provided by the present invention. Detailed Implementation
[0048] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.
[0049] Example 1
[0050] Please see Figure 1 This embodiment provides an adaptive coma wake-up vehicle system 100 after a vehicle accident, including a central controller 104, and a millimeter-wave radar module 101, a flexible pressure sensor array 102, an environmental sensor group 103 and a multi-mode execution unit 105 connected to the central controller 104.
[0051] The millimeter-wave radar module 101 is installed inside the seat back of the vehicle and is used to collect the occupant's chest displacement in a non-contact manner to obtain respiratory signals.
[0052] A flexible pressure sensor array 102 is installed on the underside of the seat cushion and backrest cover to collect micro-vibration signals of the occupant's body surface.
[0053] The environmental sensor group 103 includes vehicle interior and exterior temperature and humidity sensors and a noise detector, used to collect environmental parameters;
[0054] The central controller 104 is used to extract respiratory rate from respiratory signals and micro-vibration energy value from body surface micro-vibration signals. Based on respiratory rate and micro-vibration energy value, it determines the degree of coma of the occupant and controls the multi-mode execution unit to execute the wake-up strategy according to the temperature and humidity inside and outside the vehicle, noise information and the degree of coma.
[0055] The multi-mode execution unit 105 includes at least a pneumatic vibration generator and a temperature stimulation module, used to implement graded wake-up of the occupants according to the instructions of the central controller.
[0056] The adaptive coma wake-up system provided in this embodiment collects occupant chest breathing signals using millimeter-wave radar embedded in the seat back, combines this with a flexible pressure sensor array embedded in the seat cushion and backrest to acquire micro-vibration signals from the body surface, and utilizes an environmental sensor array to provide data on temperature, humidity, and noise levels inside and outside the vehicle. The central controller preprocesses and extracts features from the multi-source data to obtain respiratory frequency, amplitude, and micro-vibration energy values, and determines the occupant's level of unconsciousness accordingly. Based on the level of unconsciousness and environmental conditions, it adaptively controls aerodynamic vibration and temperature stimulation modules to implement graded wake-up. This solution improves the accuracy and anti-interference capability of unconsciousness recognition through multimodal fusion of radar and pressure array; achieves more reliable unconsciousness recognition through feature extraction and graded determination; and improves the effectiveness and safety of wake-up by combining environmental and physiological states, adapting to the complex wake-up needs of vehicle accident scenarios.
[0057] In a preferred embodiment, the millimeter-wave radar module has a sampling frequency of ≥20Hz and is capable of penetrating clothing to detect chest displacement at the level of 0.1mm.
[0058] It's important to note that the normal breathing frequency is approximately 0.2Hz to 0.5Hz. A sampling rate of 20Hz is significantly higher than the Nyquist frequency, enabling clear and distortion-free acquisition of respiratory waveforms. This provides precise data support for subsequent analyses of respiratory amplitude variance and irregularity. Furthermore, in moderate or deep coma, occupant breathing becomes extremely weak and shallow, with chest displacement potentially far less than normal. An accuracy of 0.1mm ensures the system can detect these millimeter-level or even sub-millimeter-level vital signs, preventing the misinterpretation of faint breathing as respiratory arrest.
[0059] In a preferred embodiment, the flexible pressure sensor comprises at least 32 sensing units, with a spatial resolution ≤2cm and a sampling frequency ≥100Hz.
[0060] High spatial resolution can accurately distinguish the differences in force experienced by different body types and sitting postures, improving the accuracy of extracting micro-vibration energy and vibration patterns from the body surface; while high sampling frequency can capture high-frequency micro-vibration changes at the heartbeat level, avoiding signal aliasing or missed detection, making weak physiological reflexes clearer. This configuration significantly enhances the reliability and robustness of coma recognition and provides more accurate basic data for subsequent body type adaptive adjustment and wake-up strategy optimization.
[0061] In a preferred embodiment, the multi-mode execution unit further includes: a light stimulation unit and an acoustic stimulation unit;
[0062] The pneumatic vibration generator, temperature stimulation module, light stimulation module, and sound stimulation module can be activated individually or in combination. According to the instructions of the central controller, the stimulation position, action area, and parameters are adjusted according to the occupant's body shape adaptive adjustment mechanism to adapt to different body shapes, weights, and sitting postures of the occupants.
[0063] As a specific embodiment, the pneumatic vibration generator includes an airbag, an air pump, a control valve assembly, and connecting pipes. The airbag is fitted between the seat upholstery and the foam, and directly contacts the occupant's back and buttocks. It is used to generate pressure and vibration through deformation, providing mechanical stimulation to the occupant. The air pump provides an air source, compresses the air, and delivers it into the airbag. The control valve assembly precisely controls the airflow direction and speed, determining the airbags to be inflated or deflated, the inflation / deflation sequence, and the timing based on instructions from the central controller. The connecting pipes connect the air pump, the valve assembly, and each airbag, forming an air transmission path.
[0064] Furthermore, the airbag is also compatible with car seat massage functions; by sharing a single set of hardware with the high-safety-requirement pneumatic wake-up function and the comfort massage function, the complexity and manufacturing cost of the system are significantly reduced, avoiding the need to install two separate actuators.
[0065] In some embodiments, the number of backrest airbags is ≥6 and the number of seat cushion airbags is ≥4.
[0066] The airbags distributed in the backrest and seat area, together with the control valve group, enable the central controller to focus stimulation on specific areas (such as the back, waist or buttocks) according to the coma level and the real-time response of the occupant, improve the targeting of the awakening, and ensure that there are always enough airbags in effective contact with the body regardless of the occupant's body size, providing a uniform and controllable stimulation force.
[0067] Example 2
[0068] This embodiment also provides a method for controlling coma and waking up after a vehicle accident, applied to the system described in Embodiment 1, such as... Figure 2 As shown, the method includes:
[0069] Step S101: After a vehicle accident occurs, the central controller activates the millimeter-wave radar module and pressure sensor array to collect the occupants' breathing signals and micro-vibration signals on their body surface.
[0070] Step S102: The central controller extracts the respiratory rate from the respiratory signal and the micro-vibration energy value from the micro-vibration signal on the body surface based on the preset feature extraction method;
[0071] Step S103: Determine the occupant's coma status and coma level based on respiratory rate and micro-vibration energy value;
[0072] Step S104: Based on the determined coma level and the environmental data collected by the environmental sensors, control the multi-mode execution unit to implement the corresponding graded wake-up strategy.
[0073] In a preferred embodiment, in step S102, the preset feature extraction method includes:
[0074] Respiratory rate is obtained by extracting respiratory cycle information through time-domain peak detection or short-time Fourier transform;
[0075] High-frequency component filtering and statistical analysis were performed on the micro-vibration signal on the body surface to extract the intensity, frequency and spatial distribution characteristics of the micro-vibration and obtain the micro-vibration energy value.
[0076] In some embodiments, the processing of the respiratory signal further includes: quantifying the amplitude of chest displacement or pressure change in each respiratory cycle to obtain respiratory amplitude characteristics;
[0077] Respiratory energy is calculated based on the sum of squares and integrals of respiratory signals, which is used to reflect respiratory intensity and volume.
[0078] By analyzing the rising edge, falling edge, peak and trough characteristics of the respiratory waveform through signal segmentation, normalization and template matching, abnormalities in the respiratory pattern can be determined.
[0079] By quantifying respiratory amplitude characteristics, the system can accurately capture shallow breathing caused by unconsciousness; by calculating respiratory energy, it can provide robust physical indicators of respiratory intensity and volume; and waveform analysis (such as template matching) can identify typical pathological abnormal breathing patterns. These feature mining methods on raw data have yielded a rich and reliable set of physiological features, which helps to enhance personalized judgment of occupant status after an accident.
[0080] Furthermore, the micro-vibration energy value is calculated based on the intensity, energy, and spatial distribution characteristics of micro-vibrations on the body surface, and is used to reflect the level of physiological micro-reflexes of the occupant. Specifically:
[0081] The first step is to apply bandpass filtering to the signal of each sensing unit in the pressure array. (Typically at 5–25 Hz, used to extract heartbeat and micro-reflection components), its root mean square (RMS) value is calculated to reflect the intensity of local micro-vibrations, expressed by the formula:
[0082]
[0083] in, This represents the original micro-vibration displacement signal acquired by the sensing unit. This represents the instantaneous energy of the micro-vibration signal. Indicates the length of the time window.
[0084] The second step is to perform a Fast Fourier Transform (FFT) on each channel to calculate the frequency domain energy within the target frequency band:
[0085]
[0086] in, This indicates that the micro-vibration signal of the i-th sensing unit is at a frequency of Fourier transform amplitude at that point Represents the power spectral density. and These represent the lower and upper limits of frequency for energy calculation, respectively.
[0087] The third step is to calculate the weights based on the geometric layout and stress conditions of each sensing unit in the array, in order to conform to physical laws, such as: the weights are greater in areas of strong vibration, the weights are reduced when there is left-right asymmetry in the spatial distribution correction, and the weights are higher in the central area than at the edge, etc.
[0088] The fourth step is to calculate the micro-vibration energy value, expressed by the formula:
[0089]
[0090] in, Represents the intensity of micro-vibrations in the time domain. Represents the frequency domain energy of micro-vibrations. Indicates spatial distribution weights, and These are empirical coefficients used to balance the contributions of different features. This represents the number of sensing units.
[0091] As a specific embodiment, in step S103, determining the occupant's coma status and coma level based on respiratory rate and micro-vibration energy value includes:
[0092] Millimeter-wave radar is used as the primary data source, and flexible pressure sensing is used as the secondary data source.
[0093] The basic confidence level of the first data source is calculated based on the respiratory rate, and the basic confidence level of the second data source is calculated based on the micro-vibration energy.
[0094] Using Dempster's composition rules, the two data sources are merged based on the basic confidence assignments of the first and second data sources to obtain the merged basic confidence assignments;
[0095] The fusion belief degree is determined based on the basic credibility allocation after fusion. The state with the highest fusion belief degree and greater than the decision threshold is used as the judgment criterion to determine whether the occupant is in a coma.
[0096] The above judgment process will be explained in detail below with a specific example.
[0097] First, define the set of all possible, mutually exclusive crew state hypotheses (identification frame):
[0098] ;
[0099] After the central controller preprocesses the acquired data, it establishes the relationship between the respiratory rate f and the confidence level based on the data from the primary data source (millimeter-wave radar) using a nonlinear function (such as a sigmoid function or a Gaussian function). The mapping, where the closer f is to the threshold center of a certain coma level, the higher the basic confidence level assigned to that level. The higher the value, the better. For example:
[0100] Assuming f ≥ 12 breaths / minute, the range is considered conscious; 8 ≤ f < 12, the range is considered mild coma; 4 ≤ f < 8, the range is considered moderate coma; and f < 4, the range is considered deep coma. Based on the nonlinear function used for mapping, if the actual respiratory rate is detected... =5 times / minute, located near the center of the interval [4, 8), then assign a basic confidence level of 0.7 to m1({severe coma}); and assign the uncertainty to the remaining adjacent hypotheses, such as m1({mild coma})=0.1, m1({uncertain})=0.2.
[0101] Similarly, for the second data source (flexible pressure sensor array), the micro-vibration energy is divided into intervals according to a preset threshold, and a mapping function is used to determine the confidence assignment value m2.
[0102] Next, Dempster's composition rule is used to fuse m1 and m2, generating a new fused basic confidence assignment m, expressed by the formula:
[0103]
[0104] in, All are recognition frameworks a subset of and The basic confidence levels for hypotheses B and C are assigned to the first and second data sources, respectively. The conflict coefficient measures the degree of contradiction between two data sources. The calculation formula is:
[0105]
[0106] Belief Based on the basic credibility allocation after fusion The result, expressed by the formula, is:
[0107]
[0108] Where X represents the frame The single hypothesis in the framework is that A is the identification framework. The non-empty subset of.
[0109] As a specific implementation, if the millimeter-wave radar determines two states, namely moderate coma C and unknown state U, and according to the mapping function, =0.6, =0.4; the flexible pressure sensor array determined three states: moderate coma (C), mild coma (B), and unknown state (U). =0.7, =0.2, =0.1; then:
[0110] First, calculate the conflict coefficient. Conflict occurs when the intersection of the hypothetical sets supported by the two data sources is an empty set, i.e.: This could be a distinction between moderate coma (from m1) and mild coma (from m2), in which case the conflict coefficient... For product =0.6 0.2 = 0.12;
[0111] Next, it was determined that it could produce All possible combinations, and calculate the fusion confidence m(A) under the corresponding state A;
[0112] For example: If A is in a moderate coma, then... The combinations are: {C}∩{C}, {C}∩{U}, {U}∩{C}; the sum of the products. = (0.6*0.7) + (0.6*0.1) + (0.4*0.7) = 0.76, therefore, the fused m(A) = 0.864 can be calculated.
[0113] Similarly, the product of A in a mild coma state can be calculated to be 0.08, and the product of A in an uncertain state can be calculated to be 0.04, with corresponding m(A) of 0.091 and 0.045 respectively.
[0114] Finally, based on m(A), the belief level Bel is obtained. It can be seen that Bel({C}) = 0.864 is much higher than the likelihood upper limit of all other hypotheses and exceeds the preset decision threshold (such as 0.7). Therefore, the central controller can draw a high-confidence conclusion: the occupant is in a moderate coma.
[0115] By applying the Dempster synthesis rule fusion method, the uncertainties and conflicts arising from multiple data sources can be resolved. This quantifies and separates the degree of inconsistency between sensors, avoiding the misleading influence of simple averaging or majority voting mechanisms, ultimately yielding highly reliable and robust decision indicators. This also ensures that the central controller can arrive at a single, high-confidence conclusion even when faced with discrepancies in data support.
[0116] In some embodiments, if the hardware configuration of the central processing unit is low, a simple and fast threshold judgment method can be used, specifically including: when f < 8 times / minute and E < 0.01g² / Hz, the occupant is directly judged to be in a coma.
[0117] Furthermore, the coma state includes mild coma, moderate coma, and deep coma; the criteria for determining the coma state include:
[0118] When 8 ≤ f < 12 beats / minute and 0.01 ≤ E < 0.05 g² / Hz, it is considered a mild coma.
[0119] When 4 ≤ f < 8 times / minute and 0.005 ≤ E < 0.01 g² / Hz are met, the condition is considered moderate coma.
[0120] When f < 4 times / minute and E < 0.005 g² / Hz, it is determined to be a deep coma;
[0121] Where f represents the breathing frequency and E represents the micro-vibration energy value.
[0122] To further improve the accuracy of state assessment, in some embodiments, the method for determining coma also includes:
[0123] Before the accident, the occupants' normal respiratory rate f0 and normal micro-vibration energy value E0 were continuously monitored and stored as individual baselines; after the accident, the respiratory rate f0 was extracted in real time. r and micro-vibration energy value E r Coma is determined based on dynamic thresholds, specifically:
[0124] When f r Compared to the percentage decrease of f0 exceeding a first preset threshold (e.g., 30%) and E r When the percentage decrease of E0 exceeds a first preset threshold (e.g., 35%), the occupant is determined to be in a coma.
[0125] By comparing the degree of decline in real-time physiological indicators after an accident relative to their normal state, the system can avoid misjudging healthy individuals with lower baseline physiological parameters and ensure that injured individuals with higher baseline parameters are not missed. This relative decline assessment method ensures a more accurate and reliable determination of coma status, greatly improving the system's robustness.
[0126] As a preferred embodiment, the graded wake-up strategy includes:
[0127] When a patient is determined to be in a mild coma, activate the pneumatic vibration generator to inflate each airbag for 3 seconds.
[0128] Moderate coma: Activate the pneumatic vibration generator (inflate each airbag for 6 seconds);
[0129] Deep coma: Activate the pneumatic vibration generator (inflate each airbag for 9 seconds), and simultaneously perform the following operations alternately through the temperature stimulation module: cold water spray at 8-15℃ for 2 seconds, followed by hot air blowing at 40-45℃ for 3 seconds.
[0130] Furthermore, the method also includes an environmental collaborative optimization step:
[0131] When the outside temperature is <5℃, reduce the cold water spray temperature to 15-20℃; when the inside humidity is >80%, disable the cold water spray function and only use the hot air and vibration combination mode.
[0132] In some embodiments, to ensure the effectiveness, safety, and comfort of the wake-up stimulus and overcome the limitations of traditional in-vehicle systems under individual differences and complex sitting postures, this application also designs an adaptive hierarchical wake-up strategy. Specifically:
[0133] Using pressure distribution data collected by a flexible pressure sensor array, the accurate location of the occupant's pressure center point, the effective bearing area on the seat contact surface, and the total pressure integral value of the seat contact area are calculated.
[0134] Based on the calculated pressure center point and effective bearing area, the central controller dynamically determines and selects the optimal grouping of stimulation airbags, namely:
[0135] Exclude airbags that do not make effective contact with the occupant's torso or only make contact with the limbs;
[0136] Only activate the airbags located below the occupant's back, waist, or hips to ensure that mechanical stimulation is applied to the core torso.
[0137] Finally, using the total pressure integral value, the equivalent weight of the occupant is estimated through a preset model; and the inflation pressure and / or inflation duration of the pneumatic vibration generator in the wake-up strategy are adjusted compensatorily based on the equivalent weight.
[0138] When the equivalent body weight is higher than the average, increase the inflation pressure or extend the inflation time appropriately;
[0139] When the equivalent body weight is below the average, reduce the inflation pressure or shorten the inflation time appropriately.
[0140] Furthermore, in some embodiments, the method further includes: adjusting the activation sequence or cycle of the airbag in real time according to the positional deviation of the pressure center point, so that the somatosensory vibration center of the mechanical stimulation can always be concentrated on the occupant's torso area.
[0141] By calculating the pressure center point and effective load-bearing area, the system can dynamically and accurately locate the occupant's core torso, eliminating stimulation of irrelevant limbs or vulnerable areas and ensuring the effectiveness and safety of mechanical stimulation. Simultaneously, by estimating equivalent body weight using the total pressure integral value and compensating for adjustments to the airbag inflation parameters, the system ensures fairness in arousal intensity, avoiding understimulation or overstimulation of occupants of different body types. Real-time correction of the airbag activation sequence ensures that the somatosensory vibration center remains concentrated in the torso, improving the system's robustness in non-standard occupant seating positions and contributing to increased survival rates after vehicle accidents.
[0142] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.
Claims
1. An adaptive coma-awakening vehicle system after a vehicle accident, characterized in that, It includes a central controller, as well as a millimeter-wave radar module, a flexible pressure sensing array, an environmental sensor group, and a multi-mode execution unit connected to the central controller; The millimeter-wave radar module is installed inside the vehicle seat back to collect the occupant's chest displacement and obtain respiratory signals in a non-contact manner. A flexible pressure sensor array is installed on the underside of the seat cushion and backrest cover to collect micro-vibration signals of the occupant's body surface. An environmental sensor array, including interior and exterior temperature and humidity sensors and a noise detector, is used to collect environmental parameters; The central controller is used to extract respiratory rate from respiratory signals and micro-vibration energy value from micro-vibration signals on the body surface. Based on respiratory rate and micro-vibration energy value, it determines the degree of coma of the occupant and controls the multi-mode execution unit to execute the wake-up strategy according to the temperature and humidity inside and outside the vehicle, noise information and the degree of coma. The multi-mode execution unit includes at least a pneumatic vibration generator and a temperature stimulation module, used to implement graded awakening of the occupants according to the instructions of the central controller; the multi-mode execution unit also includes a light stimulation unit and an acoustic stimulation unit. The pneumatic vibration generator, temperature stimulation module, light stimulation module and sound stimulation module can be started individually or in combination. According to the instructions of the central controller, the stimulation position, action area and parameters are adjusted according to the occupant's body shape adaptive adjustment mechanism to adapt to the different body shapes, weights and sitting postures of the occupants. The pneumatic vibration generator includes an airbag, an air pump, a control valve assembly, and connecting pipes. The airbag is fitted between the seat upholstery and the foam, and directly contacts the occupant's back and buttocks. It is used to generate pressure and vibration through deformation, providing mechanical stimulation to the occupant. The air pump provides an air source, compresses the air, and delivers it into the airbag. The control valve assembly precisely controls the airflow direction and speed, determining which airbags need to be inflated or deflated, the inflation / deflation sequence, and the timing based on instructions from the central controller. The connecting pipes connect the air pump, valve assembly, and each airbag, forming an air transmission path. The airbag is also compatible with car seat massage functions.
2. The adaptive coma-awakening vehicle system after a vehicle accident according to claim 1, characterized in that, The sampling frequency of the millimeter-wave radar module is no less than 20Hz.
3. The adaptive coma-awakening vehicle system after a vehicle accident according to claim 1, characterized in that, The flexible pressure sensing array contains at least 32 sensing units, with a spatial resolution of no more than 2 cm and a sampling frequency of no less than 100 Hz.
4. A method for controlling unconsciousness and waking up after a vehicle accident, applied to the system described in any one of claims 1-3, characterized in that, include: After a vehicle accident occurs, the central controller activates the millimeter-wave radar module and pressure sensor array to collect the occupants' breathing signals and micro-vibration signals on their body surface. The central controller extracts respiratory frequency from respiratory signals and micro-vibration energy values from body surface micro-vibration signals based on a preset feature extraction method. The occupant's coma status and coma level are determined based on respiratory rate and micro-vibration energy values, including: Millimeter-wave radar is used as the primary data source, and a flexible pressure sensing array is used as the secondary data source. The basic confidence level of the first data source is calculated based on the respiratory rate, and the basic confidence level of the second data source is calculated based on the micro-vibration energy. Using Dempster's composition rules, the two data sources are merged based on the basic confidence assignments of the first and second data sources to obtain the merged basic confidence assignments; The fusion belief degree is determined based on the basic credibility allocation after fusion. The state with the highest fusion belief degree and greater than the decision threshold is used as the judgment basis to determine the coma state of the occupants. Based on the determined level of coma and environmental data collected by environmental sensors, the multi-mode execution unit is controlled to implement the corresponding graded wake-up strategy.
5. The method according to claim 4, characterized in that, The preset feature extraction method includes: Respiratory rate is obtained by extracting respiratory cycle information through time-domain peak detection or short-time Fourier transform; High-frequency component filtering and statistical analysis were performed on the micro-vibration signal on the body surface to extract the intensity, frequency and spatial distribution characteristics of the micro-vibration and obtain the micro-vibration energy value.
6. The method according to claim 4, characterized in that, The method of determining the occupant's coma level based on respiratory rate and micro-vibration energy value also includes: When 8 ≤ f < 12 beats / minute and 0.01 ≤ E < 0.05 g² / Hz, it is considered a mild coma. When 4 ≤ f < 8 times / minute and 0.005 ≤ E < 0.01 g² / Hz are met, the condition is considered moderate coma. When f < 4 times / minute and E < 0.005 g² / Hz, it is determined to be a deep coma; Where f represents the breathing frequency and E represents the micro-vibration energy value.
7. The method according to claim 4, characterized in that, The graded wake-up strategy includes: When a patient is determined to be in a mild coma, the pneumatic vibration generator is activated, and the inflation time is set to the first duration. When the patient is determined to be in a moderate coma, activate the pneumatic vibration generator and set the inflation time to the second duration. When a patient is determined to be in a deep coma, the pneumatic vibration generator is activated and the inflation time is set to the third duration. At the same time, the temperature stimulation module alternately sprays cold water and blows hot air. The temperature of the cold water is set according to the outside temperature of the vehicle, and the temperature of the hot air is set according to the humidity inside the vehicle.
8. The method according to claim 7, characterized in that, The method for setting the temperature of the cold water includes: When the humidity inside the vehicle is ≤80%, determine whether the outside temperature is below 5℃; when the outside temperature is <5℃, set the temperature of the cold water to 15-20℃; when the outside temperature is ≥5℃, set the temperature of the cold water to 8-15℃. When the humidity inside the vehicle exceeds 80%, the cold water spray function will be disabled, and only the hot air and vibration combination mode will be enabled.
Citation Information
Patent Citations
Safe driving assistance method, vehicle-mounted terminal, vehicle and readable storage medium
CN117818660A
Passenger compartment gas rescue system, method, equipment and medium
CN118952963A
