A method and device for active intervention in fatigue driving of an electric vehicle

Abstract

This invention relates to the field of electric vehicle driving technology, specifically to a method and device for active intervention to prevent fatigue driving in electric vehicles. The method uses a flexible dry electrode array pre-installed on the left and right handlebars of the electric vehicle to collect surface electromyography (EMG) signals from the driver's forearm in real time. At preset intervals, multidimensional feature values ​​are extracted from the EMG signals to determine the driver's state. When mild fatigue is detected, linear resonant actuators inside the left and right handlebars are activated to emit pulse vibrations. When severe fatigue is detected, the maximum speed limit is smoothly reduced to a preset limit within a preset time period, while the dead zone damping of the throttle is increased. This technical solution utilizes microscopic changes in EMG signals as a fatigue criterion, achieving high-precision early warning with good timeliness. The two-level intervention strategy significantly improves safety and enables non-intrusive monitoring, resulting in high user compliance.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle driving technology, specifically to a method and device for active intervention to prevent fatigue driving in electric vehicles. Background Technology

[0002] With the rapid development of food delivery and instant logistics industries, as well as the continued growth in urban long-distance commuting demand, two-wheeled electric vehicles have become an important means of transportation for short-distance urban travel due to their flexibility, convenience, and low operating costs. However, prolonged continuous riding can easily lead to driver fatigue, such as muscle fatigue, inattention, and slow reaction time, resulting in frequent traffic accidents and becoming a major cause of threats to urban road traffic safety.

[0003] Current fatigue driving prevention technologies in the two-wheeled electric vehicle field mainly include: collecting driver facial features through onboard cameras to judge fatigue status by behaviors such as closing eyes, nodding, and yawning; collecting vehicle driving data by relying on GPS and IMU inertial measurement units to judge fatigue by abnormal trajectories such as S-shaped driving and vehicle swaying; and using high-decibel buzzers and voice broadcasts to provide fatigue reminders.

[0004] However, monitoring schemes that collect driver facial features are highly dependent on ambient lighting conditions and cannot function properly at night or in inclement weather. Monitoring schemes that collect vehicle driving data require at least a 30-second delay to identify fatigue levels and are easily affected by external factors such as road bumps, strong crosswinds, and uneven road surfaces, frequently generating false alarms and thus having poor practicality. Furthermore, both of these monitoring schemes are reactive, only triggering alarms after the driver exhibits obvious signs of fatigue; by then, vehicle control has already deviated, and accidents are often unavoidable. Fatigue warnings using high-decibel buzzers and voice announcements are easily masked by sound and light signals in noisy traffic environments, rendering the warnings ineffective. Sudden high-decibel sounds can also startle drivers in a fatigued or neurasthenic state, triggering instinctive braking, loss of vehicle control, and other secondary accidents.

[0005] In summary, existing electric vehicle fatigue driving prevention technologies suffer from serious deficiencies in warning timeliness, failing to predict driver fatigue in advance; poor safety of intervention methods, typically only providing alarm functions or employing aggressive vehicle control methods such as direct power cut-off, increasing safety hazards; and poor user compliance, with most existing solutions requiring users to wear additional devices such as wristbands and helmet sensors, making large-scale promotion difficult in actual operation. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to provide an active intervention method and device for preventing fatigue driving in electric vehicles, so as to solve the problems of insufficient warning timeliness, poor safety of intervention methods, and poor user compliance in the existing electric vehicle fatigue driving prevention technology solutions.

[0007] According to a first aspect of the present invention, a method for active intervention to prevent fatigue driving in electric vehicles is provided, comprising:

[0008] The surface electromyography (EMG) signals of the driver's forearm are collected in real time by a flexible dry electrode array pre-set on the left and right handlebars of the electric vehicle, and the surface EMG signals are pre-processed. Every preset period, multidimensional feature values ​​are extracted from the preprocessed surface electromyography signal. The multidimensional feature values ​​include root mean square value, median frequency and sample entropy. The driver's condition is determined based on the root mean square value, median frequency, and sample entropy, including normal, mild fatigue, and severe fatigue. When the driver is detected to be slightly fatigued, the linear resonant actuators inside the left and right handlebars of the electric vehicle emit pulse vibrations. When the driver is detected to be severely fatigued, the maximum speed limit will be smoothly reduced to the preset speed limit within a preset time period, while the dead zone damping of the throttle will be increased.

[0009] Preferably, the surface electromyography signal is preprocessed, including: An adaptive notch filter is used to remove 50Hz or 60Hz power frequency interference from the surface electromyography signal; Using a Butterworth bandpass filter of 20Hz-450Hz, the effective frequency band of electromyography in the surface electromyography signal is preserved, while low-frequency motion artifacts and high-frequency noise are filtered out. The surface electromyography signal is rectified by full wave and smoothed using a moving average window.

[0010] Preferably, the root mean square value is extracted from the preprocessed surface electromyography signal, using the following formula:

[0011] in, The root mean square value, Let i be the total number of sampling points within the moving average window, and i be the sampling point index. denoted as the amplitude of the surface electromyography signal at the i-th sampling point within the moving average window.

[0012] Preferably, extracting the median frequency from the preprocessed surface electromyography signal includes: Frequency domain analysis was performed on the surface electromyography signal within the moving average window to obtain the power spectral density curve; The total power is obtained by numerically integrating the power spectral density curve within the effective frequency band. The power spectral density curve is gradually accumulated from the low-frequency end to the high-frequency end of the effective frequency band. When the accumulated power reaches half of the total power, the corresponding frequency point is the median frequency.

[0013] Preferably, the method further includes: The collected surface electromyography (EMG) signals were from the driver's forearm flexor and extensor muscle groups.

[0014] Preferably, the linear resonant actuators inside the left and right handlebars of the electric vehicle emit pulse vibrations, including: The linear resonant actuators inside the left and right handlebars of the electric vehicle emit specific rhythmic pulse vibrations of 30Hz-50Hz, so that the pulse vibrations are directly transmitted to the cochlea of ​​the inner ear through the bones of the driver's hand.

[0015] According to a second aspect of the present invention, an active intervention device for preventing fatigue driving in electric vehicles is provided, comprising: A flexible dry electrode array, a controller, and a linear resonant actuator; both the flexible dry electrode array and the linear resonant actuator are mounted on the left and right handlebars of the electric vehicle. The controller performs any of the methods described above.

[0016] Preferably, the flexible dry electrode array is positioned at the location where the inner side of the left and right handlebars of the electric vehicle contacts the web of the hand and the flexor muscles of the forearm.

[0017] Preferably, the electrode material of the flexible dry electrode array is conductive silicone or nano-silver coated fabric.

[0018] Preferably, the electrode material of the flexible dry electrode array contacts the driver's skin through a microporous rubber layer, or the electrode material of the flexible dry electrode array directly contacts the driver's skin as the grip surface.

[0019] The technical solution provided by this invention may include the following beneficial effects: Understandably, the technical solution presented in this invention can collect surface electromyography (EMG) signals of the driver's forearm in real time through a flexible dry electrode array pre-installed on the left and right handlebars of an electric vehicle. At preset intervals, multi-dimensional feature values ​​are extracted from the EMG signals to determine the driver's state. When mild fatigue is detected, the linear resonant actuators inside the left and right handlebars of the electric vehicle emit pulse vibrations. When severe fatigue is detected, the maximum speed limit is smoothly reduced to a preset limit within a preset time period, while simultaneously increasing the dead zone damping of the throttle. This technical solution utilizes microscopic changes in EMG signals as a fatigue criterion, achieving high-precision early warning with good timeliness; the two-level intervention strategy significantly improves safety; and it enables non-intrusive monitoring, resulting in high user compliance.

[0020] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0022] Figure 1 This is a schematic diagram illustrating the steps of an active intervention method for preventing fatigue driving in an electric vehicle, according to an exemplary embodiment. Figure 2 This is a schematic diagram of a workflow illustrated according to an exemplary embodiment; Figure 3 It is a trend curve of characteristic parameters changing over time according to an exemplary embodiment. Detailed Implementation

[0023] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0024] In one embodiment, Figure 1 This is a schematic diagram illustrating the steps of an active intervention method for preventing driver fatigue in an electric vehicle, according to an exemplary embodiment. See also... Figure 1 and Figure 2 A method for active intervention to prevent fatigue driving in electric vehicles is provided, including: Step S11: Real-time acquisition of surface electromyography (EMG) signals of the driver's forearm through a flexible dry electrode array pre-set on the left and right handlebars of the electric vehicle, and signal preprocessing of the surface EMG signals.

[0025] The acquisition of surface electromyography (EMG) signals forms the sensing layer of the entire technical solution. Through a flexible dry electrode array in close contact with the rider's skin, this non-invasive acquisition structure allows for the seamless, real-time acquisition of EMG signals from the forearm flexor and extensor muscle groups while the rider naturally grips the handlebars. The sampling frequency is set at 500Hz-1000Hz, with a resolution better than 12 bits, ensuring accurate capture of microvolt-level muscle discharge signals.

[0026] In a preferred embodiment, a flexible dry electrode array is pre-embedded into the inner side of the left and right handlebars of the electric vehicle at the positions corresponding to the web of the hand and the flexor muscles of the forearm through in-mold injection molding or bonding process. The electrodes are made of conductive silicone or nano-silver coated fabric material, which can be in close contact with the driver's skin through the microporous rubber layer or directly as the surface of the handlebar.

[0027] The collected surface electromyography (EMG) signals were from the driver's forearm flexor and extensor muscle groups.

[0028] After the surface electromyography (EMG) signal is acquired, it enters the processing layer of the entire technical solution. The processing layer can be built into the vehicle controller (ECU) or a separate low-power MCU to perform detailed data preprocessing, multi-dimensional feature extraction, and driver status determination.

[0029] In a preferred embodiment, the surface electromyography (EMG) signal undergoes signal preprocessing, including: removing 50Hz or 60Hz power frequency interference from the EMG signal using an adaptive notch filter; retaining the effective frequency band of the EMG signal using a Butterworth bandpass filter (20Hz-450Hz) to filter out low-frequency motion artifacts (caused by vehicle vibration) and high-frequency noise (caused by environmental electromagnetic interference); and performing full-wave rectification on the EMG signal and smoothing it using a moving average window (e.g., 200ms). Through this preprocessing, various interference factors are eliminated, resulting in a clean and stable EMG signal, providing reliable data support for subsequent multidimensional fatigue feature extraction.

[0030] Step S12: Every preset period, extract multidimensional feature values ​​from the preprocessed surface electromyography signal. The multidimensional feature values ​​include root mean square value, median frequency, and sample entropy.

[0031] In practice, after preprocessing the surface electromyography (SEMG) signal, the system uses a fixed preset period of 200ms to synchronously extract three types of multidimensional feature values ​​from the preprocessed SEMG signal through the fatigue feature analysis and decision-making unit built into the vehicle controller (ECU) or independent low-power MCU. This provides a precise quantitative basis for subsequent driver fatigue state classification. These multidimensional feature values ​​specifically include three types of parameters: root mean square (RMS), median frequency (MDF), and sample entropy (SampEn). The RMS, as a time-domain feature, reflects the muscle contraction strength; the MDF, as a frequency-domain feature, reflects the leftward shift of the signal spectrum caused by muscle fatigue; and the sample entropy, as a nonlinear feature, characterizes the complexity of the muscle control signal. These three features comprehensively depict the physiological state of the driver's forearm muscles from different dimensions, ensuring the comprehensiveness and accuracy of fatigue assessment.

[0032] In a preferred embodiment, the root mean square value is extracted from the preprocessed surface electromyography signal, using the following formula:

[0033] in, The root mean square value, Let i be the total number of sampling points within the moving average window, and i be the sampling point index. denoted as the amplitude of the surface electromyography signal at the i-th sampling point within the moving average window.

[0034] This technical solution first sums the squares of the signal amplitudes of all sampling points within the window, then calculates the average value and takes the square root to obtain the root mean square value of the electromyography signal for that time period. The system completes real-time calculation and update of the root mean square value every 200ms. By monitoring the dynamic fluctuation of this feature, it distinguishes between regular amplitude changes under normal driving operation and abnormal high-frequency tremors that cannot be voluntarily suppressed during severe fatigue, providing accurate temporal feature support for subsequent fatigue grading.

[0035] In a preferred embodiment, extracting the median frequency from the preprocessed surface electromyography (SEMG) signal includes: performing frequency domain analysis on the SEMG signal within a moving average window to obtain a power spectral density curve; performing numerical integration on the power spectral density curve within the effective frequency band to obtain the total power; and gradually accumulating the power of the power spectral density curve from the low-frequency end to the high-frequency end of the effective frequency band. When the accumulated power reaches half of the total power, the corresponding frequency point is the median frequency.

[0036] The median frequency (MDF) is used as a frequency domain feature for determining the driver's muscle fatigue state. Its extraction process is based on standardized frequency domain calculations performed on preprocessed surface electromyography (EMG) signals and a 200ms moving average window. First, frequency domain analysis is performed on the preprocessed EMG signals within the moving average window. Power spectral density (PSD) analysis is used to obtain the power spectral density curve, which characterizes the power distribution of the EMG signal at each frequency component. Then, within the effective EMG frequency band of 20Hz-450Hz, numerical integration is performed on the PSD curve to obtain the total power of the EMG signal in the current time period. Finally, the power values ​​corresponding to the PSD curve are gradually accumulated from the low-frequency end to the high-frequency end of the effective frequency band. When the accumulated power reaches half of the total power, the frequency point corresponding to that position is the median frequency of the current EMG signal. This feature can accurately reflect the leftward shift of the spectrum caused by muscle fatigue, providing a key frequency domain basis for the system's fatigue grading determination.

[0037] The extraction of sample entropy is based on the preprocessed surface electromyography time-series signal. The system synchronously completes the extraction and updating of sample entropy with a calculation cycle of 200ms. The sample entropy value is calculated using a standard algorithm known in the field. This feature can accurately reflect the degree of orderliness of the driver's neuromuscular control signal. When the muscles are fatigued, the control mode tends to be simple and rigid, and the sample entropy value decreases accordingly, providing key nonlinear feature support for the system's fatigue grading judgment.

[0038] Step S13: Determine the driver's state based on the root mean square value, median frequency, and sample entropy, as follows: Normal state determination: If the median frequency is within the fluctuation range of the first person's baseline value (i.e., the median frequency is stable), and the sample entropy is within the fluctuation range of the second person's baseline value (the sample entropy value is higher), and the root mean square value shows regular fluctuations in normal driving operation, then the driver's state is normal.

[0039] When the system detects that the median frequency (MDF) is within the normal fluctuation range of the first human baseline value established for the driver, without significant drift or decrease, and the sample entropy (SampEn) is simultaneously within the corresponding second human baseline fluctuation range, maintaining the high complexity of muscle control signals and the flexibility of neural regulation, and at the same time, the root mean square value of electromyography (RMS) shows regular and normal fluctuations with routine driving operations such as steering, gripping, acceleration, and deceleration, without abnormal amplitude abrupt changes or high-frequency tremors, that is, when the above three characteristic conditions are met simultaneously, the system determines that the driver is currently in a normal driving state. In this state, the driver's forearm flexor and extensor muscle groups are in normal physiological state, with no signs of muscle fatigue, the brain is clear and focused, the vehicle is stable and there is no safety risk, the system only continuously collects electromyography signals and dynamically maintains personal baseline parameters, without triggering any warning or intervention actions, and the vehicle maintains normal power output and handling characteristics.

[0040] Mild fatigue state determination: If the decrease in median frequency relative to the driver's personal baseline value exceeds a first preset proportion within a preset monitoring time window, and the sample entropy is greater than a preset threshold, and the root mean square value shows regular fluctuations in normal driving operation, then the driver's state is mild fatigue; the preset threshold is less than the second personal baseline value.

[0041] In practice, when the system uses a 1-minute preset monitoring time window for real-time judgment, if it detects that the driver's median frequency (MDF) has decreased by more than 15% relative to their normal driving personal baseline (first preset ratio), and the sample entropy (SampEn) value is greater than the system's preset judgment threshold (this preset threshold is less than the second personal baseline value representing a normal state, indicating that the sample entropy has only slightly decreased), and the root mean square value of electromyography (RMS) still fluctuates regularly with normal driving operations such as steering, gripping, acceleration, and deceleration, without any abnormal high-frequency tremors, and all of the above joint judgment conditions are met, the system will accurately determine that the driver is currently in a state of mild fatigue (warning period). In this state, the driver only experiences early physiological muscle fatigue, and there are slight changes in muscle energy metabolism and muscle fiber recruitment patterns, but the neural control function is not damaged, and the driver can still control the vehicle autonomously and stably. The system only triggers a first-level bone conduction vibration warning to achieve a private and gentle alert awakening, without implementing power restriction intervention.

[0042] Severe fatigue state determination: If the median frequency decreases more than the driver's personal baseline value within the preset monitoring time window, and the sample entropy is less than or equal to the preset threshold, and the root mean square value is abnormal, then the driver's state is severe fatigue.

[0043] In practice, if the median frequency (MDF) decreases by more than 30% relative to the driver's personal baseline value for normal driving (second preset ratio), the electromyographic signal spectrum shows a significant left shift, and the sample entropy (SampEn) value is less than or equal to the system's preset fatigue judgment threshold, it indicates that the complexity of muscle control signals is extremely low, the neuromuscular regulation mode is highly rigid, and the root mean square value (RMS) of electromyography shows abnormal high-frequency tremors, corresponding to the driver's forearm muscles producing continuous tremors that cannot be voluntarily suppressed. If the above three characteristic conditions are met simultaneously, the system will determine that the driver is currently in a state of severe fatigue (intervention period). In this state, the driver has entered a state of deep physiological fatigue, the neuromuscular transmission efficiency has decreased significantly, the brain reaction is slow, the attention is severely scattered, and there is an extremely high safety risk to vehicle control. The system will immediately activate the secondary power smoothing reconfiguration intervention mechanism to ensure driving safety.

[0044] See Figure 3 The graph, with continuous driving time (minutes) on the horizontal axis and normalized feature parameter values ​​on the vertical axis, visually presents the dynamic changes of the three major features—median frequency (MDF), sample entropy (SampEn), and electromyographic amplitude (RMS)—as fatigue intensifies. This confirms the rationality of the three-level fatigue grading model in this embodiment: In the initial stage of driving, the driver is in a Level 0 normal state, with stable MDF, high SampEn, and RMS fluctuating regularly with normal driving operations; as driving time increases, the driver enters the Level 1 mild fatigue stage, with MDF decreasing by more than 15% relative to the individual baseline value, SampEn decreasing slightly, and RMS remaining normal; when fatigue continues to accumulate to a severe state, reaching the Level 2 intervention trigger condition, MDF decreases by more than 30%, SampEn drops below the judgment threshold, and RMS shows obvious abnormal high-frequency tremors. The graph clearly shows the correspondence between the three features and the fatigue level, visually demonstrating the advantage of electromyographic signals in capturing the golden time window of physiological precursors of fatigue, and verifying the scientificity and accuracy of multi-feature joint judgment.

[0045] Step S14: When the driver is detected to be slightly fatigued, the linear resonant actuators inside the left and right handlebars of the electric vehicle emit pulse vibrations.

[0046] This step is a primary intervention for drivers experiencing mild fatigue. When the system confirms that the driver is in a state of mild fatigue through a joint determination of three electromyographic characteristics—median frequency, sample entropy, and root mean square value—the vehicle controller (ECU) or the built-in low-power MCU will immediately output a control command to drive the linear resonant actuators (LRAs) integrated inside the left and right handlebars of the electric vehicle to start operating, causing them to emit specific rhythmic pulse vibrations of 30Hz-50Hz. This vibration, through the bone conduction effect, is directly transmitted to the cochlea of ​​the inner ear through the bones of the driver's hand holding the handlebars, forming a private tactile warning that only the rider can perceive (such as a "buzzing" sound). There is no airborne noise, and it will not startle the driver in a fatigued state. It quickly enhances the driver's alertness with gentle tactile stimulation, while not interfering with the normal operation of the vehicle, thus achieving an early safety warning of fatigue.

[0047] Step S15: When the driver is detected to be severely fatigued, the maximum speed limit is smoothly reduced to the preset speed limit within a preset time period, while the dead zone damping of the throttle is increased.

[0048] This step is a secondary safety intervention for drivers in a state of severe fatigue. When the driver is confirmed to be in a state of severe fatigue through a combination of electromyography (EMG) characteristics, the vehicle control unit (ECU) immediately activates the power smoothing reconfiguration mechanism. Over a preset time period of 1-2 minutes, it uses a linear and gradual control method to smoothly and without jerking the vehicle's maximum speed limit down to the preset speed limit of 20 km / h, avoiding the risk of sudden power loss and vehicle imbalance caused by traditional direct power cut-off. At the same time, the ECU linearly increases the dead zone damping of the throttle, making the throttle feel "heavier," and conveying a fatigue warning to the driver through physical touch. The entire intervention process is executed silently without triggering any audible or visual alarms, so as not to startle the driver in a fatigued state. While preserving the driver's control of the vehicle, it guides the driver to safely pull over to the side of the road through gentle power constraints and control adjustments, minimizing the risk of traffic accidents caused by severe fatigue driving.

[0049] This invention possesses multiple outstanding technical advantages, comprehensively addressing the core pain points of existing technologies: Firstly, it achieves a golden time window for advanced warning. Changes in electromyographic physiological signals, compared to visible fatigue-related movement distortions such as driver nodding or S-shaped vehicle trajectories, can be accurately identified by the system 30-60 seconds in advance, providing valuable emergency response time for traffic accident prevention. Secondly, the warning method is private and non-alarming; bone conduction vibrations can only be perceived by the rider, avoiding the embarrassment of noise in public places and preventing panic-induced operational errors caused by sudden stimuli in a fatigued state through gentle tactile feedback. The solution addresses several issues. It constructs a complete active safety loop, fundamentally changing the passive protection approach of traditional solutions that only issue alarms but don't control the vehicle. By smoothly limiting power, it eliminates high-risk driving conditions from a physical perspective. Furthermore, the system is adaptable to all weather conditions, unaffected by darkness, tunnels, or rain / snow that can blur the camera. It operates stably as long as the rider holds the handlebars, demonstrating extremely high overall reliability. Finally, the solution boasts high user compliance, employing an integrated, non-contact monitoring design on the handlebars. Riders do not need to wear any additional equipment, effectively solving the problems of cumbersome charging and uncomfortable wearing of wearable devices.

[0050] In another embodiment, an active intervention device for preventing driver fatigue in electric vehicles is provided, comprising: A flexible dry electrode array, a controller, and a linear resonant actuator; both the flexible dry electrode array and the linear resonant actuator are mounted on the left and right handlebars of an electric vehicle; the controller performs the method described in any of the above descriptions.

[0051] In a preferred embodiment, the flexible dry electrode array is positioned at the location where the inner side of the left and right handlebars of the electric vehicle contacts the web of the hand and the flexor muscles of the forearm.

[0052] In a preferred embodiment, the electrode material of the flexible dry electrode array is conductive silicone or nano-silver coated fabric.

[0053] In a preferred embodiment, the electrode material of the flexible dry electrode array contacts the driver's skin through a microporous rubber layer, or the electrode material of the flexible dry electrode array directly contacts the driver's skin as the grip surface.

[0054] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0055] It should be noted that in the description of this invention, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this invention, unless otherwise stated, "a plurality of" means at least two.

[0056] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0057] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0058] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0059] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0060] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0061] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0062] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for active intervention to prevent fatigue driving in electric vehicles, characterized in that, include: The surface electromyography (EMG) signals of the driver's forearm are collected in real time by a flexible dry electrode array pre-set on the left and right handlebars of the electric vehicle, and the surface EMG signals are pre-processed. Every preset period, multidimensional feature values ​​are extracted from the preprocessed surface electromyography signal. The multidimensional feature values ​​include root mean square value, median frequency and sample entropy. The driver's condition is determined based on the root mean square value, median frequency, and sample entropy, including normal, mild fatigue, and severe fatigue. When the driver is detected to be slightly fatigued, the linear resonant actuators inside the left and right handlebars of the electric vehicle emit pulse vibrations. When the driver is detected to be severely fatigued, the maximum speed limit will be smoothly reduced to the preset speed limit within a preset time period, while the dead zone damping of the throttle will be increased.

2. The method according to claim 1, characterized in that, The surface electromyography (EMG) signal is preprocessed, including: An adaptive notch filter is used to remove 50Hz or 60Hz power frequency interference from the surface electromyography signal; Using a Butterworth bandpass filter of 20Hz-450Hz, the effective frequency band of electromyography in the surface electromyography signal is preserved, while low-frequency motion artifacts and high-frequency noise are filtered out. The surface electromyography signal is rectified by full wave and smoothed using a moving average window.

3. The method according to claim 2, characterized in that, The root mean square value is extracted from the preprocessed surface electromyography signal using the following formula: in, The root mean square value, Let i be the total number of sampling points within the moving average window, and i be the sampling point index. denoted as the amplitude of the surface electromyography signal at the i-th sampling point within the moving average window.

4. The method according to claim 2, characterized in that, Extracting the median frequency from the preprocessed surface electromyography signal includes: Frequency domain analysis was performed on the surface electromyography signal within the moving average window to obtain the power spectral density curve; The total power is obtained by numerically integrating the power spectral density curve within the effective frequency band. The power spectral density curve is gradually accumulated from the low-frequency end to the high-frequency end of the effective frequency band. When the accumulated power reaches half of the total power, the corresponding frequency point is the median frequency.

5. The method according to claim 1, characterized in that, Also includes: The collected surface electromyography (EMG) signals were from the driver's forearm flexor and extensor muscle groups.

6. The method according to claim 1, characterized in that, The linear resonant actuators inside the left and right handlebars of the electric vehicle emit pulse vibrations, including: The linear resonant actuators inside the left and right handlebars of the electric vehicle emit specific rhythmic pulse vibrations of 30Hz-50Hz, so that the pulse vibrations are directly transmitted to the cochlea of ​​the inner ear through the bones of the driver's hand.

7. An active intervention device for preventing fatigue driving in electric vehicles, characterized in that, include: Flexible dry electrode arrays, controllers, and linear resonant actuators; Both the flexible dry electrode array and the linear resonant actuator are mounted on the left and right handlebars of the electric vehicle. The controller performs the method as described in any one of claims 1 to 6.

8. The device according to claim 7, characterized in that, The flexible dry electrode array is positioned on the inside of the left and right handlebars of the electric vehicle, where it contacts the web of the hand and the flexor muscles of the forearm.

9. The device according to claim 7, characterized in that, The electrode material of the flexible dry electrode array is conductive silicone or nano-silver coated fabric.

10. The device according to claim 7, characterized in that, The electrode material of the flexible dry electrode array contacts the driver's skin through a microporous rubber layer, or the electrode material of the flexible dry electrode array directly contacts the driver's skin as the grip surface.

Images