Driver fatigue state evaluation method

Abstract

The invention discloses a driver fatigue state evaluation method, and relates to the technical field of data monitoring, and the method comprises the steps: obtaining a plurality of fatigue reference data used for evaluating the fatigue state of a driver, the multiple pieces of fatigue reference data comprise dynamic feature data of a vehicle, torque feature data of a steering wheel and driving duration feature data of the driver, and the dynamic feature data comprise lane departure confidence and / or trigger frequency of a lane management system; aiming at each piece of fatigue reference data, weighting the fatigue reference data based on a preset first weight to obtain weighted fatigue data, and the first weights of at least two pieces of fatigue reference data are different; a total fatigue value of the driver is obtained by combining the multiple pieces of weighted fatigue data; the total fatigue value is compared with a preset fatigue threshold value, and the fatigue evaluation grade of the driver is determined.

Description

Technical Field

[0001] This application relates to the technical field of data monitoring, and particularly to a method for evaluating a driver's fatigue state. Background Art

[0002] Currently, existing fatigue detection technologies mainly rely on analyzing a driver's facial features or physiological features and calculating the driver's fatigue state in combination with their behavior. Most of the existing technologies are mainly non-contact fatigue detection technologies, that is, inferring the driver's current fatigue state based on a monitoring camera in front of the driver for the driver's facial features, eye signals, head motility, etc. The contact fatigue detection technology requires wearing sensors that contact the body throughout the driving process and collecting physiological signals such as electrocardiogram, respiration, and skin electricity for extracting the driver's fatigue indicators. There are also some new fatigue detection technologies that use a variety of in-vehicle native data to determine the driver's current fatigue indicators.

[0003] However, in non-contact fatigue detection technologies, when the driver wears sunglasses or a mask, the accuracy of the visual recognition model will be unable to recognize or there will be errors in the recognition results. In contact fatigue detection technologies, the driver needs to wear new sensors every time they drive, and the driver may feel uncomfortable and resist wearing additional sensors. And the physiological signals come from the human body, and measurement noise may be generated due to human movements. And when the physiological signals cannot be accurately measured, the model detection accuracy is greatly affected. In addition, traditional contact / non-contact fatigue monitoring technologies all have potential privacy infringement problems in data collection. In new fatigue detection technologies, one of the judgment conditions requires inputting the number of passengers in the vehicle each time, and this condition is not directly proportional to the driver's driving state, so it should not be a judgment reason, and it additionally increases the driver's operation. In addition, voice recognition is not one of the core interactions between the driver and the vehicle, and inaccurate results are likely to be generated. There is no module in the judgment conditions that can judge the current road driving state, and there are no judgment conditions for key road information. Simple linear multiplication based on driving duration is not applicable to complex autonomous driving scenarios. Summary of the Invention

[0004] A series of simplified concepts are introduced in the Summary of the Invention section, which will be further detailed in the Detailed Description section. The Summary of the Invention section of this application does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the protection scope of the claimed technical solution.

[0005] An embodiment of this application provides a method for evaluating a driver's fatigue state, and the method includes:

[0006] Obtain a plurality of fatigue reference data for evaluating the driver's fatigue state, where the plurality of fatigue reference data includes: dynamic characteristic data of the vehicle, torque characteristic data of the steering wheel, and driving duration characteristic data of the driver, and the dynamic characteristic data includes lane offset confidence and / or the trigger frequency of the lane management system;

[0007] For each of the fatigue reference data, weight the fatigue reference data based on a preset first weight to obtain weighted fatigue data, where the first weights of at least two of the fatigue reference data are different;

[0008] Combine a plurality of the weighted fatigue data to obtain the total fatigue value of the driver;

[0009] Compare the total fatigue value with a preset fatigue threshold to determine the fatigue evaluation level of the driver.

[0010] In an embodiment of the present invention, the lane offset confidence is obtained in the following manner:

[0011] Obtain at least one offset feature sequence of the vehicle relative to the lane reference line within a first time window, where the at least one offset feature sequence includes: one or more of an offset amount sequence, an offset speed sequence, and an offset acceleration, and the offset feature sequence includes a plurality of offset feature values;

[0012] For each of the offset feature sequences, determine the lateral swing amplitude value of the feature sequence according to the numerical fluctuation of the plurality of offset feature values, and weight the lateral swing amplitude value based on a preset second weight to obtain a weighted swing amplitude value;

[0013] Determine the lane offset confidence according to at least one of the weighted swing amplitude values.

[0014] In an embodiment of the present invention, the step of determining the lateral swing amplitude value of the feature sequence according to the numerical fluctuation of the plurality of offset feature values includes:

[0015] Based on the offset amount sequence, the offset speed sequence, and the offset acceleration sequence, determine the offset amount standard deviation, the offset speed standard deviation, and the offset acceleration standard deviation;

[0016] Use one or more of the offset amount standard deviation, the offset speed standard deviation, and the offset acceleration standard deviation as the lateral swing amplitude value.

[0017] In an embodiment of the present invention, the trigger frequency of the lane management system is obtained in the following manner:

[0018] Obtain the total number of triggers of the lane management system;

[0019] Based on the ratio of the total number of triggers and a preset second time window, as the trigger frequency.

[0020] In an embodiment of the present invention, the torque characteristic data is obtained by the following method:

[0021] Obtain a torque sequence within a preset third time window, the torque sequence including a plurality of steering wheel torque values;

[0022] Determine the ratio of the duration during which the steering wheel torque value within the third time window is less than or equal to a preset torque value to the duration of the third time window;

[0023] Based on the torque sequence, determine a torque mutation value of the steering wheel, the torque mutation value being used to characterize the torque change of the steering wheel;

[0024] Based on a third weight, respectively weight the duration ratio and the torque mutation value to obtain weighted torque characteristic data.

[0025] In an embodiment of the present invention, the driving duration characteristic data is a duration fatigue value related to the driving duration, and the driving duration characteristic data is obtained by the following method:

[0026] Obtain the current driving duration of the driver;

[0027] Input the driving duration into a preset model, so that the preset model obtains the duration fatigue value that increases non-linearly as the driving duration increases on the basis of the fatigue time starting point.

[0028] In an embodiment of the present invention, the steps after inputting the driving duration into the preset model, so that the preset model obtains the duration fatigue value that increases non-linearly as the driving duration increases on the basis of the fatigue time starting point include:

[0029] Based on a multiplicative correction factor, perform dynamic weighted correction on the driving duration characteristic data, and the multiplicative correction factor is obtained based on an adjustment coefficient, the lane departure confidence, the trigger frequency, and the torque characteristic data.

[0030] In an embodiment of the present invention, the preset fatigue thresholds include a first preset threshold, a second preset threshold, and a third preset threshold, the first preset threshold being less than the second preset threshold, and the second preset threshold being less than the third preset threshold;

[0031] The step of comparing the total fatigue value with the preset fatigue thresholds to determine the fatigue assessment level of the driver includes:

[0032] When the total fatigue value is less than the first preset threshold, the fatigue assessment level of the driver is no fatigue, and the assessment result corresponding to the fatigue assessment level is safe driving;

[0033] When the total fatigue value is greater than the first preset threshold and less than the second preset threshold, the fatigue assessment level of the driver is mild fatigue, and the assessment result corresponding to the fatigue assessment level is a first-level risk;

[0034] When the total fatigue value is greater than the second preset threshold and less than the third preset threshold, the fatigue assessment level of the driver is moderate fatigue, and the assessment result corresponding to the fatigue assessment level is a second-level risk;

[0035] When the total fatigue value is greater than the third preset threshold, the fatigue assessment level of the driver is severe fatigue, and the assessment result corresponding to the fatigue assessment level is a third-level risk.

[0036] In an embodiment of the present invention, the method further includes:

[0037] When the assessment result is a first-level risk, a second-level risk or a third-level risk, determine whether intelligent driving is turned on;

[0038] If intelligent driving is turned on, reduce the weighted weight of the lane departure confidence, reduce the weighted weight of the trigger frequency, and increase the weighted weight of the torque feature data.

[0039] In an embodiment of the present invention, before comparing the total fatigue value with a preset fatigue threshold, the method further includes:

[0040] Obtain the driving style of the driver, where the driving style includes agile driving, conventional driving, and comfortable driving;

[0041] Determine the preset fatigue threshold based on the driving style of the driver, where, in the case where the driving styles are sorted as agile driving, conventional driving, and comfortable driving, the fatigue threshold shows a decreasing trend.

[0042] In summary, a driver fatigue state evaluation method according to an embodiment of the present application performs fatigue detection on the current driver based on lane deviation confidence, the triggering frequency of the lane management system, torque feature data, and driving duration feature data. Different from traditional contact-type fatigue detection solutions, it does not require the driver to wear any detection devices during driving. Throughout the driving process, the present application does not impose any burden on the driver. The driver only needs to drive the vehicle normally, and the present application can perform weighted summation on the lane deviation confidence, the triggering frequency of the lane management system, the torque feature data, and the driving duration feature data to obtain a total fatigue value to determine the current fatigue index of the driver.

[0043] For the driver fatigue state evaluation method proposed in the present application, other advantages, objectives, and features of the present application will be partially reflected by the following description, and partially will also be understood by those skilled in the art through the research and practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of illustrating the preferred embodiments and are not considered to be a limitation of this specification. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0045] Figure 1 It is a schematic flowchart of a driver fatigue state evaluation method provided by an embodiment of the present application;

[0046] Figure 2 It is a schematic structural diagram of an electronic device for driver state evaluation provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] In order to better understand the technical solutions provided by the embodiments of this specification, the technical solutions of the embodiments of this specification will be described in detail below through the drawings and specific embodiments. It should be understood that the specific features in the embodiments of this specification and the embodiments are detailed descriptions of the technical solutions of the embodiments of this specification, rather than limitations on the technical solutions of this specification. Without conflict, the technical features in the embodiments of this specification and the embodiments can be combined with each other.

[0048] In this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element. The term "more than two" includes two or more than two cases.

[0049] Please refer to Figure 1 , which is a schematic flow chart of a method for evaluating a driver's fatigue state provided by an embodiment of this application, and specifically may include:

[0050] S110. Obtain a plurality of fatigue reference data for evaluating the driver's fatigue state, where the plurality of fatigue reference data includes: dynamic characteristic data of the vehicle, torque characteristic data of the steering wheel, and driving duration characteristic data of the driver, and the dynamic characteristic data includes lane deviation confidence and / or trigger frequency of the lane management system;

[0051] Exemplarily, in order to comprehensively and accurately evaluate the driver's fatigue state, it is necessary to collect a plurality of data in different dimensions. These data reflect the state information of the driver during driving from different angles. The dynamic performance of the vehicle during driving is an important basis for evaluating the driver's fatigue. Among them, the lane deviation confidence reflects whether the vehicle keeps driving within the normal lane and the degree and stability of the deviation. For example, if the driver is fatigued, it will cause the vehicle to deviate from the center line of the lane unconsciously, and the lane deviation confidence can quantify this deviation. The trigger frequency of the lane management system, such as the trigger times of the LKA (Lane Keeping Assist) and LDW (Lane Departure Warning) systems, can also reflect whether the driving state of the vehicle is abnormal. Frequent triggering of these systems means that there is a problem with the driver's control of the vehicle's driving trajectory, and the driver may be driving fatigued.

[0052] The torque change when the driver operates the steering wheel can directly reflect the strength and stability of the operation. When driving fatigued, the driver may have situations such as too small torque (weak operation) or sudden torque change (sudden large change), and these abnormalities can be captured by obtaining the torque characteristic data of the steering wheel. Driving duration is an important factor affecting the driver's fatigue. As the driving time increases, the driver's fatigue level will gradually rise. Therefore, obtaining the driving duration characteristic data helps to evaluate the accumulation of fatigue.

[0053] Evaluating by integrating data from multiple dimensions can more comprehensively and accurately reflect the driver's fatigue state. Single data cannot fully reflect the complexity of fatigue, while multiple different types of data complement each other, can more comprehensively cover various factors affecting fatigue assessment, and improve the reliability of the assessment results.

[0054] S120. For each of the fatigue reference data, weight the fatigue reference data based on a preset first weight to obtain weighted fatigue data, where the first weights of at least two of the fatigue reference data are different;

[0055] Exemplarily, assign a preset first weight to each fatigue reference data, and then multiply each data by its corresponding weight to obtain weighted fatigue data. The magnitude of the weight represents the relative importance of the data in the overall fatigue assessment. The weights of at least two fatigue reference data are different because different types of data reflect the driver's fatigue state to different extents. For example, the lane deviation confidence may be more critical for judging fatigue in certain situations, so a higher weight is assigned; while the driving duration is important, but relying on it alone to judge fatigue is not precise enough, so the weight is relatively low. By setting different weights, the role of key data can be highlighted, and at the same time, the influence of each data can be reasonably balanced.

[0056] Through the weighting process, different treatments can be given according to the importance of different data in evaluating the fatigue state. This can more precisely reflect the contribution of each data in the overall assessment, making the assessment results more in line with the actual situation. If no weighting is performed and all data are treated equally, the influence of key factors may be ignored, resulting in inaccurate assessment results.

[0057] S130. Combine multiple weighted fatigue data to obtain the total fatigue value of the driver;

[0058] Exemplarily, comprehensively calculate each fatigue reference data after weighting. Finally, a total fatigue value representing the overall fatigue degree of the driver is obtained. This total fatigue value is the result of comprehensively considering data from multiple dimensions, reflecting the comprehensive fatigue state of the driver. The total fatigue value is shown in Equation (1).

[0059] S total = w1S deviation + w2T steer + w3f trigger + w4S time_adjusted (1);

[0060] Among them, w1, w2, w3, and w4 are the weights of each feature, satisfying w1 + w2 + w3 + w4 = 1 (tentatively set w1 = 0.25, w2 = 0.25, w3 = 0.25, w4 = 0.25, and adjusted according to subsequent fleet test data), S deviation is the lane departure confidence, T steer is the torque feature data, f trigger is the trigger frequency, S time_adjusted is the driving duration feature.

[0061] A single total fatigue value is obtained, which is convenient for quantifying and visually comparing the fatigue state of the driver. Whether it is between different drivers or the fatigue state of the same driver at different time points, it can be clearly judged and analyzed through this total fatigue value. It provides a clear numerical basis for determining the fatigue assessment level subsequently, making the assessment process more standardized and operable.

[0062] S140. Compare the total fatigue value with a preset fatigue threshold to determine the fatigue assessment level of the driver.

[0063] Exemplarily, different fatigue thresholds are preset, and these thresholds divide the range of the total fatigue value into different intervals. Compare the calculated total fatigue value of the driver with these preset thresholds, and determine the fatigue assessment level of the driver according to which interval the total fatigue value falls into. For example, three thresholds are set, corresponding to no fatigue, mild fatigue, moderate fatigue, and severe fatigue levels respectively. The total fatigue value lower than the lowest threshold is no fatigue, and being in different threshold intervals corresponds to the corresponding fatigue levels. The preset fatigue thresholds and the corresponding fatigue assessment levels provide a clear and unified standard for fatigue state assessment. It makes the assessment process objective and repeatable. Different assessment systems or operators can obtain consistent results by comparing and judging according to the same rules, avoiding subjective arbitrariness and inconsistent assessment results.

[0064] The clear fatigue assessment level can help relevant personnel quickly understand the fatigue degree of the driver, and thus take corresponding and targeted measures. For drivers with no fatigue, they can continue to drive normally; for drivers with mild fatigue, they can be reminded to pay attention to rest and appropriately adjust the driving rhythm; for drivers with moderate and severe fatigue, more strict measures need to be taken, such as forced rest or replacing the driver, etc. This way of taking different measures according to different levels can more effectively ensure driving safety.

[0065] In summary, the driver fatigue state assessment method proposed in the embodiments of the present application performs fatigue detection on the current driver based on the lane deviation confidence, the trigger frequency of the lane management system, the torque feature data, and the driving duration feature data. Different from traditional contact fatigue detection solutions, it does not require the driver to wear any detection devices during driving. The present application does not impose any burden on the driver throughout the driving process. The driver only needs to drive the vehicle normally, and the present application can perform weighted summation on the lane deviation confidence, the trigger frequency of the lane management system, the torque feature data, and the driving duration feature data to obtain the total fatigue value to determine the current fatigue index of the driver.

[0066] In some examples, the lane deviation confidence is obtained in the following manner:

[0067] Obtain at least one offset feature sequence of the vehicle relative to the lane reference line within a first time window, where the at least one offset feature sequence includes one or more of: an offset amount sequence, an offset speed sequence, and an offset acceleration, and the offset feature sequence includes multiple offset feature values;

[0068] For each of the offset feature sequences, determine the lateral swing amplitude value of the feature sequence according to the numerical fluctuation of the multiple offset feature values, and weight the lateral swing amplitude value based on a preset second weight value to obtain a weighted swing amplitude value;

[0069] Determine the lane deviation confidence according to at least one of the weighted swing amplitude values.

[0070] Exemplarily, first collect the offset feature sequences of the vehicle relative to the lane reference line within a specific first time window. These sequences include one or more of the offset amount, offset speed, and offset acceleration, and each sequence consists of multiple corresponding feature values. Due to the requirements of intelligent driving, at least one front-view camera is installed on the vehicle to detect the lane condition in real time. According to the information extracted by the front camera, it can be determined whether the vehicle has an offset phenomenon. From the lane detection module of the front-view camera, the system periodically obtains the lateral offset Δd(t) (t = 50 ms, the non-fixed value can be set and debugged).

[0071] Under normal circumstances, the vehicle drives smoothly within the lane or basically along the center of the lane, and Δd(t) fluctuates within a small range, and the fluctuation range is not too high. When the vehicle is driving with continuous offset, Δd(t) will show obvious and periodic offsets and be accompanied by large-amplitude and high-frequency left-right swings. Taking the center line of the road as the origin, the left offset is negative (Δd(t) < 0), and the right offset is positive (Δd(t) > 0). Within the first time window T, record the offset amount sequence of the vehicle relative to the center line of the lane:

[0072] Δd(t1), Δd(t2),..., Δd(t i )

[0073] wherein, Δd(t i ) is the offset eigenvalue, and Δt is the sampling interval. The offset speed is:

[0074]

[0075] wherein, v y (t) is the offset speed, and Δd(t) is the offset eigenvalue. The offset acceleration is:

[0076]

[0077] wherein, a y (t) is the offset acceleration, and Δd(t) is the offset eigenvalue. If, within a given time period, |v y | or |a y | exceeds a certain empirical threshold multiple times, it indicates that the driver or the vehicle is in an unstable lateral operation state.

[0078] Then, according to the fluctuation of these offset eigenvalues, the lateral swing amplitude value of the feature sequence is determined, and the lateral swing amplitude value is weighted by a preset second weight value. Finally, based on the weighted swing amplitude value, the lane offset confidence is determined, wherein the lane offset confidence is shown in Equation (4).

[0079]

[0080] wherein, S deviation is the lane offset confidence, m is the total number of features, i is the index variable representing the current calculated feature sequence number, w i is the importance of each feature for the continuous lane offset recognition, that is, the second weight value, satisfying ∑w i = 1, and T i is the result of segmented processing. By an accurate calculation process, the lane offset confidence is obtained, considering the offset features and their fluctuations in multiple dimensions during the vehicle's driving process, enabling the lane offset confidence to more accurately reflect the vehicle's driving stability, and thus providing a more reliable basis for evaluating the driver's fatigue state. Because when the driver is fatigued, the vehicle's driving trajectory often shows unstable offsets, and this way of obtaining the lane offset confidence can sensitively capture these changes.

[0081] In some examples, the step of determining the lateral swing amplitude value of the feature sequence according to the numerical fluctuations of the multiple offset eigenvalues includes:

[0082] Based on the offset sequence, offset velocity sequence, and offset acceleration sequence, determine the standard deviation of the offset, the standard deviation of the offset velocity, and the standard deviation of the offset acceleration;

[0083] Use one or more of the standard deviation of the offset, the standard deviation of the offset velocity, and the standard deviation of the offset acceleration as the lateral swing amplitude value.

[0084] Exemplarily, based on the offset sequence, offset velocity sequence, and offset acceleration sequence, calculate the corresponding standard deviations respectively, and then select one or more of these standard deviations as the lateral swing amplitude value.

[0085] Among them, the standard deviation of the offset sequence is:

[0086]

[0087] Among them, σ Δd is the standard deviation of the offset sequence, k is the total number of sampling points of the lateral offset within the time window, Δd(t i ) is the offset sequence, and μ Δd is the mean value of the offset sequence. The standard deviation can well reflect the degree of dispersion of the data, that is, the fluctuation situation. Therefore, through this method, the lateral swing degree of the vehicle in different aspects can be quantified. Calculate the lane offset degree based on the standard deviation, and perform normalization processing on the features in the lane offset degree:

[0088]

[0089] Among them, N i is the result of normalizing the features in the lane offset degree, X i,min is the minimum value calibrated by historical data, and X i,max is the maximum value calibrated by historical data.

[0090] For each normalized feature N i , perform segmentation processing to obtain the result of segmentation processing:

[0091]

[0092] Among them, α1 and α2 are segmentation thresholds, β is the weight of the intermediate interval, and T i is the result of segmentation processing. The specific method for quantifying the lateral swing amplitude value is clarified, making the calculation of this index consistent and repeatable. Using the standard deviation to measure the fluctuation situation can objectively reflect the stability change of the vehicle driving trajectory, which helps to improve the accuracy and scientificity of fatigue state assessment and avoid errors caused by subjective judgment.

[0093] In some examples, the triggering frequency of the lane management system is obtained in the following manner:

[0094] Obtain the total number of triggers of the lane management system;

[0095] Based on the ratio of the total number of triggers and a preset second time window, as the triggering frequency.

[0096] Exemplarily, first count the total number of triggers of the lane management system, and then divide the total number of triggers by the preset second time window. The obtained ratio is the triggering frequency. Among them, the preset second time window is fifteen minutes, and the triggering frequency is as shown in Equation (8):

[0097]

[0098] Among them, f trigger is the triggering frequency, N trigger is the total number of triggers of the statistically lane management system, and Δt is the preset second time window. Set an empirical upper limit f max , that is, the number of times most drivers trigger LKA and LDW within a fixed time period. This way of obtaining the triggering frequency is simple and direct, and can clearly reflect the frequency of the lane management system being triggered within a certain time. The triggering of the lane management system is often related to abnormal vehicle driving states. By calculating the triggering frequency, it can provide an intuitive reference index for evaluating the driver's fatigue state and help determine whether there is an abnormality in the driver's control of the vehicle driving trajectory.

[0099] In some examples, the torque characteristic data is obtained in the following manner:

[0100] Obtain the torque sequence within a preset third time window, and the torque sequence includes multiple steering wheel torque values;

[0101] Determine the ratio of the duration in which the steering wheel torque value within the third time window is less than or equal to the preset torque value to the duration of the third time window;

[0102] Based on the torque sequence, determine the torque mutation value of the steering wheel, and the torque mutation value is used to characterize the torque change of the steering wheel;

[0103] Based on the third weight, respectively weight the duration ratio and the torque mutation value to obtain the weighted torque characteristic data.

[0104] Exemplarily, a torque sequence of the steering wheel is obtained within a third time window, and the sequence consists of multiple steering wheel torque values. Next, calculate the proportion of the duration during which the steering wheel torque value is less than or equal to a preset torque value in the total duration of the third time window, and simultaneously determine the mutation value of the steering wheel torque during this period. Finally, use the third weight to weight these two indicators respectively, so as to obtain comprehensive torque characteristic data, and the torque characteristic data is shown in Equation (9).

[0105] T steer =w low *f(r low )+w var *g(σ M )(9);

[0106] Among them, T steer is the torque characteristic data, σ M is the standard deviation quantifying the fluctuation amplitude of the torque value, w low and w var are the third weights, and f(·) and g(·) are non-linear functions used to amplify the influence of outliers.

[0107]

[0108] Among them, σ M is the standard deviation quantifying the fluctuation amplitude of the torque value, M steer (t i ) is the steering wheel torque value, is the mean value quantifying the fluctuation amplitude of the torque value, and k is the total number of sampling points of the lateral offset within the third time window.

[0109] It comprehensively considers the distribution of the steering wheel torque in the time dimension (duration proportion) and the instantaneous change situation (torque mutation value). By weighting and integrating these two factors, it can more comprehensively and meticulously reflect the characteristics of the driver's operation of the steering wheel. When driving fatigued, the driver's operation of the steering wheel will show abnormal situations such as too small torque or mutation. This way of obtaining torque characteristic data can effectively capture these abnormalities and provide strong support for fatigue state assessment.

[0110] In some examples, the driving duration characteristic data is a duration fatigue value related to the driving duration, and the driving duration characteristic data is obtained through the following method:

[0111] Obtain the current driving duration of the driver;

[0112] Input the driving duration into a preset model so that the preset model obtains the duration fatigue value that increases non-linearly with the increase of the driving duration based on the fatigue time starting point.

[0113] Exemplarily, first, obtain the actual driving duration of the driver from ignition to engine shutdown. Then, input this driving duration data into a pre-set model. This pre-set model can simulate and capture the growth trend of the driver fatigue value as the driving duration increases, and can also identify the time point when the driver fatigue value begins to increase significantly. The result output after model processing is the driving duration feature, and the non-linear growth of the driving duration is shown in Equation (11).

[0114]

[0115] Among them, S time is the non-linear growth of the driving duration, t is the driving duration, k1 is the parameter controlling the steepness of the function, and t0 is the midpoint of the significant increase in fatigue (tentatively set t0 to 120 minutes, which can be adjusted later). By processing the driving duration through the pre-set model, the complex non-linear relationship between the fatigue value and the driving duration is fully considered, rather than simply assuming that the fatigue value increases linearly with the driving duration. This method can more accurately describe the impact of the driving duration on the driver's fatigue state, provide a more realistic data basis for accurately evaluating the driver's fatigue state later, and make the evaluation result more reliable.

[0116] In some examples, after the step of inputting the driving duration into the pre-set model to enable the pre-set model to obtain the duration fatigue value that non-linearly increases as the driving duration grows based on the fatigue time starting point, the steps include:

[0117] Dynamically weight and correct the driving duration feature data based on a multiplicative correction factor, and the multiplicative correction factor is obtained based on the adjustment coefficient, the lane deviation confidence level, the trigger frequency, and the torque feature data.

[0118] Exemplarily, in order to achieve the linkage between the driving duration and other features, a dynamic weighting mechanism is introduced. The influence of the driving duration is adjusted through a multiplicative correction factor, that is, the obtained driving duration feature is dynamically weighted and corrected through a multiplicative correction factor. This multiplicative correction factor is not fixed. It is jointly calculated based on the adjustment coefficient and other factors related to driver fatigue, such as the lane deviation confidence level, the trigger frequency of the lane management system, and the torque feature data. Among them, the adjustment coefficient is a key parameter for dynamically balancing the influence of the driving duration and other features, and its value is determined through experimental calibration or dynamic strategies for accurate fatigue state assessment. Through this dynamic weighting and correction, the driving duration feature can be better correlated and integrated with other evaluation factors.

[0119] Normalize the trigger frequency and the torque feature data, and the normalization result of the trigger frequency is shown in Equation (12).

[0120]

[0121] Among them, f trigger is the trigger frequency, f max is the experience upper limit, f min is the experience lower limit, is the normalization result of the trigger frequency. The normalization result of the torque characteristic data is shown in Equation (13).

[0122]

[0123] Among them, f trigger is the torque characteristic data, T max is the experience upper limit, T min is the experience lower limit, is the normalization result of the torque characteristic data. The multiplicative correction factor is shown in Equation (14).

[0124]

[0125] Among them, β1 is the multiplicative correction factor, α is the adjustment coefficient, S deviation is the lane departure confidence level, is the normalization result of the trigger frequency, is the normalization result of the torque characteristic data. The driving duration characteristic is shown in Equation (15).

[0126] S time_adjusted = S time *β1 (15);

[0127] Among them, β1 is the multiplicative correction factor, S time is the non-linear growth of the driving duration. Introducing a dynamic weighted correction mechanism makes the role of the driving duration factor in fatigue assessment no longer isolated. It can be adjusted according to real-time changing factors such as vehicle dynamic characteristics and driver operation characteristics, and more comprehensively reflect the actual fatigue condition of the driver. This can make the fatigue assessment model more flexible and adaptive, improve the accuracy and effectiveness of the assessment, and thus more timely and accurately identify the fatigue state of the driver.

[0128] In some examples, the preset fatigue threshold includes a first preset threshold, a second preset threshold, and a third preset threshold, the first preset threshold is less than the second preset threshold, and the second preset threshold is less than the third preset threshold;

[0129] The step of determining the fatigue assessment level of the driver by comparing the total fatigue value with the preset fatigue threshold includes:

[0130] When the total fatigue value is less than the first preset threshold, the fatigue assessment level of the driver is no fatigue, and the assessment result corresponding to the fatigue assessment level is safe driving;

[0131] When the total fatigue value is greater than the first preset threshold and less than the second preset threshold, the fatigue assessment level of the driver is mild fatigue, and the assessment result corresponding to the fatigue assessment level is a first-level risk;

[0132] When the total fatigue value is greater than the second preset threshold and less than the third preset threshold, the fatigue assessment level of the driver is moderate fatigue, and the assessment result corresponding to the fatigue assessment level is a second-level risk;

[0133] When the total fatigue value is greater than the third preset threshold, the fatigue assessment level of the driver is severe fatigue, and the assessment result corresponding to the fatigue assessment level is a third-level risk.

[0134] Exemplarily, a preset fatigue threshold system is defined, including a first preset threshold, a second preset threshold, and a third preset threshold, and these three thresholds are in an increasing relationship. After obtaining the total fatigue value of the driver, it is compared with these preset thresholds to determine the fatigue assessment level of the driver. When the total fatigue value is less than the first preset threshold, it is determined that the driver is in a non-fatigued state, corresponding to the assessment result of safe driving; when the total fatigue value is between the first preset threshold and the second preset threshold, it is evaluated as mild fatigue, corresponding to the assessment result of a first-level risk; when it is between the second preset threshold and the third preset threshold, it is moderate fatigue, corresponding to the assessment result of a second-level risk; when it is greater than the third preset threshold, it is severe fatigue, corresponding to the assessment result of a third-level risk. In this application, the first preset threshold is 0.8, the second preset threshold is 1.2, and the third preset threshold is 1.5.

[0135] An acoustic reminder is activated when the assessment result is at the first-level risk, an acoustic reminder, an optical reminder, and a tactile reminder are activated when the assessment result is at the second-level risk, and it is indicated that the current situation has a relatively high risk when the assessment result is at the third-level risk. At this time, an emergency acoustic reminder, an emergency optical reminder, and an emergency tactile reminder different from those at the second-level risk are activated. Among them, the third-level risk indicates that the driver is in a state of high fatigue or serious abnormal operation, and continuing to rely on intelligent driving may cause an accident. Exiting intelligent driving is a necessary safety measure.

[0136] This clear threshold setting and level division provide a clear and quantitative assessment standard for the driver's fatigue state. It makes the fatigue assessment result have a clear definition, facilitating the system's automatic judgment and the operator's understanding and taking corresponding measures. Different fatigue levels correspond to different risk prompts, which helps to take appropriate intervention measures in a timely manner to ensure driving safety.

[0137] In some examples, the method further includes:

[0138] When the evaluation result is a first-level risk, a second-level risk, or a third-level risk, determine whether intelligent driving is enabled;

[0139] If intelligent driving is enabled, reduce the weighted weight of the lane deviation confidence, reduce the weighted weight of the trigger frequency, and increase the weighted weight of the torque feature data.

[0140] Exemplarily, on the basis of determining the driver fatigue evaluation level, if the evaluation result is a first-level risk, a second-level risk, or a third-level risk, that is, when the driver has a certain degree of fatigue, it is necessary to further determine whether intelligent driving is enabled. If intelligent driving is in the enabled state, in order to more accurately evaluate the driver fatigue state and adapt to the changes in the driver's operation in the intelligent driving scenario, the weights in the evaluation model will be adjusted. Specifically, reduce the weighted weights of the lane deviation confidence and the trigger frequency, and at the same time increase the weighted weight of the torque feature data.

[0141] Specifically: During the period when intelligent driving is enabled, adjust the thresholds and weights of various indicators to accurately reflect the driver's fatigue state and effectively cope with the situation of reduced driver operation after intelligent driving is enabled. After intelligent driving is enabled, the driver's operation on the steering wheel decreases, and the weighted weight of the torque feature data of the steering wheel should be appropriately increased to monitor abnormal fluctuations of the steering wheel in real time and reduce misjudgment. After intelligent driving is enabled, the operation on the pedal decreases, and the pedal fluctuation weight should be increased. After intelligent driving is enabled, the vehicle driving state is mainly controlled by the program, and the trigger frequency weight is reduced. After intelligent driving is enabled, the vehicle driving state is mainly controlled by the program, and the lane deviation confidence weight is reduced. After intelligent driving is enabled, the driving duration feature data remains unchanged, and the dynamic weighting mechanism is still used. During the period when intelligent driving is enabled, the calculation model remains unchanged, and only the final calculation weights and judgment thresholds are adjusted. Therefore, during the period when intelligent driving is enabled, the total intelligent driving fatigue value is calculated by calculating the lane deviation confidence, the trigger frequency of the lane management system, the torque feature data of the steering wheel, and the driving duration feature data of the driver after adjusting the weights. During the period when intelligent driving is enabled, each threshold is increased accordingly to avoid false recognition and false alarms from affecting the user experience.

[0142] In some examples, before comparing the total fatigue value with a preset fatigue threshold, the method further includes:

[0143] Obtain the driving style of the driver, where the driving style includes agile driving, regular driving, and comfortable driving;

[0144] Determine the preset fatigue threshold based on the driving style of the driver, where, in the case where the driving styles are sorted as agile driving, regular driving, and comfortable driving, the fatigue threshold shows a decreasing trend.

[0145] Exemplarily, due to different driving styles of different drivers, the present application classifies driving styles into the following three types, namely agile driving, normal driving, and comfortable driving. Then, the preset fatigue threshold is determined according to the driver's driving style. Among them, drivers with an agile driving style usually have a quick reaction, a high operation frequency, and relatively frequent adjustments to the vehicle dynamics. Drivers with this driving style also have a high operation frequency in the non-fatigued state. Therefore, a higher threshold needs to be set to avoid false alarms and affect the user experience. Drivers with a normal driving style operate smoothly, have a moderate reaction, and have a relatively balanced adjustment to the vehicle dynamics. Therefore, the threshold is set to medium. Drivers with a comfortable driving style operate relatively slowly and have a relatively gentle adjustment to the vehicle dynamics. Therefore, more attention should be paid when an abnormality is recognized, and the corresponding preset fatigue threshold is set to the lowest.

[0146] There are significant differences in the behavior patterns of drivers with different driving styles during normal driving. If a unified fatigue threshold is used to evaluate all drivers, misjudgment is likely to occur. For example, drivers with an agile driving style may be misjudged as fatigued due to their normal high operation frequency, while drivers with a comfortable driving style may not be detected as fatigued in time due to their normal low operation activity. By determining different preset fatigue thresholds according to the driving style, these differences can be fully considered, making the fatigue assessment more in line with the actual situation of each driver, thus greatly improving the accuracy of the assessment.

[0147] For drivers with an agile driving style, the higher threshold reduces unnecessary fatigue reminders and avoids interfering with their normal driving; for drivers with a comfortable driving style, the lower threshold can more timely detect potential fatigue risks, and at the same time, they will not be unable to get timely reminders when they are fatigued due to an unreasonably high threshold. This can provide more appropriate and personalized services for drivers with different driving styles, improving the acceptance and satisfaction of users with the fatigue assessment system.

[0148] As Figure 2 shown, the embodiment of the present application also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored on the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, the steps of any of the above methods for evaluating the driver's fatigue state are implemented.

[0149] Since the electronic device introduced in this embodiment is the device used to implement a driver fatigue state assessment device in the embodiments of the present application, based on the method introduced in the embodiments of the present application, those skilled in the art can understand the specific implementation manners and various variations of the electronic device in this embodiment. Therefore, the specific implementation of how this electronic device implements the method in the embodiments of the present application will not be described in detail here. As long as the device used by those skilled in the art to implement the method in the embodiments of the present application falls within the scope of protection of the present application.

[0150] In the specific implementation process, when the computer program 311 is executed by the processor, it can implement Figure 1 any implementation manner in the corresponding embodiment.

[0151] It should be noted that in the above embodiments, the descriptions of each embodiment have their own focuses. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0152] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-readable program code.

[0153] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0154] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0155] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process or multiple processes and / or one block or multiple blocks in the flow. Figure 1 One process or multiple processes and / or blocks Figure 1 Steps for implementing the functions specified in one block or multiple blocks.

[0156] The embodiment of the present application also provides a computer program product, which includes computer software instructions. When the computer software instructions run on a processing device, the processing device is enabled to execute the process of the LDPC decoding method of the solid-state drive controller.

[0157] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions according to the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, a computer, a server, or a data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be stored by a computer or a data storage device such as a server or a data center that includes one or more integrated available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid-state disk (SSD)), etc.

[0158] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0159] In several embodiments provided by the present application, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of devices or units can be in electrical, mechanical, or other forms.

[0160] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0161] In addition, each functional unit in various embodiments of the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0162] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present application. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, and other various media that can store program codes.

[0163] The above embodiments are only used to illustrate the technical solution of the present application, rather than to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of various embodiments of the present application.

[0164] Although the preferred embodiments of this specification have been described, additional changes and modifications can be made to these embodiments by those skilled in the art once they learn of the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications that fall within the scope of this specification.

[0165] Obviously, those skilled in the art can make various changes and modifications to this specification without departing from the spirit and scope of this specification. Thus, if these modifications and variations of this specification fall within the scope of the claims of this specification and their equivalent technologies, this specification is also intended to include these modifications and variations.

Claims

1. A method for evaluating driver fatigue status, characterized in that: The method comprises: Acquiring a plurality of fatigue reference data for evaluating a driver's fatigue state, the plurality of fatigue reference data comprising: dynamic characteristic data of the vehicle, torque characteristic data of the steering wheel, and driving time characteristic data of the driver, the dynamic characteristic data comprising lane deviation confidence and / or triggering frequency of a lane management system; For each of the fatigue reference data, weighting the fatigue reference data based on a preset first weight to obtain weighted fatigue data, wherein the first weights of at least two of the fatigue reference data are different; Combining a plurality of the weighted fatigue data to obtain a total fatigue value of the driver; The fatigue assessment level of the driver is determined by comparing the total fatigue value with a preset fatigue threshold.

2. The method for evaluating driver fatigue status according to claim 1, characterized in that: The lane deviation confidence is obtained in the following way: Acquire at least one offset feature sequence of the vehicle relative to the lane reference line within a first time window, wherein the at least one offset feature sequence comprises: one or more of an offset sequence, an offset velocity sequence, and an offset acceleration sequence, and the offset feature sequence comprises a plurality of offset feature values; For each of the offset feature sequences, according to the numerical fluctuations of the plurality of offset feature values, a lateral swing amplitude value of the feature sequence is determined, and the lateral swing amplitude value is weighted based on a preset second weight value to obtain a weighted swing amplitude value; The lane deviation confidence is determined based on at least one of the weighted swing amplitude values.

3. The method for evaluating driver fatigue status according to claim 2, characterized in that: The step of determining the lateral swing amplitude value of the characteristic sequence according to the numerical fluctuation of the plurality of offset characteristic values ​​comprises: Based on the offset sequence, the offset velocity sequence and the offset acceleration sequence, determining an offset standard deviation, an offset velocity standard deviation and an offset acceleration standard deviation; One or more of the offset standard deviation, the offset velocity standard deviation and the offset acceleration standard deviation is used as the lateral swing amplitude value.

4. The method for evaluating driver fatigue status according to claim 1, characterized in that: The trigger frequency of the lane management system is obtained in the following way: Get the total number of triggers of the lane management system; The ratio of the total number of trigger times to a preset second time window is used as the trigger frequency.

5. The method for evaluating driver fatigue status according to claim 1, characterized in that: The torque characteristic data is obtained by: Acquire a torque sequence within a preset third time window, wherein the torque sequence includes a plurality of steering wheel torque values; Determine the duration of the steering wheel torque value being less than or equal to the preset torque value within the third time window, and the ratio of the duration of the third time window; Based on the torque sequence, determining a torque mutation value of the steering wheel, wherein the torque mutation value is used to characterize a torque change of the steering wheel; The duration ratio and the torque mutation value are weighted based on the third weight to obtain weighted torque feature data.

6. The method for evaluating driver fatigue status according to claim 1, characterized in that: The driving time characteristic data is a time fatigue value related to the driving time, and the driving time characteristic data is obtained in the following manner: Obtaining the current driving time of the driver; The driving duration is input into a preset model so that the preset model obtains the duration fatigue value which increases nonlinearly with the increase of the driving duration based on the fatigue time starting point.

7. The method for evaluating driver fatigue status according to claim 6, characterized in that: The step after inputting the driving duration into a preset model so that the preset model obtains the duration fatigue value that increases nonlinearly with the increase of the driving duration based on the fatigue time starting point includes: The driving duration characteristic data is dynamically weighted corrected based on a multiplicative correction factor, wherein the multiplicative correction factor is obtained based on an adjustment coefficient, the lane deviation confidence, the trigger frequency, and the torque characteristic data.

8. The method for evaluating driver fatigue status according to claim 1, characterized in that: The preset fatigue threshold includes a first preset threshold, a second preset threshold and a third preset threshold, the first preset threshold is smaller than the second preset threshold, and the second preset threshold is smaller than the third preset threshold; The step of determining the driver's fatigue assessment level by comparing the total fatigue value with a preset fatigue threshold comprises: When the total fatigue value is less than the first preset threshold, the fatigue assessment level of the driver is no fatigue, and the assessment result corresponding to the fatigue assessment level is safe driving; When the total fatigue value is greater than the first preset threshold value and less than the second preset threshold value, the fatigue assessment level of the driver is mild fatigue, and the assessment result corresponding to the fatigue assessment level is level one risk; When the total fatigue value is greater than the second preset threshold and less than the third preset threshold, the fatigue assessment level of the driver is moderate fatigue, and the assessment result corresponding to the fatigue assessment level is a level 2 risk; When the total fatigue value is greater than the third preset threshold, the fatigue assessment level of the driver is severe fatigue, and the assessment result corresponding to the fatigue assessment level is a third-level risk.

9. The method for evaluating driver fatigue status according to claim 8, characterized in that: The method further comprises: When the assessment result is level one risk, level two risk or level three risk, determining whether intelligent driving is turned on; If the intelligent driving is turned on, the weighted weight of the lane deviation confidence is reduced, the weighted weight of the trigger frequency is reduced, and the weighted weight of the torque characteristic data is increased.

10. The method for evaluating driver fatigue status according to claim 1, characterized in that: Before comparing the total fatigue value with a preset fatigue threshold, the method further includes: acquiring a driving style of a driver, wherein the driving style includes agile driving, regular driving, and comfortable driving; The preset fatigue threshold is determined based on the driving style of the driver, wherein when the driving styles are ranked as the agile driving, the conventional driving, and the comfortable driving, the fatigue threshold shows a decreasing trend.

Images